INFO-VAX Tue, 29 May 2007 Volume 2007 : Issue 292 Contents: =?KOI8-R?Q?=EB=F7=E9_=E5=C7=C5=DD=CB=DB=C6=C4?= Re: Active Batch from Advanced Systems Concepts Re: Active Batch from Advanced Systems Concepts Re: Anyone know why the Alpha market is so so quiet? Re: Anyone know why the Alpha market is so so quiet? Re: Anyone know why the Alpha market is so so quiet? Re: Anyone know why the Alpha market is so so quiet? Re: Anyone know why the Alpha market is so so quiet? Re: Anyone know why the Alpha market is so so quiet? Re: Anyone know why the Alpha market is so so quiet? Re: Anyone know why the Alpha market is so so quiet? RE: Anyone know why the Alpha market is so so quiet? ASE - HP AlphaServer Systems - OpenVMS credentials being retired? Re: ASE - HP AlphaServer Systems - OpenVMS credentials being retired? AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full CONVERTing RMS CR/LF files to explicit byte Stream format Re: CONVERTing RMS CR/LF files to explicit byte Stream format Re: CONVERTing RMS CR/LF files to explicit byte Stream format Re: CONVERTing RMS CR/LF files to explicit byte Stream format Re: CONVERTing RMS CR/LF files to explicit byte Stream format Re: Do Not Call Registry Re: DS10L ATI 7500 PCI Graphics Problems Re: DS10L ATI 7500 PCI Graphics Problems Re: DS10L ATI 7500 PCI Graphics Problems Re: DS10L ATI 7500 PCI Graphics Problems Re: DSSI allocation class puzzle Re: DSSI allocation class puzzle Re: DSSI allocation class puzzle Re: DSSI allocation class puzzle Re: DSSI allocation class puzzle Re: DSSI allocation class puzzle Re: DSSI allocation class puzzle Re: DSSI allocation class puzzle Re: DSSI allocation class puzzle Re: FalconStor Re: FalconStor and WAN repliaction Re: FalconStor and WAN repliaction Re: FalconStor and WAN repliaction Re: FalconStor and WAN repliaction Re: FalconStor and WAN repliaction Re: FalconStor and WAN repliaction Re: Hobbyist Licenses for ISV Layered Products Re: Hobbyist Licenses for ISV Layered Products Re: Hobbyist Licenses for ISV Layered Products Re: Hobbyist Licenses for ISV Layered Products Re: Hobbyist Licenses for ISV Layered Products Re: Hobbyist Licenses for ISV Layered Products Re: Hobbyist Licenses for ISV Layered Products Re: Indexed file (RMS) IRC$V_RRV question Re: Indexed file (RMS) IRC$V_RRV question Re: Indexed file (RMS) IRC$V_RRV question Re: Is VMS losing the Financial Sector, also? RE: Is VMS losing the Financial Sector, also? Re: Is VMS losing the Financial Sector, also? Re: Is VMS losing the Financial Sector, also? Re: Is VMS losing the Financial Sector, also? Re: Is VMS losing the Financial Sector, also? Re: Is VMS losing the Financial Sector, also? Re: Is VMS losing the Financial Sector, also? Re: Is VMS losing the Financial Sector, also? Re: Is VMS losing the Financial Sector, also? Re: Is VMS losing the Financial Sector, also? RE: Is VMS losing the Financial Sector, also? Looking for a H7868 power supply Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OM Group acquired by Nasdaq - VMS probably out Re: OpenVMS 2007 Bootcamp Re: OpenVMS 2007 Bootcamp Re: OpenVMS 2007 Bootcamp openvms pascal rdb "IN" predicate parameterparsing Re: openvms pascal rdb "IN" predicate parameterparsing OT Helmets: Hobbyist Licenses for ISV Layered Products PCSI problems on 8.3 - what to do? Re: PCSI problems on 8.3 - what to do? PCSI, disk space, UNDO, unseen dangers etc Re: PCSI, disk space, UNDO, unseen dangers etc Re: PCSI, disk space, UNDO, unseen dangers etc Re: PCSI, disk space, UNDO, unseen dangers etc Re: PCSI, disk space, UNDO, unseen dangers etc Re: PCSI, disk space, UNDO, unseen dangers etc Re: PCSI, disk space, UNDO, unseen dangers etc Re: PCSI, disk space, UNDO, unseen dangers etc php error from command line Problem with NFS client connecting to Windows 2003 server Re: Problem with NFS client connecting to Windows 2003 server Re: Problem with NFS client connecting to Windows 2003 server Re: Problem with NFS client connecting to Windows 2003 server Re: Problem with NFS client connecting to Windows 2003 server Re: Problem with NFS client connecting to Windows 2003 server PURGE Enhancement Suggestion (was Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header RDB turorial RDB Tutorial Recommended Graphic Adapter for Hobbyist Itanic? Re: Recommended Graphic Adapter for Hobbyist Itanic? Re: Recommended Graphic Adapter for Hobbyist Itanic? Re: Recommended Graphic Adapter for Hobbyist Itanic? Re: Recommended Graphic Adapter for Hobbyist Itanic? Re: Recommended Graphic Adapter for Hobbyist Itanic? RMS_TUNE_CHECK - ROOT: Primary key index root level is high Suggestion: SET PROMPT/POST="string" Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: World's Fastest Processor? For Now, Anyway Re: Xpdf 3.02 for OpenVMS/Itanium ---------------------------------------------------------------------- Date: Tue, 29 May 2007 14:22:39 +0400 From: "Ruslan R. Laishev" Subject: =?KOI8-R?Q?=EB=F7=E9_=E5=C7=C5=DD=CB=DB=C6=C4?= Message-ID: <3995DF2D429C1314ED2D60760705A2C1@NNTP.DeltaTel.RU> Hello, All! I looking for RDB tutorial "for beginners". Thanks for any pointers. -- + WBR, OpenVMS [Sys|Net] HardWorker ............. Skype: SysMan-One + Delta Telecom JSC, IMT-MC-450(CDMA2000) cellular operator Russia,191119,St.Petersburg,Transportny per. 3 Cel: +7 (812) 716-3222 +http://starlet.deltatelecom.ru ............. Frying on OpenVMS only + ------------------------------ Date: Mon, 28 May 2007 14:40:56 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Active Batch from Advanced Systems Concepts Message-ID: <00A6849D.2306BF15@SendSpamHere.ORG> In article <465A433A.21E4EEBC@spam.comcast.net>, David J Dachtera writes: > > >VAXman-, @SendSpamHere.ORG wrote: >> >> In article <4659d10b$0$21924$157c6196@dreader1.cybercity.dk>, "Dr. Dweeb" writes: >> {...snip...} >> >>>Dweeb >> >> >> >> I don't see anything in the logs that looks like dweeb.net. >> >> >> >> Can you tell me the time in the header of the bounceback message? >> >> >> > >> >For obvious reasons I have removed my email from below. >> >> I think I have it. Try sending again. > >That's the same message I was getting when everything I sent you from both work >and home was being bounced. If you get a bounceback, try including the following in the subject line of your message: [PASS:VAXman] I have a filter looking for this and it will allow your message past all of the other filters. Wouldn't it be great if SMTP wasn't such a Simple Mail Transport Protocol? Then we wouldn't need to have all of this anti- SPAM shit interfering with our lives and business. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: Mon, 28 May 2007 23:39:00 +0200 From: "Dr. Dweeb" Subject: Re: Active Batch from Advanced Systems Concepts Message-ID: <465b4bf6$0$21929$157c6196@dreader1.cybercity.dk> "John Vottero" wrote in message news:I6F5i.11428$rO7.5929@newssvr25.news.prodigy.net... > "Dr. Dweeb" wrote in message > news:46560936$0$7604$157c6196@dreader2.cybercity.dk... >> Folks, >> >> We are evaluation a product called Active Batch >> http://h21007.www2.hp.com/dspp/mop/mop_partner_product_detail_IDX/1,1331,5496,00.html >> for use in our Win2003AS environment. It all looks very familiar, like a >> graphical front end to the VMS queue manager, with concepts I know. >> >> Are there any users of the product here who would like to offer me the >> benefit of their experience with the product and equally importantly, >> their experiences when dealing with the company for support etc. >> > > I don't know anything about ActiveBatch but, I hope you'll take a look at > our competing product, JAMS. > I have directed the analyst to give it a whirl. He is a .NET specialist, so he might like what he sees. I will let you know what we decide - but you are officially in the game now. Dweeb > We've been creating enterprise grade scheduling software for over 20 years > and we're now trying to bring that expertise to the Windows platform. > JAMS for OpenVMS is still marketed, developed and supported and I think > most (all?) of our OpenVMS customers are happy (with JAMS at least). JAMS > for Windows .NET is the same basic architecture as JAMS for OpenVMS but, > it's all new code for the Windows platform. > > You can download the free Developer's Edition at: > > http://www.mvpsi.com/Free.html > > There's also a link on that page that will get you a full blown > evaluation. > > Thanks for listening. > > John Vottero > JAMS Technical Support > ------------------------------ Date: Mon, 28 May 2007 11:47:49 -0500 From: pechter@pechter.dyndns.org (William Pechter) Subject: Re: Anyone know why the Alpha market is so so quiet? Message-ID: In article , Ron Johnson wrote: >On 05/27/07 09:35, Main, Kerry wrote: >[snip] >> >> And we ignore that enthusiasts have to eat and to eat they need >> to make money. >> >>> (we ignore the fact that cars require materials per copy, >>> because software does not) >>> >> >> Software requires time and skilled resources. Both are limited >> resources. As the old saying goes - "Linux is free as long as you >> value your time as free as well." > >Unadulterated hog poop. There's a whole lot of paid worker bees in >there, including some from your own company. > >http://lwn.net/Articles/222773/ > >Top lines changed by employer >(Unknown) 66154 19.0% >Red Hat 44527 12.8% >(None) 38099 11.0% >IBM 25244 7.3% >Astaro 15306 4.4% >Linux Foundation 13638 3.9% >Qumranet 12108 3.5% >Novell 11930 3.4% >Intel 11652 3.4% >SANPeople 9888 2.8% >NetXen 9607 2.8% >Sony 8497 2.4% >Broadcom 8349 2.4% >Tensilica 8195 2.4% >Nokia 5581 1.6% >MontaVista 4394 1.3% >Uni. Aberdeen 4324 1.2% >LWN.net 3975 1.1% >Secretlab 3370 1.0% >HP 3211 0.9% > >[snip] > >> I would be willing to bet that a large majority of IT shops today >> have only a small fraction of the applicable (after review) RH >> Linux security patches identified at the following RH site >> applied to all their Dev/QA/Test/Prod systems. >> >> https://www.redhat.com/archives/enterprise-watch-list/ [click on >> thread for each month and add them up - 34 security patches so >> far in May 2007.. 34!!!] > >Not long. > >Is gimp installed on my server? No. >Is vixie-cron installed? Maybe. >Is Evolution installed? No. >Is libpng installed? Maybe, probably not. >Is squirrelmail installed? Maybe, probably not. >Is bluez-utils installed? No. >Is samba installed? Maybe. >Is freeradius installed? No. >Is php installed? Maybe. > >That took me 3 minutes, most of it typing. It would take me less >time to eyeball-scan the list for my server's relevant apps. > >> Case in point - if you have RH Linux systems - have you reviewed >> the 34 security patches released this month to see if they apply >> to your environment? >> >> At what point does someone wake up and say "we can not afford >> this platform!!" > >As someone who has used Linux for 7 years, and VMS for 16, I've got >to say that your arguments just don't hold water. > >Ron I've given up this argument. Kerry's spouting the company line. I've been there. I no longer work for a computer/os vendor. I can speak what I see in small and medium sized IT shops. The truth is RH Linux is no different from Solaris or HP-UX from a security perspective and VMS doesn't run on any hardware platform at my employer. (We're moving off an IBM 390 series box and the rest is either AIX (also going away over time) and RHEL Advanced Server. Not my first choice -- but I can live with it as a reasonable choice these days... bill -- "When I think back on all the crap I learned in Vax school It's a wonder I fixed anything at all." (to the tune of Kodachrome) pechter-at-gmail.com ------------------------------ Date: Mon, 28 May 2007 11:56:21 -0500 From: pechter@pechter.dyndns.org (William Pechter) Subject: Re: Anyone know why the Alpha market is so so quiet? Message-ID: In article <_vm6i.168407$nh4.140265@newsfe20.lga>, Ron Johnson wrote: >On 05/27/07 14:33, Main, Kerry wrote: >>> -----Original Message----- >>> From: Ron Johnson [mailto:ron.l.johnson@cox.net] >>> Sent: May 27, 2007 12:11 PM >>> To: Info-VAX@Mvb.Saic.Com >>> Subject: Re: Anyone know why the Alpha market is so so quiet? >>> >>> On 05/27/07 09:35, Main, Kerry wrote: >>> [snip] >>>> And we ignore that enthusiasts have to eat and to eat they need >>>> to make money. >>>> >>>>> (we ignore the fact that cars require materials per copy, >>>>> because software does not) >>>>> >>>> Software requires time and skilled resources. Both are limited >>>> resources. As the old saying goes - "Linux is free as long as you >>>> value your time as free as well." >>> Unadulterated hog poop. There's a whole lot of paid worker bees in >>> there, including some from your own company. >>> >> >> You missed the point. I was referring to the *maint* of Linux (and >> Windows) environments. Of course there are paid folks creating Linux >> from many different sources. > >Oh, ok. > >Still, "Linux is free" is a straw man that should have been burned >years ago. Even the GPL explicitly states that you shouldn't just >throw up your code on expensive bandwidth, and even "offer warranty >protection in exchange for a fee". > >Freedom of speech and freedom to look at and modify the code is >primary. How many times on this list have you seen "help! my data >is locked up in this undocumented format from 1982 and the company >who wrote the s/w is long gone!" Having the code gives you at least >a *fighting* chance at retrieving your old data. > >Free as in beer is a nice, but secondary, artifact. And serves to >spread the pollen, like MSFT turning a blind eye to piracy for 25 years. > >>> http://lwn.net/Articles/222773/ >>> >>> Top lines changed by employer >>> (Unknown) 66154 19.0% >>> Red Hat 44527 12.8% >>> (None) 38099 11.0% >>> IBM 25244 7.3% >>> Astaro 15306 4.4% >>> Linux Foundation 13638 3.9% >>> Qumranet 12108 3.5% >>> Novell 11930 3.4% >>> Intel 11652 3.4% >>> SANPeople 9888 2.8% >>> NetXen 9607 2.8% >>> Sony 8497 2.4% >>> Broadcom 8349 2.4% >>> Tensilica 8195 2.4% >>> Nokia 5581 1.6% >>> MontaVista 4394 1.3% >>> Uni. Aberdeen 4324 1.2% >>> LWN.net 3975 1.1% >>> Secretlab 3370 1.0% >>> HP 3211 0.9% >>> >>> [snip] >>> >>>> I would be willing to bet that a large majority of IT shops today >>>> have only a small fraction of the applicable (after review) RH >>>> Linux security patches identified at the following RH site >>>> applied to all their Dev/QA/Test/Prod systems. >>>> >>>> https://www.redhat.com/archives/enterprise-watch-list/ [click on >>>> thread for each month and add them up - 34 security patches so >>>> far in May 2007.. 34!!!] >>> Not long. >>> >>> Is gimp installed on my server? No. >>> Is vixie-cron installed? Maybe. >>> Is Evolution installed? No. >>> Is libpng installed? Maybe, probably not. >>> Is squirrelmail installed? Maybe, probably not. >>> Is bluez-utils installed? No. >>> Is samba installed? Maybe. >>> Is freeradius installed? No. >>> Is php installed? Maybe. >>> >>> That took me 3 minutes, most of it typing. It would take me less >>> time to eyeball-scan the list for my server's relevant apps. >>> >> >> Ok, so put your self in the position of a typical Customer who has 100+ >> small Linux and Windows servers to manage. They are running a mix of RH >> Linux ES3, ES4 and even ES5 as well as a number of current and legacy >> versions of RH Linux WS. > >The same way he upgrades Windows? No, just kidding. > >> Some are managed by the same group - others are self managed by >> developers who think central IT do not know what they are doing. > >Well, heck, that's his problem. Let him run "yum update". > >> On top of this, like all well run IT shops with important applications, >> they need to test their Apps before they release OS patches. >> >> Now, how do you propose to handle this environment - and this is very >> typical. > >See below. > >>>> Case in point - if you have RH Linux systems - have you reviewed >>>> the 34 security patches released this month to see if they apply >>>> to your environment? >>>> >>>> At what point does someone wake up and say "we can not afford >>>> this platform!!" >>> As someone who has used Linux for 7 years, and VMS for 16, I've got >>> to say that your arguments just don't hold water. >>> >>> Ron >> >> And as someone who has extensive experience (25+) in large multi-vendor, >> multi-platform data centers, perhaps you have only dealt with some small >> server environments (less than 20 perhaps?). >> >> If you have experience in large DC's - how do you recommend Cust's >> continually test their app's and keep all these different Linux versions >> current with all of the 5-20 security patches per month? >> >> I just do not see a lot of practical experience in what you are stating >> when you look at how large DC's operate today. > >I'd start by writing (in Perl or Python) a script to: > 1) scrape the relevant errata page, grabbing only packages since > you last upgraded, > 2) examine the packages installed on that system, > 3) compare the two sets, > 4) display a list of relevant packages sorted by priority. > >If you've got 100 Linux boxes, they're each probably dedicated to a >single task (or are all clones like in a scaled-out web "server"), >which means that each month's errata list per machine is pretty small. > >KISS. If you've got 100 servers, they shouldn't be loaded up with >unnecessary packages. > >-- >Ron Johnson, Jr. >Jefferson LA USA > >Give a man a fish, and he eats for a day. >Hit him with a fish, and he goes away for good! Yup... yum update is good. What's better is a tool like I've got. Apply the patches, test. Make a chrooted tree with all fixes on that platform. Deploy with rsync to all boxes of the same type on the net. All changes from the original distribution can be rsync'd from the tree to take a kickstarted raw box (minimal system) to full functionality with all packages/apps installed in one upgrade rsync. Upgrade 20 machines in 10 minutes or less. This includes applying any necessary customizations for the applications where the correct configs based on machine name get sync'd with the updates. It becomes a very easy process to update all web servers with one command all mail servers with the next. They can be combined to --servers ALL (but I like doing them in smaller groups). This has been used to manage app changes on Solaris as well as Linux and SCO Unix. Sometimes we used yum update for the os stuff and just this deployment tool for our customized versions of stuff like sendmail. Not too bad if you know how to do it. BTW -- we also can push out a copy of the rpm's to a temp directory with the program and post-script run the rpm -Uvh to install the newer packages without yum. Lots of ways to automate this. Just requires a little thought and experience. Bill -- "When I think back on all the crap I learned in Vax school It's a wonder I fixed anything at all." (to the tune of Kodachrome) pechter-at-gmail.com ------------------------------ Date: Mon, 28 May 2007 16:58:33 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: Anyone know why the Alpha market is so so quiet? Message-ID: <465b4266$0$90274$14726298@news.sunsite.dk> Main, Kerry wrote: >>> Yeah, but how long would Toyota and Ford be in business if they >>> gave away cars for free and only sold $1000/year support >>> licenses? >> If the cars got build by enthusiasts which did it for fun, then >> they would probably be in good shape. >> (we ignore the fact that cars require materials per copy, because >> software does not) > And we ignore that enthusiasts have to eat and to eat they need to > make money. The enthusiasts will still make money. I see no reason to worry about the incomes for the developers. But I would be very worried as a manager or other overhead in a company depending on revenue from OS sales, because the days of huge profits on OS's are over. > Software requires time and skilled resources. Both are limited > resources. As the old saying goes - "Linux is free as long as you > value your time as free as well." But those subscribing to that philosophy will in 20 years either be working on IBM mainframe or flipping burgers. It is the equivalent of 12-14 years ago when I heard a DEC reseller when asked about whether it was smart of DEC to make an alliance with MS replied "MS is just a small company - who do they think they are.". >>> And as has been noted here in the past - if an OS platform has >>> 5-20 security patches released each and every month, given the >>> huge QA/testing for App certifications required, can a company >>> actually afford that platform? >> Apparently yes. > > Actually, you can not fault the Operations folks for not keeping the > hundreds of servers they maintain with 5-20 security patches per > month. The sheer volume is over whelming in many cases. > > Remember that proper IT processes require their apps to be tested > before any new OS patches are released into production. If you make a reality check, then you will see that people are switching to Linux. >>> Unfortunately, as most security analysts will tell you, approx >>> 50-60% >>> of all security incidents are internal related. >> How many of those use holes that should have been patched ?? > > See above - in many cases, So you are claiming that all those 50-60% uses unpatched security holes and none use valid accounts and priviliges ???? > At what point does someone wake up and say "we can not afford this > platform!!" Same month as people wake up and decide that VMS is best OS for the desktop to run word processing and spreadsheets. Arne ------------------------------ Date: Mon, 28 May 2007 17:15:07 -0500 From: Ron Johnson Subject: Re: Anyone know why the Alpha market is so so quiet? Message-ID: On 05/28/07 15:58, Arne Vajhøj wrote: > Main, Kerry wrote: [snip] >> Software requires time and skilled resources. Both are limited >> resources. As the old saying goes - "Linux is free as long as you >> value your time as free as well." > > But those subscribing to that philosophy will in 20 years > either be working on IBM mainframe or flipping burgers. > > It is the equivalent of 12-14 years ago when I heard a > DEC reseller when asked about whether it was smart of DEC > to make an alliance with MS replied "MS is just a small company - who > do they think they are.". Except now DEC is gone and it's products are gone or disappearing, and MSFT is the most profitable company in the known multiverse. -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Tue, 29 May 2007 01:54:04 +0200 From: "Dr. Dweeb" Subject: Re: Anyone know why the Alpha market is so so quiet? Message-ID: <465b6b9e$0$21926$157c6196@dreader1.cybercity.dk> "Ron Johnson" wrote in message news:MvI6i.88784$vE1.9157@newsfe24.lga... > On 05/28/07 15:58, Arne Vajhøj wrote: >> Main, Kerry wrote: > [snip] >>> Software requires time and skilled resources. Both are limited >>> resources. As the old saying goes - "Linux is free as long as you >>> value your time as free as well." >> >> But those subscribing to that philosophy will in 20 years >> either be working on IBM mainframe or flipping burgers. >> >> It is the equivalent of 12-14 years ago when I heard a >> DEC reseller when asked about whether it was smart of DEC >> to make an alliance with MS replied "MS is just a small company - who >> do they think they are.". > > Except now DEC is gone and it's products are gone or disappearing, and > MSFT is the most profitable company in the known multiverse. > It's easy to make money as a monopoly Dweeb > -- > Ron Johnson, Jr. > Jefferson LA USA > > Give a man a fish, and he eats for a day. > Hit him with a fish, and he goes away for good! ------------------------------ Date: Tue, 29 May 2007 13:06:09 +0800 From: "Richard Maher" Subject: Re: Anyone know why the Alpha market is so so quiet? Message-ID: Hi Kerry, Ron Sorry to but in, and I'm not trying to put words in Kerry's mouth, but WRT: - [On top of this, like all well run IT shops with important applications, they need to test their Apps before they release OS patches.] and [If you have experience in large DC's - how do you recommend Cust's continually test their app's and keep all these different Linux versions current with all of the 5-20 security patches per month?] I think it would help if Ron pointed out exactly what testing he performs on each system and the hosted applications, after each one of these yum fests. 1) Let's just do it and see what happens on Monday? 2) System manager reboots and checks the console for errors? 3) Fire up the applications and make a couple of quick queries? 4) Run a comprehensive regression test suite ? 5) Check performance, resource utilization, output all the same? 6) Get Users involved to verify results? 7) Run an End-of-Year? I'm not saying all of those things should be done for every minor version upgrade, and I have seen a few VMS sites that never do more than (1), but if Kerry could say what he expects should be done, and Ron could say what he actually does then it might speed things up a little bit :-) Personally, I've always been a bit pissed off that the *NIX and Windows crowds never had to face the same strict change-control procedures that the VMS side of the fence does. We have to have sign-off on each stage of testing and fall-back procedures whereas the Unix and Windows people just seem to get away with "It's an upgrade! Whaddaya gonna do - just ignore it?". Obviously for urgent fixes and outages there are retrospective change controls; so maybe everything on *nix is simply urgent? If something goes wrong and a change has to be rolled-back (or a fix for the fix applied) don't you guys get to go to those lovely post-mortem meetings with finger pointing all culminating with the contractor sacrifice ritual? If there's no reckoning or accountability then shit yeah just put the six-guns and Stetson on and upgrade those suckers li'l missy! Just look at VMSINSTAL; if the change works on Itanium then VMS engineering sees no reason to waste time repeating them on Alpha. I mean why should it be any different eh? (And even if it does break; in a hundred years who'll give a shit?) My favourite was a very big bank in the UK that worked out early on that things usually only go wrong after you've changed something, so set about making it very difficult to make any change at all. When it came to the DR tests we had to fail over to the other data centre for a week and then fail back (obviously) with no loss of data. The Unix guys failed over, did a couple of queries and then failed back immediately (else lose whatever they did in DR state). Then there was that whole Rapid Application Development thing where the VMS guys had to actually design stuff and come up with comprehensive functional specifications and more user sign-off, whereas the other guys just started cutting code and bypassing release management. I dunno, maybe we should stop being the only cyclist in the Tour de France that doesn't have a performance boosting cocktail running through their veins? Cheers Richard Maher "Main, Kerry" wrote in message news:FA60F2C4B72A584DBFC6091F6A2B8684023E7DC0@tayexc19.americas.cpqcorp.net... > -----Original Message----- > From: Ron Johnson [mailto:ron.l.johnson@cox.net] > Sent: May 27, 2007 12:11 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: Anyone know why the Alpha market is so so quiet? > > On 05/27/07 09:35, Main, Kerry wrote: > [snip] > > > > And we ignore that enthusiasts have to eat and to eat they need > > to make money. > > > >> (we ignore the fact that cars require materials per copy, > >> because software does not) > >> > > > > Software requires time and skilled resources. Both are limited > > resources. As the old saying goes - "Linux is free as long as you > > value your time as free as well." > > Unadulterated hog poop. There's a whole lot of paid worker bees in > there, including some from your own company. > You missed the point. I was referring to the *maint* of Linux (and Windows) environments. Of course there are paid folks creating Linux from many different sources. > http://lwn.net/Articles/222773/ > > Top lines changed by employer > (Unknown) 66154 19.0% > Red Hat 44527 12.8% > (None) 38099 11.0% > IBM 25244 7.3% > Astaro 15306 4.4% > Linux Foundation 13638 3.9% > Qumranet 12108 3.5% > Novell 11930 3.4% > Intel 11652 3.4% > SANPeople 9888 2.8% > NetXen 9607 2.8% > Sony 8497 2.4% > Broadcom 8349 2.4% > Tensilica 8195 2.4% > Nokia 5581 1.6% > MontaVista 4394 1.3% > Uni. Aberdeen 4324 1.2% > LWN.net 3975 1.1% > Secretlab 3370 1.0% > HP 3211 0.9% > > [snip] > > > I would be willing to bet that a large majority of IT shops today > > have only a small fraction of the applicable (after review) RH > > Linux security patches identified at the following RH site > > applied to all their Dev/QA/Test/Prod systems. > > > > https://www.redhat.com/archives/enterprise-watch-list/ [click on > > thread for each month and add them up - 34 security patches so > > far in May 2007.. 34!!!] > > Not long. > > Is gimp installed on my server? No. > Is vixie-cron installed? Maybe. > Is Evolution installed? No. > Is libpng installed? Maybe, probably not. > Is squirrelmail installed? Maybe, probably not. > Is bluez-utils installed? No. > Is samba installed? Maybe. > Is freeradius installed? No. > Is php installed? Maybe. > > That took me 3 minutes, most of it typing. It would take me less > time to eyeball-scan the list for my server's relevant apps. > Ok, so put your self in the position of a typical Customer who has 100+ small Linux and Windows servers to manage. They are running a mix of RH Linux ES3, ES4 and even ES5 as well as a number of current and legacy versions of RH Linux WS. Some are managed by the same group - others are self managed by developers who think central IT do not know what they are doing. On top of this, like all well run IT shops with important applications, they need to test their Apps before they release OS patches. Now, how do you propose to handle this environment - and this is very typical. > > Case in point - if you have RH Linux systems - have you reviewed > > the 34 security patches released this month to see if they apply > > to your environment? > > > > At what point does someone wake up and say "we can not afford > > this platform!!" > > As someone who has used Linux for 7 years, and VMS for 16, I've got > to say that your arguments just don't hold water. > > Ron And as someone who has extensive experience (25+) in large multi-vendor, multi-platform data centers, perhaps you have only dealt with some small server environments (less than 20 perhaps?). If you have experience in large DC's - how do you recommend Cust's continually test their app's and keep all these different Linux versions current with all of the 5-20 security patches per month? I just do not see a lot of practical experience in what you are stating when you look at how large DC's operate today. Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT) OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Tue, 29 May 2007 02:23:13 -0500 From: Ron Johnson Subject: Re: Anyone know why the Alpha market is so so quiet? Message-ID: On 05/29/07 00:06, Richard Maher wrote: > Hi Kerry, Ron > > Sorry to but in, and I'm not trying to put words in Kerry's mouth, but > WRT: - > > [On top of this, like all well run IT shops with important applications, > they need to test their Apps before they release OS patches.] > > and > > [If you have experience in large DC's - how do you recommend Cust's > continually test their app's and keep all these different Linux versions > current with all of the 5-20 security patches per month?] > > I think it would help if Ron pointed out exactly what testing he performs on > each system and the hosted applications, after each one of these yum fests. > > 1) Let's just do it and see what happens on Monday? > 2) System manager reboots and checks the console for errors? > 3) Fire up the applications and make a couple of quick queries? > 4) Run a comprehensive regression test suite ? > 5) Check performance, resource utilization, output all the same? > 6) Get Users involved to verify results? > 7) Run an End-of-Year? > > I'm not saying all of those things should be done for every minor version > upgrade, and I have seen a few VMS sites that never do more than (1), but if > Kerry could say what he expects should be done, and Ron could say what he > actually does then it might speed things up a little bit :-) As you say, it depends on the patch. Application changes go thru a QA sign-off after being run on the QA cluster. The larger the changes, the bigger the QA effort. Some VMS & Rdb patches, though, can't be *adequately* tested on the QA cluster because it's not as powerful as the main cluster and can't simulate the same load. -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Tue, 29 May 2007 09:38:53 +0200 From: Michael Kraemer Subject: Re: Anyone know why the Alpha market is so so quiet? Message-ID: Arne Vajhøj schrieb: > It is the equivalent of 12-14 years ago when I heard a > DEC reseller when asked about whether it was smart of DEC > to make an alliance with MS replied "MS is just a small company - who > do they think they are.". which was a rather stupid statement even back then. 12 years ago, that's 1995, and at this time M$ was already the same order of magnitude as DEC. And only 3 years later a small company called DEC was swallowed by much bigger PC maker Compaq. ------------------------------ Date: Tue, 29 May 2007 12:17:57 -0400 From: "Main, Kerry" Subject: RE: Anyone know why the Alpha market is so so quiet? Message-ID: > -----Original Message----- > From: William Pechter [mailto:pechter@pechter.dyndns.org] > Sent: May 28, 2007 12:56 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: Anyone know why the Alpha market is so so quiet? >=20 snip..=B8 >=20 > Yup... yum update is good. >=20 > What's better is a tool like I've got. >=20 > Apply the patches, test. So is this where the full business app testing occurs or is this just = the system manager test ? > Make a chrooted tree with all fixes on that platform. > Deploy with rsync to all boxes of the same type on the net. >=20 > All changes from the original distribution can be rsync'd from the > tree > to take a kickstarted raw box (minimal system) to full functionality > with all packages/apps installed in one upgrade rsync. >=20 > Upgrade 20 machines in 10 minutes or less. >=20 > This includes applying any necessary customizations for the > applications > where the correct configs based on machine name get sync'd with the > updates. >=20 As I mentioned earlier, it is not the roll-out of the patches that is = the issue. Heck, that is relatively minor as you can even easily do this = with all of the Windows security patches. The big issue by far is the re-certification and testing of important = business applications with all of the monthly OS security patches. For small and some medium businesses with small numbers of users, this = is not an issue as they simply apply the patch and reboot. If a OS = security patch breaks the kernel or an application, then they can simply = roll-back with minimal impact as the numbers of users are not that = large. That is usually not the case with large IT environments with mission = critical environments. Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20 OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Tue, 29 May 2007 10:38:17 -0400 From: sol gongola Subject: ASE - HP AlphaServer Systems - OpenVMS credentials being retired? Message-ID: <1WW6i.256$Jl3.210@newsfe12.lga> This is the text of an email i recieved from Are they killing all VMS certifications? ------------------------------------------------------------------------------------------------------------- The HP Certified Professional credential you currently hold ASE - HP AlphaServer Systems - OpenVMS will become inactive effective 30 April 2007. The retirement of this credential from our certification portfolio is a result of the courseware and exams becoming obsolete with no plans for future development or updates. What does this status change mean to you? · You will no longer see this title on our list of active credentials, and no new candidates will be accepted. · You still hold this credential and will remain an active HP Certified Professional at the Integration level, with all applicable Program benefits. · There are currently no migration requirements. For now, no action is required of you. If you are interested in achieving an active credential, please refer to our web site at: www.hp.com/go/certification/americas to search our portfolio. We want to reassure you of your value to the HP Certified Professional Program, and we look forward to maintaining a successful relationship with you. If you have additional questions about this change, please contact the Americas HP Certified Professional Program Office at: certification.americas@hp.com . Regards, Americas HP Certified Professional Program Team . ------------------------------ Date: 29 May 2007 08:11:26 -0700 From: IanMiller Subject: Re: ASE - HP AlphaServer Systems - OpenVMS credentials being retired? Message-ID: <1180451486.468391.326190@q66g2000hsg.googlegroups.com> No, current VMS certifications are based on current hardware and software i.e. OpenVMS I64 V8.3 like it or not they don't sell alphaservers any more. ------------------------------ Date: Mon, 28 May 2007 16:22:13 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: When running AUTOGEN, I got the dreaded %SYSTEM-W-HEADERFULL, file header is full ** WARNING ** - Error creating SYS$SYSROOT:[SYSEXE]SYSDUMP.DMP. ** ERROR ** - error creating SYSDUMP.DMP. SYSDUMP.DMP needs to be created manually with 393258 blocks Presumably, I would get the same error when trying to create it manually, right? $ dir SYS$SYSROOT:[SYSEXE]SYSDUMP.DMP/siz Directory SYS$SYSROOT:[SYSEXE] SYSDUMP.DMP;4 215910 OK, it wants a larger dump file, and there is room: $ sh dev sys$sysdevice Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt DSA133: Mounted 0 ALPHASYS_3 1070271 488 3 $33$DKB100: (GLADIA) ShadowSetMember 0 (member of DSA133:) $33$DKB600: (GLADIA) ShadowSetMember 0 (member of DSA133:) $ help/messa HEADERFULL HEADERFULL, file header is full Facility: SYSTEM, System Services Explanation: The file header map area on the volume is full and cannot be extended. This error occurs only when the file header extension is inhibited, for example, when the index file is being extended. User Action: Compress the volume by copying it with BACKUP. This condition can be avoided by increasing the value for the /HEADER qualifier of the INITIALIZE command. For ODS-2 disks, the value of the /HEADER qualifier should equal the approximate number of files you expect to store on the disk. Three questions. First, I'm sure that the INITIALIZE had a value for /HEADER which is OK. I remember thinking a lot about this several years ago and coming up with a table based on disk size and use of the disk (system, user, software, data, scratch---this, together with the size, determining the maximum number of files). Is there some other reason I could be getting this error? I'm pretty sure that I don't have too many files with respect to the /HEADER value. Second, I want to reboot as a prelude to installing patches. (To be on the safe side, I shut down, remove one member of the system-disk shadow set, reboot (might as well do an AUTOGEN before that, as now), apply the patches, reboot and, if all looks OK after a while, add the backup member of the shadow set back in. The idea is to have a valid backup disk in case something goes really wrong.) Can the HEADERFULL error lead to any problems during reboot? Third, is BACKUP really the only way to solve the problem? I could of course boot from CD and BACKUP/NOINIT one member of the shadow set to the other, presumably after having done INIT/HEADER on the target, then boot from the fresh disk. If the HELP is correct, then a proper INIT/HEADER will solve the problem. However, if there is some other cause of the problem (fragmentation?), how sure can I be that BACKUP will solve it? $ pipe sh dev sys$sysdevice/fu |sea sys$pipe block Total blocks 8380080 Sectors per track 113 Free blocks 1070271 Maximum files allowed 419004 Extent cache size 64 Maximum blocks in extent cache 107027 File ID cache size 64 Blocks in extent cache 74880 BACKUP.SYS;1 0/0 BADBLK.SYS;1 0/0 BADLOG.SYS;1 0/0 BITMAP.SYS;1 229/234 CONTIN.SYS;1 0/0 CORIMG.SYS;1 0/0 INDEXF.SYS;1 38608/100143 SECURITY.SYS;1 1/9 VOLSET.SYS;1 0/0 $ sh def SYS$SYSDEVICE:[000000] $ dir/gra [*...]/exc=[.sys*.syscommon...] Grand total of 647 directories, 20312 files. $ dir/gra [*...] Grand total of 1697 directories, 65327 files. ------------------------------ Date: 28 May 2007 09:27:36 -0700 From: Hein RMS van den Heuvel Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: <1180369656.358439.295700@u30g2000hsc.googlegroups.com> The old extention algortime for INDEXF.SYS was poorly done amd coudl cause the header for it to become full easily. That was addressed, but you may well have a system disk from before those changes. BACKUP + re-init will help, and help for good. As a workaroudn you may want to delete some files of look for files with excessive headers. DFU REPORT may just help with that. Hein. ------------------------------ Date: Mon, 28 May 2007 16:42:36 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: In article <1180369656.358439.295700@u30g2000hsc.googlegroups.com>, Hein RMS van den Heuvel writes: Wow, that was quick! Of course, you are problem one of the best people in the world to advise on this issue! Keep monitoring the newsgroup until I feel able to reboot! :-) > The old extention algortime for INDEXF.SYS was poorly done amd coudl > cause the header for it to become full easily. That was addressed, but > you may well have a system disk from before those changes. BACKUP + > re-init will help, and help for good. I'm at 7.3-2 UPDATE 10. Old, but not THAT old, right? The reason for the reboot etc is to apply update 11 and the 9 patches which have since appeared as well as DEC-AXPVMS-DWMOTIF_ECO03-V0103-1-4.PCSI and DEC-AXPVMS-TCPIP_ECO-V0504-156-4.PCSI (due to the known problem, I'll replace TCPIP$IPC_SHR.EXE and UCX$IPC_SHR.EXE with the current versions from ECO 5). > As a workaroudn you may want to delete some files There are no obvious files I could delete. There is no non-system stuff on the system disk, and all files which one can move elsewhere and define a logical to point to them (SYSUAF etc) are not on the system disk. I tried creating a few small files and a 1000-block file and that worked. However, I really don't want the HEADERFULL message when applying the patches! > of look for files > with excessive headers. > DFU REPORT may just help with that. Installing DFU is on my list of things to do. Sigh! Any other way (PIPE with DUMP?) to find files with excessive headers? My main concerns are a) solving the problem enough so that I have a bootable disk after shutting down and removing one member of the shadow set as a backup), b) solving the problem enough so that I can apply the patches and c) solving the problem for good. ------------------------------ Date: Mon, 28 May 2007 16:58:53 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: In article <1180369656.358439.295700@u30g2000hsc.googlegroups.com>, Hein RMS van den Heuvel writes: > The old extention algortime for INDEXF.SYS was poorly done amd coudl > cause the header for it to become full easily. That was addressed, but > you may well have a system disk from before those changes. BACKUP + > re-init will help, and help for good. > > As a workaroudn you may want to delete some files of look for files > with excessive headers. Presumably, if the number of files IS the problem (which, based on the data I posted, I don't think is the case), is it the current number of files, or could high activity, i.e. a large amount of creation/deletion but with roughly constant number of files, cause a problem? SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_RECV_RUN.LOG gets up to about ;4000 every day (spam!). They get purged automatically, and every night I renumber them, so in itself not a problem, but it is a lot of turnover. ------------------------------ Date: Mon, 28 May 2007 17:09:07 GMT From: John Santos Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: In article , helbig@astro.multiCLOTHESvax.de says... > When running AUTOGEN, I got the dreaded > > %SYSTEM-W-HEADERFULL, file header is full > > > ** WARNING ** - Error creating SYS$SYSROOT:[SYSEXE]SYSDUMP.DMP. > > ** ERROR ** - error creating SYSDUMP.DMP. SYSDUMP.DMP needs > to be created manually with 393258 blocks > > Presumably, I would get the same error when trying to create it > manually, right? > > $ dir SYS$SYSROOT:[SYSEXE]SYSDUMP.DMP/siz > > Directory SYS$SYSROOT:[SYSEXE] > > SYSDUMP.DMP;4 215910 > > OK, it wants a larger dump file, and there is room: > > $ sh dev sys$sysdevice > > Device Device Error Volume Free Trans Mnt > Name Status Count Label Blocks Count Cnt > DSA133: Mounted 0 ALPHASYS_3 1070271 488 3 > $33$DKB100: (GLADIA) ShadowSetMember 0 (member of DSA133:) > $33$DKB600: (GLADIA) ShadowSetMember 0 (member of DSA133:) > > $ help/messa HEADERFULL > > HEADERFULL, file header is full > > Facility: SYSTEM, System Services > > Explanation: The file header map area on the volume is full and cannot > be extended. This error occurs only when the file header > extension is inhibited, for example, when the index file is > being extended. > > User Action: Compress the volume by copying it with BACKUP. This condition > can be avoided by increasing the value for the /HEADER > qualifier of the INITIALIZE command. For ODS-2 disks, the > value of the /HEADER qualifier should equal the approximate > number of files you expect to store on the disk. > > Three questions. > > First, I'm sure that the INITIALIZE had a value for /HEADER which is OK. The "HELP" text is accurate, but deceptive. The "Explanation" part is correct, AFAIK. However, the "User Action" only applies to the specific example given in the Explanation, I.e. you were trying to extend the index file ([000000]INDEXF.SYS) and its header was full. You are trying to extend the dump file, which like the index file, is only allowed to have one header (extension headers prohibited.) So the detailed instructions about preallocating a larger index file are completely irrelevant. > I remember thinking a lot about this several years ago and coming up > with a table based on disk size and use of the disk (system, user, > software, data, scratch---this, together with the size, determining the > maximum number of files). Is there some other reason I could be getting > this error? I'm pretty sure that I don't have too many files with > respect to the /HEADER value. See above. The stuff about the index file is just a "for example", and in this case is completely irrelevant. > > Second, I want to reboot as a prelude to installing patches. (To be on > the safe side, I shut down, remove one member of the system-disk shadow > set, reboot (might as well do an AUTOGEN before that, as now), apply the > patches, reboot and, if all looks OK after a while, add the backup > member of the shadow set back in. The idea is to have a valid backup > disk in case something goes really wrong.) Can the HEADERFULL error > lead to any problems during reboot? > No. You'll just continue to use the old, too small, dump file. If you have selective dumps enabled, you'll just get less stuff in it, so if you have a crash, critical debugging info might be missing, but then again, maybe not. This is always the case with selective dumps. The recommended value is just AUTOGEN's best guess as to how big the dump needs to be to capture all the critical info most of the time. If you have full dumps enabled, and the dump file isn't as big as physical memory, then you'll get a truncated dump in the event of a crash, which may be worthless for debugging the crash. Depends on what ends up in the dump file. The dump file size is a compromise between completeness, disk space, and reboot speed. (The bigger the dump file, the longer it takes to copy memory to it.) There is no exact "right size". > Third, is BACKUP really the only way to solve the problem? I could of > course boot from CD and BACKUP/NOINIT one member of the shadow set to > the other, presumably after having done INIT/HEADER on the target, then > boot from the fresh disk. If the HELP is correct, then a proper > INIT/HEADER will solve the problem. However, if there is some other > cause of the problem (fragmentation?), how sure can I be that BACKUP > will solve it? > If you have enough free, mostly contiguous space on the disk to create a new dump file of the right size with few enough extents to fit in one header, then you could do that instead. First rename the existing dump file to something else so it doesn't try to extend it. (For example, sys$system:sysdump.dmp-old). Then execute @sys$update:swapfiles.com and tell it to create a new dump file of the correct size. Or you could execute the trivial SYSGEN command to do the same thing. > $ pipe sh dev sys$sysdevice/fu |sea sys$pipe block > Total blocks 8380080 Sectors per track 113 > Free blocks 1070271 Maximum files allowed 419004 > Extent cache size 64 Maximum blocks in extent cache 107027 > File ID cache size 64 Blocks in extent cache 74880 You have enough space. The question is, how fragmented is it? > > BACKUP.SYS;1 0/0 > BADBLK.SYS;1 0/0 > BADLOG.SYS;1 0/0 > BITMAP.SYS;1 229/234 > CONTIN.SYS;1 0/0 > CORIMG.SYS;1 0/0 > INDEXF.SYS;1 38608/100143 > SECURITY.SYS;1 1/9 > VOLSET.SYS;1 0/0 > > $ sh def > SYS$SYSDEVICE:[000000] > $ dir/gra [*...]/exc=[.sys*.syscommon...] > > Grand total of 647 directories, 20312 files. > > $ dir/gra [*...] > > Grand total of 1697 directories, 65327 files. > > -- John ------------------------------ Date: Mon, 28 May 2007 17:30:09 GMT From: John Santos Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: In article , helbig@astro.multiCLOTHESvax.de says... > In article <1180369656.358439.295700@u30g2000hsc.googlegroups.com>, Hein > RMS van den Heuvel writes: > > Wow, that was quick! Of course, you are problem one of the best people > in the world to advise on this issue! Keep monitoring the newsgroup > until I feel able to reboot! :-) > > > The old extention algortime for INDEXF.SYS was poorly done amd coudl > > cause the header for it to become full easily. That was addressed, but > > you may well have a system disk from before those changes. BACKUP + > > re-init will help, and help for good. Hein is (of course) right, but only if the problem is actually the index file running out of free headers. > > I'm at 7.3-2 UPDATE 10. Old, but not THAT old, right? The reason for > the reboot etc is to apply update 11 and the 9 patches which have since > appeared as well as DEC-AXPVMS-DWMOTIF_ECO03-V0103-1-4.PCSI and > DEC-AXPVMS-TCPIP_ECO-V0504-156-4.PCSI (due to the known problem, I'll > replace TCPIP$IPC_SHR.EXE and UCX$IPC_SHR.EXE with the current versions > from ECO 5). > The system disk doesn't get reallocated or rebuilt when you upgrade, unless you do a backup/image and then upgrade the copy, so the version of VMS is almost irrelevant. I'm running V8.3, but I have a disk on my Alpha (backup/image from an MV-II originally) Directory $1$DKC500:[000000] 000000.DIR;1 36 12-FEB-1981 03:33:07.65 [1,1] (The backup/image was just a couple of years ago, but it could easily be 15 years old!) > > As a workaroudn you may want to delete some files > > There are no obvious files I could delete. There is no non-system stuff > on the system disk, and all files which one can move elsewhere and > define a logical to point to them (SYSUAF etc) are not on the system > disk. > > I tried creating a few small files and a 1000-block file and that > worked. However, I really don't want the HEADERFULL message when > applying the patches! You probably won't get this error if it was due to the dump file's header being full rather than the index file's header being full. DUMP/HEADER/BLOCK=COUNT=0 [0,0]INDEXF.SYS to see how much free space there is in the index file's retrieval pointer area, and the same for sys$system:sysdump.dmp to see if the dump file is out of space in its retrieval pointer area. I bet you'll find the latter is full, not the former. Normally, when the retrieval area gets full, the file system just allocates an extension header (from INDEXF.SYS) and puts the additional retrieval pointers there. However, there are some special files that aren't allowed to have extension headers, usually because the primitive boot file system needs to map them and can't deal with the complexity. INDEXF.SYS and the system dump file are two of these special cases. > > > of look for files > > with excessive headers. > > DFU REPORT may just help with that. > > Installing DFU is on my list of things to do. Sigh! Any other way > (PIPE with DUMP?) to find files with excessive headers? I forgot about DFU... It can tell you lots of useful info, including how fragmented various files are, and how fragmented your free space is, and how many headers in indexf.sys are in use and how many are available. > > My main concerns are a) solving the problem enough so that I have a > bootable disk after shutting down and removing one member of the shadow > set as a backup), b) solving the problem enough so that I can apply the > patches and c) solving the problem for good. Doing a backup/image will solve all these problems as well, but I guess you are trying to avoid that due to the time required. (It won't solve an under-allocated INDEXF.SYS, but it will minimize the number of extents in it, the first time you try to extend it, by creating the n+1st file, it will find a nice big contiguous chunk of disk for its new extent. Or you could initialize the new disk by hand with a big /headers value, and then backup/image/noinit to it.) -- John ------------------------------ Date: Mon, 28 May 2007 12:32:22 -0500 From: David J Dachtera Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: <465B1226.30F3F8D7@spam.comcast.net> Phillip Helbig---remove CLOTHES to reply wrote: > > In article <1180369656.358439.295700@u30g2000hsc.googlegroups.com>, Hein > RMS van den Heuvel writes: > > Wow, that was quick! Of course, you are problem one of the best people > in the world to advise on this issue! Keep monitoring the newsgroup > until I feel able to reboot! :-) > > > The old extention algortime for INDEXF.SYS was poorly done amd coudl > > cause the header for it to become full easily. That was addressed, but > > you may well have a system disk from before those changes. BACKUP + > > re-init will help, and help for good. > > I'm at 7.3-2 UPDATE 10. Old, but not THAT old, right? The reason for > the reboot etc is to apply update 11 and the 9 patches which have since > appeared as well as DEC-AXPVMS-DWMOTIF_ECO03-V0103-1-4.PCSI and > DEC-AXPVMS-TCPIP_ECO-V0504-156-4.PCSI (due to the known problem, I'll > replace TCPIP$IPC_SHR.EXE and UCX$IPC_SHR.EXE with the current versions > from ECO 5). > > > As a workaroudn you may want to delete some files > > There are no obvious files I could delete. There is no non-system stuff > on the system disk, and all files which one can move elsewhere and > define a logical to point to them (SYSUAF etc) are not on the system > disk. > > I tried creating a few small files and a 1000-block file and that > worked. However, I really don't want the HEADERFULL message when > applying the patches! > > > of look for files > > with excessive headers. > > DFU REPORT may just help with that. > > Installing DFU is on my list of things to do. Sigh! Any other way > (PIPE with DUMP?) to find files with excessive headers? > > My main concerns are a) solving the problem enough so that I have a > bootable disk after shutting down and removing one member of the shadow > set as a backup), b) solving the problem enough so that I can apply the > patches and c) solving the problem for good. Well, try this: DJAS02::DDACHTERA$ type dcl$path:more.com $ ipt := sys$input $ if f$trnlnm( "sys$pipe" ) .nes. "" then - $ ipt := sys$pipe $ if p1 .eqs. "" then p1 = ipt $ if f$type( more_pages ) .eqs. "" then - $ more_pages = 64 $ type/page=save='more_pages' &p1 $ exit DJAS02::DDACHTERA$ ..., and then use: $ pipe dump/header/blo=cou=0 | @more -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Mon, 28 May 2007 18:34:20 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: In article , John Santos writes: > The "HELP" text is accurate, but deceptive. The "Explanation" part is > correct, AFAIK. However, the "User Action" only applies to the specific > example given in the Explanation, I.e. you were trying to extend the > index file ([000000]INDEXF.SYS) and its header was full. OK, that explains things and answers questions 1 and 2, i.e. my value for /HEADERS was OK and b) I can reboot with no problem. > The dump file size is a compromise between completeness, disk space, > and reboot speed. (The bigger the dump file, the longer it takes to > copy memory to it.) There is no exact "right size". > If you have enough free, mostly contiguous space on the disk to create > a new dump file of the right size with few enough extents to fit in > one header, then you could do that instead. First rename the existing > dump file to something else so it doesn't try to extend it. (For > example, sys$system:sysdump.dmp-old). Then execute > @sys$update:swapfiles.com and tell it to create a new dump file of the > correct size. Or you could execute the trivial SYSGEN command to do > the same thing. As for the third question, maybe it is a fragmentation problem. However, I can live with the dump file as it is now, so will do a backup to defragment things later on (probably before and/or after upgrading to 8.3, hopefully this summer). ------------------------------ Date: Mon, 28 May 2007 14:35:39 -0400 From: JF Mezei Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: <8dd41$465b211d$cef8887a$28611@TEKSAVVY.COM> Phillip Helbig---remove CLOTHES to reply wrote: > There are no obvious files I could delete. Try ANA/DISK/REPAIR Also, consider some of the very large log files. Those tend to grow and grow and grow, and eat up extents. Purging them, deleting them and starting a new one might free up some space. > I tried creating a few small files and a 1000-block file and that > worked. However, I really don't want the HEADERFULL message when > applying the patches! How full is your disk ? Looks to me like you may have a few files that have eaten up all large free areas, so creating any large file requires a lot of extents instead of one clean continguous area. > Installing DFU is on my list of things to do. Sigh! Any other way > (PIPE with DUMP?) to find files with excessive headers? DIR/SELECT=SIZE=MIN=20000/NOHEAD disk:[000000...] If this system disk is not loaded with excess bagage, it shouldn't have header full errors, unless you have some terribly large log files. (is the web server log file there ? That can eat up disk extents over time) ------------------------------ Date: Mon, 28 May 2007 14:38:28 -0400 From: JF Mezei Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: <214ed$465b21c6$cef8887a$28611@TEKSAVVY.COM> Phillip Helbig---remove CLOTHES to reply wrote: > SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_RECV_RUN.LOG gets up to about ;4000 > every day (spam!). They get purged automatically, and every night I > renumber them, so in itself not a problem, but it is a lot of turnover. $SET DIRECTORY SYS$SPECIFIC:[TCPIP$SMTP]/VERSION=10 $PURGE SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_RECV_RUN.LOG $set file SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_RECV_RUN.LOG;*/version=10 ------------------------------ Date: Mon, 28 May 2007 18:43:45 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: In article , John Santos writes: > > I'm at 7.3-2 UPDATE 10. Old, but not THAT old, right? The reason for > > the reboot etc is to apply update 11 and the 9 patches which have since > > appeared as well as DEC-AXPVMS-DWMOTIF_ECO03-V0103-1-4.PCSI and > > DEC-AXPVMS-TCPIP_ECO-V0504-156-4.PCSI (due to the known problem, I'll > > replace TCPIP$IPC_SHR.EXE and UCX$IPC_SHR.EXE with the current versions > > from ECO 5). > > The system disk doesn't get reallocated or rebuilt when you upgrade, > unless you do a backup/image and then upgrade the copy, so the version > of VMS is almost irrelevant. I'm running V8.3, but I have a disk > on my Alpha (backup/image from an MV-II originally) I started out with 7.1: DEC AXPVMS VMS V7.1 Oper System Install 21-APR-1997 10:37:21 Directory SYS$SYSDEVICE:[000000] 000000.DIR;1 21-APR-1997 09:47:56.89 I don't know if that qualifies as old in this context. However, that was on a 1-GB disk. I later moved to a 2-GB disk and am now at a 4-GB disk. Each time, I did a fresh INIT/HEADERS of the target disk and BACKUP/NOINIT the smaller disk to the larger one. (I might move to 9-GB disks around the time I move to 8.3. I have enough disks; I just need to make sure I have enough shelves, replacements etc.) > You probably won't get this error if it was due to the dump file's > header being full rather than the index file's header being full. Right. > DUMP/HEADER/BLOCK=COUNT=0 [0,0]INDEXF.SYS to see how much free space > there is in the index file's retrieval pointer area, and the same > for sys$system:sysdump.dmp to see if the dump file is out of space > in its retrieval pointer area. I bet you'll find the latter is full, > not the former. Right. ------------------------------ Date: Mon, 28 May 2007 19:10:54 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: In article <214ed$465b21c6$cef8887a$28611@TEKSAVVY.COM>, JF Mezei writes: > Phillip Helbig---remove CLOTHES to reply wrote: > > SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_RECV_RUN.LOG gets up to about ;4000 > > every day (spam!). They get purged automatically, and every night I > > renumber them, so in itself not a problem, but it is a lot of turnover. > > $SET DIRECTORY SYS$SPECIFIC:[TCPIP$SMTP]/VERSION=10 > $PURGE SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_RECV_RUN.LOG > $set file SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_RECV_RUN.LOG;*/version=10 This keeps the number of files low, but there is still the turnover. And you still have to rename when you get to ;32767. What happens in your case when all 10 files are locked and a new one wants to get created? ------------------------------ Date: Mon, 28 May 2007 23:05:48 +0200 From: "P. Sture" Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: In article <8dd41$465b211d$cef8887a$28611@TEKSAVVY.COM>, JF Mezei wrote: > Phillip Helbig---remove CLOTHES to reply wrote: > > There are no obvious files I could delete. > > Try ANA/DISK/REPAIR > > Also, consider some of the very large log files. Those tend to grow and > grow and grow, and eat up extents. Purging them, deleting them and > starting a new one might free up some space. > > > > I tried creating a few small files and a 1000-block file and that > > worked. However, I really don't want the HEADERFULL message when > > applying the patches! > > How full is your disk ? Looks to me like you may have a few files that > have eaten up all large free areas, so creating any large file requires > a lot of extents instead of one clean continguous area. > > > Installing DFU is on my list of things to do. Sigh! Any other way > > (PIPE with DUMP?) to find files with excessive headers? > > DIR/SELECT=SIZE=MIN=20000/NOHEAD disk:[000000...] > > If this system disk is not loaded with excess bagage, it shouldn't have > header full errors, unless you have some terribly large log files. (is > the web server log file there ? That can eat up disk extents over time) Other candidates are the operator log, audit journal, errlog.sys, other tcp/ip component logs, layered component logs and ftp logs in user accounts. Are there any logs created by batch jobs submitted at startup? -- Paul Sture ------------------------------ Date: Tue, 29 May 2007 08:48:08 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: In article , "P. Sture" writes: > > Also, consider some of the very large log files. Those tend to grow and > > grow and grow, and eat up extents. Purging them, deleting them and > > starting a new one might free up some space. > > If this system disk is not loaded with excess bagage, it shouldn't have > > header full errors, unless you have some terribly large log files. (is > > the web server log file there ? That can eat up disk extents over time) > > Other candidates are the operator log, audit journal, errlog.sys, other > tcp/ip component logs, layered component logs and ftp logs in user > accounts. Are there any logs created by batch jobs submitted at startup? All of this stuff is off the system disk. It turns out that the HELP text is somewhat misleading. Everything is fine except the dump file itself. There IS some "wasted" space, namely the PCSI$UNDO stuff, but that is the subject of another post. ------------------------------ Date: Tue, 29 May 2007 12:43:58 +0200 From: Jur van der Burg <"vdburg at hotmail dot com"> Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: <465c03e8$0$327$e4fe514c@news.xs4all.nl> The real problem is just disk fragmentation. Autogen attempted to extend the dumpfile, but the dumpfile header was full and no more mapping pointers are allowed. The dumpfile is special in the sense that no extension headers are allowed for that file. Jur. Phillip Helbig---remove CLOTHES to reply wrote, On 28-5-2007 18:22: > When running AUTOGEN, I got the dreaded > > %SYSTEM-W-HEADERFULL, file header is full > > > ** WARNING ** - Error creating SYS$SYSROOT:[SYSEXE]SYSDUMP.DMP. > > ** ERROR ** - error creating SYSDUMP.DMP. SYSDUMP.DMP needs > to be created manually with 393258 blocks > > Presumably, I would get the same error when trying to create it > manually, right? > > $ dir SYS$SYSROOT:[SYSEXE]SYSDUMP.DMP/siz > > Directory SYS$SYSROOT:[SYSEXE] > > SYSDUMP.DMP;4 215910 > > OK, it wants a larger dump file, and there is room: > > $ sh dev sys$sysdevice > > Device Device Error Volume Free Trans Mnt > Name Status Count Label Blocks Count Cnt > DSA133: Mounted 0 ALPHASYS_3 1070271 488 3 > $33$DKB100: (GLADIA) ShadowSetMember 0 (member of DSA133:) > $33$DKB600: (GLADIA) ShadowSetMember 0 (member of DSA133:) > > $ help/messa HEADERFULL > > HEADERFULL, file header is full > > Facility: SYSTEM, System Services > > Explanation: The file header map area on the volume is full and cannot > be extended. This error occurs only when the file header > extension is inhibited, for example, when the index file is > being extended. > > User Action: Compress the volume by copying it with BACKUP. This condition > can be avoided by increasing the value for the /HEADER > qualifier of the INITIALIZE command. For ODS-2 disks, the > value of the /HEADER qualifier should equal the approximate > number of files you expect to store on the disk. > > Three questions. > > First, I'm sure that the INITIALIZE had a value for /HEADER which is OK. > I remember thinking a lot about this several years ago and coming up > with a table based on disk size and use of the disk (system, user, > software, data, scratch---this, together with the size, determining the > maximum number of files). Is there some other reason I could be getting > this error? I'm pretty sure that I don't have too many files with > respect to the /HEADER value. > > Second, I want to reboot as a prelude to installing patches. (To be on > the safe side, I shut down, remove one member of the system-disk shadow > set, reboot (might as well do an AUTOGEN before that, as now), apply the > patches, reboot and, if all looks OK after a while, add the backup > member of the shadow set back in. The idea is to have a valid backup > disk in case something goes really wrong.) Can the HEADERFULL error > lead to any problems during reboot? > > Third, is BACKUP really the only way to solve the problem? I could of > course boot from CD and BACKUP/NOINIT one member of the shadow set to > the other, presumably after having done INIT/HEADER on the target, then > boot from the fresh disk. If the HELP is correct, then a proper > INIT/HEADER will solve the problem. However, if there is some other > cause of the problem (fragmentation?), how sure can I be that BACKUP > will solve it? > > $ pipe sh dev sys$sysdevice/fu |sea sys$pipe block > Total blocks 8380080 Sectors per track 113 > Free blocks 1070271 Maximum files allowed 419004 > Extent cache size 64 Maximum blocks in extent cache 107027 > File ID cache size 64 Blocks in extent cache 74880 > > BACKUP.SYS;1 0/0 > BADBLK.SYS;1 0/0 > BADLOG.SYS;1 0/0 > BITMAP.SYS;1 229/234 > CONTIN.SYS;1 0/0 > CORIMG.SYS;1 0/0 > INDEXF.SYS;1 38608/100143 > SECURITY.SYS;1 1/9 > VOLSET.SYS;1 0/0 > > $ sh def > SYS$SYSDEVICE:[000000] > $ dir/gra [*...]/exc=[.sys*.syscommon...] > > Grand total of 647 directories, 20312 files. > > $ dir/gra [*...] > > Grand total of 1697 directories, 65327 files. > ------------------------------ Date: 29 May 2007 08:11:48 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: <4VBDl6yz3SPR@eisner.encompasserve.org> In article , helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: > > Third, is BACKUP really the only way to solve the problem? I could of > course boot from CD and BACKUP/NOINIT one member of the shadow set to > the other, presumably after having done INIT/HEADER on the target, then > boot from the fresh disk. If the HELP is correct, then a proper > INIT/HEADER will solve the problem. However, if there is some other > cause of the problem (fragmentation?), how sure can I be that BACKUP > will solve it? > INIT/HEADER will not solve this problem as it controls how many headers a volume can have, not how many allocations a header can track. Your problem has more to do with the number of allocations for a single file. The problem can sometimes be solved through the judicious use of IO$_MOVE. IIRC I put a small program which will do an IO$_MOVE on a single file on the Freeware a few years back. Pagefiles and swapfiles are typically marked /NOMOVE, so simply doing an IO$_MOVE will fail. You need to get the system to stop using that file temporarily, the set the file /MOVE, then make it bigger with SYSGEN (which will mark it /NOMOVE again). You may have to repeat this until the file is large enough. Only once did I have a disk so fragmented that this failed, in which case I did use BACKUP to defrag the entire disk. ------------------------------ Date: Tue, 29 May 2007 11:58:24 -0400 From: norm.raphael@metso.com Subject: Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header is full Message-ID: try $dump/head/bl=end=0 sys$sysdevice:[000000]indexf.sys and see how many pointers/fragments you have in the file. You may need to defrag/compress/delete some (the right) stuff. helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) wrote on 05/28/2007 12:22:13 PM: > When running AUTOGEN, I got the dreaded > > %SYSTEM-W-HEADERFULL, file header is full > > > ** WARNING ** - Error creating SYS$SYSROOT:[SYSEXE]SYSDUMP.DMP. > > ** ERROR ** - error creating SYSDUMP.DMP. SYSDUMP.DMP needs > to be created manually with 393258 blocks > > Presumably, I would get the same error when trying to create it > manually, right? > > $ dir SYS$SYSROOT:[SYSEXE]SYSDUMP.DMP/siz > > Directory SYS$SYSROOT:[SYSEXE] > > SYSDUMP.DMP;4 215910 > > OK, it wants a larger dump file, and there is room: > > $ sh dev sys$sysdevice > > Device Device Error Volume > Free Trans Mnt > Name Status Count Label > Blocks Count Cnt > DSA133: Mounted 0 ALPHASYS_3 > 1070271 488 3 > $33$DKB100: (GLADIA) ShadowSetMember 0 (member of DSA133:) > $33$DKB600: (GLADIA) ShadowSetMember 0 (member of DSA133:) > > $ help/messa HEADERFULL > > HEADERFULL, file header is full > > Facility: SYSTEM, System Services > > Explanation: The file header map area on the volume is full and cannot > be extended. This error occurs only when the file header > extension is inhibited, for example, when the index file is > being extended. > > User Action: Compress the volume by copying it with BACKUP. This condition > can be avoided by increasing the value for the /HEADER > qualifier of the INITIALIZE command. For ODS-2 disks, the > value of the /HEADER qualifier should equal the approximate > number of files you expect to store on the disk. > > Three questions. > > First, I'm sure that the INITIALIZE had a value for /HEADER which is OK. > I remember thinking a lot about this several years ago and coming up > with a table based on disk size and use of the disk (system, user, > software, data, scratch---this, together with the size, determining the > maximum number of files). Is there some other reason I could be getting > this error? I'm pretty sure that I don't have too many files with > respect to the /HEADER value. > > Second, I want to reboot as a prelude to installing patches. (To be on > the safe side, I shut down, remove one member of the system-disk shadow > set, reboot (might as well do an AUTOGEN before that, as now), apply the > patches, reboot and, if all looks OK after a while, add the backup > member of the shadow set back in. The idea is to have a valid backup > disk in case something goes really wrong.) Can the HEADERFULL error > lead to any problems during reboot? > > Third, is BACKUP really the only way to solve the problem? I could of > course boot from CD and BACKUP/NOINIT one member of the shadow set to > the other, presumably after having done INIT/HEADER on the target, then > boot from the fresh disk. If the HELP is correct, then a proper > INIT/HEADER will solve the problem. However, if there is some other > cause of the problem (fragmentation?), how sure can I be that BACKUP > will solve it? > > $ pipe sh dev sys$sysdevice/fu |sea sys$pipe block > Total blocks 8380080 Sectors per track 113 > Free blocks 1070271 Maximum files allowed 419004 > Extent cache size 64 Maximum blocks in extent > cache 107027 > File ID cache size 64 Blocks in extent cache 74880 > > BACKUP.SYS;1 0/0 > BADBLK.SYS;1 0/0 > BADLOG.SYS;1 0/0 > BITMAP.SYS;1 229/234 > CONTIN.SYS;1 0/0 > CORIMG.SYS;1 0/0 > INDEXF.SYS;1 38608/100143 > SECURITY.SYS;1 1/9 > VOLSET.SYS;1 0/0 > > $ sh def > SYS$SYSDEVICE:[000000] > $ dir/gra [*...]/exc=[.sys*.syscommon...] > > Grand total of 647 directories, 20312 files. > > $ dir/gra [*...] > > Grand total of 1697 directories, 65327 files. > ------------------------------ Date: Tue, 29 May 2007 12:18:50 +0800 From: "Richard Maher" Subject: CONVERTing RMS CR/LF files to explicit byte Stream format Message-ID: Hi, This has got to be easy but after floundering around for an hour I can't find it, so could someone please tell me how to take a variable length carriage control file and remove any implicit RMS processing and control info, so that the end result is a "What you see is What you get" UDF file of bytes? In other words if I have a Var Length CR control file, with two records hello world Then I'd like to end up with a Record Format = Undefined and Record Attributes=None helloworld I've played around with convert and FDL and can't get it to work. I don't have to write a program for this do I? Why? I want to perform block i/o on the file and will lose any rms record attributes. Cheers Richard Maher ------------------------------ Date: 29 May 2007 06:14:51 -0700 From: Hein RMS van den Heuvel Subject: Re: CONVERTing RMS CR/LF files to explicit byte Stream format Message-ID: <1180444490.989751.27060@q66g2000hsg.googlegroups.com> On May 29, 12:18 am, "Richard Maher" wrote: > Hi, > > This has got to be easy but after floundering around for an hour I can't > find it, so could someone please tell me how to take a variable length > carriage control file and remove any implicit RMS processing and control > info, so that the end result is a "What you see is What you get" UDF file of > bytes? Transfer ascii to windoze, tranfer back binary and voila ! Just kidding! When using convert I supposed you could do an FDL with RECORD; FORMAT STREAM, but then you'd need to do a SET FILE /ATTR afterwards. Yuck. The program that is supposed to deal with these style conversion is "EXCHANGE /NET" it is little known. It can do your request in once go using: $ cre tmp.fdl record; format undefined; carr none; $ del tmp.txt.* $ exchan/net/transfer=convert=(record=crlf) tmp.tmp tmp.txt/fdl=tmp Hope this helps a little, Hein. ------------------------------ Date: 29 May 2007 08:19:34 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: CONVERTing RMS CR/LF files to explicit byte Stream format Message-ID: In article , "Richard Maher" writes: > Then I'd like to end up with a Record Format = Undefined and Record > Attributes=None > > helloworld > > I've played around with convert and FDL and can't get it to work. I don't > have to write a program for this do I? > > Why? I want to perform block i/o on the file and will lose any rms record > attributes. For this I think you can get what you want by using CONVERT to change the file to stream. Use "analyze/rms/fdl" to get an FDL file describing what you've got, EDIT/FDL to change the recordtype to stream, and CONVERT/FDL to get the above result. Then if you really want RMS to look at it as UDF, just use SET FILE/ATTRIBUTES to make it so. Alternatively you could write a program to read the lines, assemble the data and the cr-lf in blocks, and use block mode write. CONVERT will not by itself do what you want since you want to lose information about the record meta-data that CONVERT preserves. ------------------------------ Date: Tue, 29 May 2007 06:21:17 -0700 From: "Jeffrey H. Coffield" Subject: Re: CONVERTing RMS CR/LF files to explicit byte Stream format Message-ID: Richard Maher wrote: > Hi, > > This has got to be easy but after floundering around for an hour I can't > find it, so could someone please tell me how to take a variable length > carriage control file and remove any implicit RMS processing and control > info, so that the end result is a "What you see is What you get" UDF file of > bytes? > > In other words if I have a Var Length CR control file, with two records > > hello > world > > Then I'd like to end up with a Record Format = Undefined and Record > Attributes=None > > helloworld > > I've played around with convert and FDL and can't get it to work. I don't > have to write a program for this do I? > > Why? I want to perform block i/o on the file and will lose any rms record > attributes. > > Cheers Richard Maher > > > $ dump hw.txt Virtual block number 1 (00000001), 512 (0200) bytes 00646C72 6F770005 006F6C6C 65680005 ..hello...world. 000000 00000000 00000000 00000000 0000FFFF ................ 000010 ... $ type hw.fdl IDENT FDL_VERSION 02 "29-MAY-2007 06:20:24 OpenVMS FDL Editor" FILE ORGANIZATION sequential RECORD BLOCK_SPAN yes CARRIAGE_CONTROL none FORMAT stream $ convert/fdl=hw.fdl hw.seq $ dump hw.seq Virtual block number 1 (00000001), 512 (0200) bytes 00000A0D 646C726F 770A0D6F 6C6C6568 hello..world.... 000000 00000000 00000000 00000000 00000000 ................ 000010 Jeff Coffield ------------------------------ Date: Tue, 29 May 2007 12:14:10 -0500 From: David J Dachtera Subject: Re: CONVERTing RMS CR/LF files to explicit byte Stream format Message-ID: <465C5F62.CE6819B4@spam.comcast.net> "Jeffrey H. Coffield" wrote: > > Richard Maher wrote: > > Hi, > > > > This has got to be easy but after floundering around for an hour I can't > > find it, so could someone please tell me how to take a variable length > > carriage control file and remove any implicit RMS processing and control > > info, so that the end result is a "What you see is What you get" UDF file of > > bytes? > > > > In other words if I have a Var Length CR control file, with two records > > > > hello > > world > > > > Then I'd like to end up with a Record Format = Undefined and Record > > Attributes=None > > > > helloworld > > > > I've played around with convert and FDL and can't get it to work. I don't > > have to write a program for this do I? > > > > Why? I want to perform block i/o on the file and will lose any rms record > > attributes. > > > > Cheers Richard Maher > > > > > > > > $ dump hw.txt > > Virtual block number 1 (00000001), 512 (0200) bytes > > 00646C72 6F770005 006F6C6C 65680005 ..hello...world. 000000 > 00000000 00000000 00000000 0000FFFF ................ 000010 > ... > > $ type hw.fdl > > IDENT FDL_VERSION 02 "29-MAY-2007 06:20:24 OpenVMS FDL Editor" > > FILE > ORGANIZATION sequential > > RECORD > BLOCK_SPAN yes > CARRIAGE_CONTROL none > FORMAT stream > > $ convert/fdl=hw.fdl hw.seq > > $ dump hw.seq > > Virtual block number 1 (00000001), 512 (0200) bytes > > 00000A0D 646C726F 770A0D6F 6C6C6568 hello..world.... 000000 > 00000000 00000000 00000000 00000000 ................ 000010 > > Jeff Coffield Alternately: $ COPY NLA0: STREAM_FILE.DAT $ SET FILE/ATTR=RFM=STM $ APPEND MYFILE.TXT STREAM_FILE.DAT $ SET FILE/ATTR=RFM=UDF ...should do what you want. -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: 29 May 2007 07:36:51 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Do Not Call Registry Message-ID: In article <5bm0adF2pal1oU1@mid.individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: > > Why bother. My home has been recistered since the beginnng but I have > seen no decrease in the number of calls I get. I registered my numbers from the begining. Since then I have received two solicitation calls, both of whom I verbally warned that further contact would lead to a $20,000 fine. I've not heard from anyone twice. And both of these started with "I'm so-and-so from X", so I know who to complain about. It would be hard for me to understand how a sales pitch could not reveal some identification about the company. As for charity and other exempt calls, I find these have not increased. There has been a voluntary registration to stop charity calls for a long time and most charities honor it. I also seem to have convinced my alma matter to stop calling by promising not to send donations in years that they call. I do generally sedn donations in years that they don't call. ------------------------------ Date: Mon, 28 May 2007 19:54:07 -0700 From: Vance Haemmerle Subject: Re: DS10L ATI 7500 PCI Graphics Problems Message-ID: <465B95CF.5040500@toyvax.Glendale.CA.US> FredK wrote: > "Craig A. Berry" wrote in message > news:mcSdnUd357Mhd8vbnZ2dnUVZ_jadnZ2d@speakeasy.net... > >>FredK wrote: >> >>>"Craig A. Berry" wrote in message >>>news:6oednbx4cPA0x8vbnZ2dnUVZ_ualnZ2d@speakeasy.net... >> >>Here's what I have: >> >>http://www.viewsonic.com/products/desktopdisplays/lcddisplays/optiquest/q22wg/ >> >>Screens are getter wider faster than they are getting taller, so I think >>1680 x 1050 is pretty common now in monitors bigger than 20". >> > > > Sure. The flat panel I am using right now for example is 1920x1200. > > It is the 1050 that is the oddball number - if they were trying to build a > HD panel you would think it would be 1080. > > > All of Apple's Cinema Display 20" and iMac 20" and last 17" PowerBook G4 and 17" MacBook Pros are 1680x1050. ------------------------------ Date: Tue, 29 May 2007 10:24:18 -0400 From: "FredK" Subject: Re: DS10L ATI 7500 PCI Graphics Problems Message-ID: "Vance Haemmerle" wrote in message news:465B95CF.5040500@toyvax.Glendale.CA.US... > FredK wrote: > > All of Apple's Cinema Display 20" and iMac 20" and last 17" PowerBook G4 > and 17" MacBook Pros are 1680x1050. Consumer electronics, odd numbers. In any case, the VMS tabled are setup for 1600x1200 If you can pass me along the modeline from a Linux/xFree86 system for the panels, I'll look at adding them. I can generate a modeline like: "1600x1050_60.00" 139.83 1600 1704 1872 2144 1050 1051 1054 1087 -HSync +Vsync or "1680x1050_60.00" 147.14 1680 1784 1968 2256 1050 1051 1054 1087 -HSync +Vsync But those are just CVT suggested timings. I am considering adding a mechanism to allow a user to load a custom entry (that is, the modeline idea from xFree86) - but with a big warning that you can fry your monitor by plugging in bogus values. ------------------------------ Date: Tue, 29 May 2007 15:47:34 +0100 From: "Richard Brodie" Subject: Re: DS10L ATI 7500 PCI Graphics Problems Message-ID: "FredK" wrote in message news:f3btjk$qrn$1@usenet01.boi.hp.com... > Sure. The flat panel I am using right now for example is 1920x1200. > > It is the 1050 that is the oddball number - if they were trying to build a HD panel you > would think it would be 1080. Why we need a 16:10 ratio as well as a 16:9 one is an interesting question. Then again, one might as well ask why having picked the HD resolution of 1920 horizontally, one would choose 1200 and not 1080 vertically. ------------------------------ Date: Tue, 29 May 2007 11:05:23 -0400 From: "FredK" Subject: Re: DS10L ATI 7500 PCI Graphics Problems Message-ID: "Richard Brodie" wrote in message news:f3hee5$ku$1@south.jnrs.ja.net... > > "FredK" wrote in message > news:f3btjk$qrn$1@usenet01.boi.hp.com... > >> Sure. The flat panel I am using right now for example is 1920x1200. >> >> It is the 1050 that is the oddball number - if they were trying to build >> a HD panel you would think it would be 1080. > > Why we need a 16:10 ratio as well as a 16:9 one is an interesting > question. > Then again, one might as well ask why having picked the HD resolution of > 1920 horizontally, one would choose 1200 and not 1080 vertically. In a computer monitor, I expect them to give me every pixel they can push it to. In a TV monitor, I expect them to try to at least give me the closest approximation to 1920 x 1080 - or something that scales to that aspect ratio. Since on the "professional" computer side of things, 1600x1200 LCD panels have been common for a long time, it is just suprising that the consumer panel is 150 pixels shorter. Must be some price/shape compromise. ------------------------------ Date: Mon, 28 May 2007 16:01:09 +0200 From: Jur van der Burg <"lddriver at digiater dot nl"> Subject: Re: DSSI allocation class puzzle Message-ID: <465ae0a7$0$322$e4fe514c@news.xs4all.nl> I would connect all disks to two buses and share them between the systems, and not worry about cluster traffic saturating the buses. Otherwise you would need mscp serving to get data from disks connected to the other system which is far less efficient than direct access on a shared bus. Be aware that DSSI disks have their own controller, and that you have to set the allocation class on such disks on the disk itself (either from the console (set host/dssi) or from vms (set host/dup with FYdriver connected). The system parameter ALLOCLASS has only meaning if you want to mscp serve the disks to another system, and that may also be different depending on the VMS version used. Jur. gerry77@no.spam.mail.com wrote: > On Sun, 27 May 2007 06:02:08 GMT, "John E. Malmberg" > wrote: > >>> I have two VAXen (4000-400 and 4000-500) that I would like to cluster >>> together by using DSSI bus 1 on each system. Both have 3 DSSI disks, >>> connected to their respective bus 0. > > [...] > >> What is your goal on converting these systems to a cluster? > > There's a little detail that disappeared from the final draft of my > post: these are hobbyist systems, everything is just for fun and to > learn something, the hardware is what we were able to pick up at the > dumpster, and we operate them in the spare time :-) > >> What is causing you to need to reserve a DSSI bus for local disk >> activity? Are you currently maxing out the 4MB/Sec bandwidth on that >> bus? Are you even coming close? > > That was just a theory. > >> Are you maintaining two system disks? Do you want to? > > Yes, at least until we'll not find some more DSSI cables. Then we'll be > able to add the DSSI tower to the cluster and we'll have a true quorum > disk and maybe we'll migrate to a common system disk configuration. > >> Are you using volume shadowing? > > A common shadowed system disk with a member on each system would be > feasible? That would not be my first system disk shadowing, but the > first one across two nodes... I'll have to check the docs. > >> If I was concerned about high availability, I would have used only >> external disks on HSD controllers, and volume shadowing. > > What about ethernet failover for cluster communications in the event of > a DSSI cable failure? Again I'll have to check if that is possible. > > Thank you, > G. > > > P.S.: I'm not english native, sorry for any mistake. ------------------------------ Date: Mon, 28 May 2007 21:53:15 -0400 From: Stephen Hoffman Subject: Re: DSSI allocation class puzzle Message-ID: gerry77@no.spam.mail.com wrote: > I have two VAXen (4000-400 and 4000-500) that I would like to cluster > together by using DSSI bus 1 on each system. Both have 3 DSSI disks, > connected to their respective bus 0. I'd connect both DSSI buses between both hosts, and I'd set the host allocation classes and the disk IDE allocation classes (all) into the same allocation class. In deference to Fibre Channel and its allocation class expectations, I tend to recommend new cluster configurations not involving FC be set up in allocation class 3 or higher; FC really wants 1 and 2 for its own use. I'd tend to suggest reserving allocation classes 1 and 2 for FC devices. In this configuration, consider the allocation class to be the way to configure a pool of disks behind two or more servers; in this case, these two VAX 4000 series nodes would be serving some or all of the disks on the DSSI buses to the cluster members accessing storage over the network. With an allocation class, you're looking to establish a pool of disks, where a disk with a particular unit visible via one path is assumed to be the same disk via any other path. It's not the buses, its a pool of disks where you have multiple paths to the storage, and where all unit numbers within the particular pool are unique. Accordingly, you will need to set up your IDE device unit numbers to unique values, and if everything is in, say, allocation class 3, and that's across both of the DSSI buses. JF mentions one scheme for assigning units, where the units are set to the DSSI IDE unit IDs. I'd tend to set the units using a different scheme and to the required unique values across both busses; I'd use units 10, 11, 12, etc on bus 1 and units 20, 21, 22, etc. on bus 2. There's no need to reserve unit 0 (or units 10 or 20, for instance) for use as a system disk. For ease of management and operation, I'd tend to set up a single system disk. Possibly shadowed, if you have the licenses for it. Each VAX 4000 would have a boot root on the shared system disk. As for the hardware, DSSI is SCSI-1 vintage and SCSI-1 speed; not very fast. For its time, it was good, and since you're using this gear, DSSI is a reasonable choice as a cluster communications bus. As for your concerns over performance, I might suggest configuring the box for your requirements and for best use and for uptime, and looking at the system performance -- if that becomes a problem -- later. Knuth was fond of suggesting that optimization be considered for deferral; that premature optimization was to be avoided. And you're not going to get particular speed out of a DSSI bus or IDE disks. But I'd not try to optimize things before you get the configuration installed and going. Starting at http://64.223.189.234/, here are a couple of topics on OpenVMS and the I/O stack: /node/10 /node/114 Cluster quorum information is here: /node/105 And the master copy of the OpenVMS FAQ is available at: /node/1 has a write-up on setting up DSSI, including accessing the IDE parameters that are the settings for the allocation class and for the user-set (forced) unit number field and the unit number value itself. For grins, you can also set the IDE node name. -- www.HoffmanLabs.com Services for OpenVMS ------------------------------ Date: Mon, 28 May 2007 22:28:04 -0400 From: JF Mezei Subject: Re: DSSI allocation class puzzle Message-ID: <3e44a$465b8fd8$cef8887a$6019@TEKSAVVY.COM> Stephen Hoffman wrote: > tend to set the units using a different scheme and to the required > unique values across both busses; I'd use units 10, 11, 12, etc on bus 1 > and units 20, 21, 22, etc. on bus 2. Aren't DSSI units numbers limited from 0 to 7 ? (With 7 usually used by one VAX controller, and using 6 for the other vax controller). And if I remember correctly, disks on the first bus wull be DIA0-7 and those on the second bus will be DIB0-7. (but not entirely sure of that). If so, it means you don't have to worry about the disks being on different allocation classes since they will inherently have different device names. ------------------------------ Date: Tue, 29 May 2007 02:50:27 GMT From: "John E. Malmberg" Subject: Re: DSSI allocation class puzzle Message-ID: JF Mezei wrote: > Stephen Hoffman wrote: > >> tend to set the units using a different scheme and to the required >> unique values across both busses; I'd use units 10, 11, 12, etc on bus >> 1 and units 20, 21, 22, etc. on bus 2. > > Aren't DSSI units numbers limited from 0 to 7 ? There are two numbers. 1 number is the electrical bus position number and is set most commonly by an ID plug. There are also dip switches on the drive for that purpose. I have never seen a configuration where the dip switches were used. The other number is the UNIT number that the IDE will report to the host. This needs to be unique to the allocation class. A common convention is bus number * 10 + plug ID, which works fine if you only have real DSSI disks. If you have HSD controllers, then you can have up to 15 devices per HSD, so a more practical algorithm is: 1000 * Bus ID + 100 * HSD unit ID + SCSI bus ID. > (With 7 usually used by one VAX controller, and using 6 for the other > vax controller). > > And if I remember correctly, disks on the first bus wull be DIA0-7 and > those on the second bus will be DIB0-7. There were several numbering conventions in use, and that can be reenabled with a SYSGEN setting. The KFQSA gives a different controller letter for each IDE unit, so you could see DIA0: DIB1: DIC2:, and you still see it from the console prompt. While it made sense from one perspective, it did not look very ordered. With DSSI built in interfaces on the VAX 3400, and the 4XXX series, the DSSI disks on bus 0 would be DIAn, and bus 1 would be DIBn to the console. The default behavior on for VMS for a while has been that all DSSI disks are DIAnnnn, where nnnnn is set on the IDE. If the UNITNUM has not been set on on the IDE, the default is for it to follow the bus position number. -John wb8tyw@qsl.network Personal Opinion Only ------------------------------ Date: Mon, 28 May 2007 23:23:58 -0400 From: Stephen Hoffman Subject: Re: DSSI allocation class puzzle Message-ID: JF Mezei wrote: > Stephen Hoffman wrote: >> tend to set the units using a different scheme and to the required >> unique values across both busses; I'd use units 10, 11, 12, etc on bus >> 1 and units 20, 21, 22, etc. on bus 2. > > Aren't DSSI units numbers limited from 0 to 7 ? That would be an incorrect assumption. "...Once you are out into the PARAMS subsystem, you can use the FORCEUNI option to force the use of the UNITNUM value and then set a unique UNITNUM inside each DSSI ISE-this causes each DSSI ISE to use the specfied [sic] unit number and not use the DSSI node as the unit number. Other parameters of interest are NODENAME and ALLCLASS, the node name and the (disk or tape) cluster allocation class." "Ensure that all disk unit numbers used within an OpenVMS Cluster disk allocation class are unique, and all tape unit numbers used within an OpenVMS Cluster tape allocation class are also unique." I'll leave you to guess the source of that text. > And if I remember correctly, disks on the first bus wull be DIA0-7 and > those on the second bus will be DIB0-7. (but not entirely sure of that). That would be an incorrect recollection. The OpenVMS host sees all DSSI IDEs as the same DIA device prefix on any release likely to be in use here. (This since VAX/VMS V5.1? or so, when the use of DIcu: was replaced with the use of the A controller letter and of the DIAu: naming across all uses.) > If so, it means you don't have to worry about the disks being on > different allocation classes since they will inherently have different > device names. As the OpenVMS host sees all as the same device prefix, your assumption here would be rather unsafe. There are specific limitations around the KFQSA Q-bus DSSI controller; the discussions here all assume the use of the integrated DSSI controllers available with various of the MicroVAX and VAX 4000 series processors. (I'm a bit fuzzy on the KFPSA series PCI controller, ISTR that controller was a full member of a DSSI bus and allowed host-to-host communications, and not the storage-only access granted via the KFQSA.) -- www.HoffmanLabs.com Services for OpenVMS ------------------------------ Date: Tue, 29 May 2007 12:27:40 GMT From: gerry77@no.spam.mail.com Subject: Re: DSSI allocation class puzzle Message-ID: <307o535mgqvvnahucc5312s72qsvrmp59l@4ax.com> On Mon, 28 May 2007 21:53:15 -0400, Stephen Hoffman wrote: > > I have two VAXen (4000-400 and 4000-500) that I would like to cluster > > together by using DSSI bus 1 on each system. Both have 3 DSSI disks, > > connected to their respective bus 0. > > I'd connect both DSSI buses between both hosts, and I'd set the host > allocation classes and the disk IDE allocation classes (all) into the > same allocation class. Actually I've got only _one_ DSSI cable (I'm just a hobbyist) so I can connect at most one bus between hosts. Which bus do you suggest? The one with the disks (bus 0) or the other one? There would be any performance difference? I'm looking for another DSSI cable, and I'll get it sooner or later. Then which will be the best connection pattern with two cables? Crossed (bus 0 on one host to bus 1 on the other and vice versa) or straight? I know that probably I'll never have any real performance issue with this cluster, and that it would work all the same with every possible bus configuration, because being a hobbyist I'll never use it at his maximum cababilities, but I would like to build it as if it were :-) Thank you, G. ------------------------------ Date: Tue, 29 May 2007 12:27:41 GMT From: gerry77@no.spam.mail.com Subject: Re: DSSI allocation class puzzle Message-ID: On Mon, 28 May 2007 23:23:58 -0400, Stephen Hoffman wrote: > JF Mezei wrote: > > Stephen Hoffman wrote: > >> tend to set the units using a different scheme and to the required > >> unique values across both busses; I'd use units 10, 11, 12, etc on bus > >> 1 and units 20, 21, 22, etc. on bus 2. > > > > Aren't DSSI units numbers limited from 0 to 7 ? > > That would be an incorrect assumption. > > "...Once you are out into the PARAMS subsystem, you can use the > FORCEUNI option to force the use of the UNITNUM value and then set a > unique UNITNUM inside each DSSI ISE-this causes each DSSI ISE to use the > specfied [sic] unit number and not use the DSSI node as the unit number. I know for sure that it's possibile to override the ID plug unit number by setting it with the PARAMS utility inside the ISE, and I've actually done that because of lack of suitable plugs. However I'm wondering if this is still true when the UNITNUM is set to any number higher than 7. The "electrical" ID set with plugs should be quite low level if compared with MSCP and friends: are you saying that during its initialization the controller does a bus scan for every possibile ID, even those higher than 7? Is that a too SCSI-oriented thought? Thank you, G. ------------------------------ Date: Tue, 29 May 2007 13:34:40 +0100 From: "Richard Brodie" Subject: Re: DSSI allocation class puzzle Message-ID: wrote in message news:o17o5398j26vvegc046qqfo13umfvmqpaj@4ax.com... > are you saying that > during its initialization the controller does a bus scan for every > possibile ID, even those higher than 7? There aren't any possible IDs higher than 7, if DSSI arbitration is anything like SCSI. It's only 8 bits wide, so only 8 IDs are available. ------------------------------ Date: Tue, 29 May 2007 13:20:21 GMT From: "John E. Malmberg" Subject: Re: DSSI allocation class puzzle Message-ID: gerry77@no.spam.mail.com wrote: > I know for sure that it's possibile to override the ID plug unit number > by setting it with the PARAMS utility inside the ISE, and I've actually > done that because of lack of suitable plugs. > > However I'm wondering if this is still true when the UNITNUM is set to > any number higher than 7. The "electrical" ID set with plugs should be > quite low level if compared with MSCP and friends: are you saying that > during its initialization the controller does a bus scan for every > possibile ID, even those higher than 7? Is that a too SCSI-oriented > thought? See my reply to J.F. which showed up before Hoff's. The plugs are setting a different thing than UNITNUM. UNITNUM defaults to the plug setting as a matter of convenience. Setting UNITNUM does not substitute for or override the setting of the ID plug. It is possible to get a used disk with UNITNUM forced to a specific and undesired value, as is the same for the allocation class. -John wb8tyw@qsl.network Personal Opinion Only ------------------------------ Date: Mon, 28 May 2007 09:10:33 +0200 From: "Gorazd Kikelj" Subject: Re: FalconStor Message-ID: wrote in message news:1180014926.704832.52970@m36g2000hse.googlegroups.com... > Does anyone out there use a product called "FalconStor". > (used for replication of data to a remote location). ... > Now comes the curious part. If I try to "init" one of the > devices, I get a "medium is offline" message. > > We are pretty much convinced that we are missing something on the > FalconStor side, which is why I am looking for anyone with experience. ... 7.3-2? Be on latest ECO or better on 8.2 latest ECO. On 8.3 there are some problems lurking arount FalconStore and waiting for ECO to be released. Best, Gorazd ------------------------------ Date: Tue, 29 May 2007 07:40:10 -0400 From: "Hal Kuff" Subject: Re: FalconStor and WAN repliaction Message-ID: Hi, OK, you sound internal to HP.... Are you (or HP) working with anyone at Falconstor..? We are in touch with the director of support and the principle developer that writes the Fibre Channel code for OpenVMS and AS400 and they seem to be fuzzy on whether OpenVMS is officially supported... They have no system for test and no customer interest to date.... (There was only one site using replication on record that they can find and that was a POC only and the customer opted out).... We are kind of curious as it looks like a good option for WAN replication where you have lots of bandwidth but high latency, as Falconstor uses UDP... While we are on the subject... High latency (50ms) repliaction is still an issue.... what's everyone doing these days...? --- Dave --- Hal (Kuff at Tessco dot Com) "Gorazd Kikelj" wrote in message news:f3dv9f$h8d$1@usenet01.boi.hp.com... > > wrote in message > news:1180014926.704832.52970@m36g2000hse.googlegroups.com... >> Does anyone out there use a product called "FalconStor". >> (used for replication of data to a remote location). > ... >> Now comes the curious part. If I try to "init" one of the >> devices, I get a "medium is offline" message. >> >> We are pretty much convinced that we are missing something on the >> FalconStor side, which is why I am looking for anyone with experience. > ... > > 7.3-2? Be on latest ECO or better on 8.2 latest ECO. On 8.3 there are > some problems lurking arount FalconStore and waiting for ECO to be > released. > > Best, Gorazd > ------------------------------ Date: Tue, 29 May 2007 15:01:38 +0200 From: "Gorazd Kikelj" Subject: Re: FalconStor and WAN repliaction Message-ID: "Hal Kuff" wrote in message news:tiU6i.190961$jt2.42852@newsfe13.lga... > Hi, > > OK, you sound internal to HP.... Are you (or HP) working with anyone at > Falconstor..? We are in touch with the director of support and the > principle developer that writes the Fibre Channel code for OpenVMS and > AS400 and they seem to be fuzzy on whether OpenVMS is officially > supported... They have no system for test and no customer interest to > date.... (There was only one site using replication on record that they > can find and that was a POC only and the customer opted out).... We are > kind of curious as it looks like a good option for WAN replication where > you have lots of bandwidth but high latency, as Falconstor uses UDP... > > While we are on the subject... High latency (50ms) repliaction is still > an issue.... what's everyone doing these days...? > > --- Dave > --- Hal (Kuff at Tessco dot Com) > I realy don't know. I don't know about support status. If it is on SPD, then it is supported, but as I know, no FalconStore is on. So it is on Falcon side to provide it. I do have a customers with IPstore in place and running on 8.2 and 8.3. That's way I know for problems with 8.3 (have a case opened). Not exactly problems you cited, but I realy strongly suggest to go with latest FIBRE_SCSI ECOs. Best, Gorazd ------------------------------ Date: 29 May 2007 06:54:49 -0700 From: Tom Callahan Subject: Re: FalconStor and WAN repliaction Message-ID: <1180446889.101851.73870@m36g2000hse.googlegroups.com> On May 29, 9:01 am, "Gorazd Kikelj" wrote: > "Hal Kuff" wrote in message > > news:tiU6i.190961$jt2.42852@newsfe13.lga... > > > > > Hi, > > > OK, you sound internal to HP.... Are you (or HP) working with anyone at > >Falconstor..? We are in touch with the director of support and the > > principle developer that writes the Fibre Channel code for OpenVMS and > > AS400 and they seem to be fuzzy on whether OpenVMS is officially > > supported... They have no system for test and no customer interest to > > date.... (There was only one site using replication on record that they > > can find and that was a POC only and the customer opted out).... We are > > kind of curious as it looks like a good option for WAN replication where > > you have lots of bandwidth but high latency, asFalconstoruses UDP... > > > While we are on the subject... High latency (50ms) repliaction is still > > an issue.... what's everyone doing these days...? > > > --- Dave > > --- Hal (Kuff at Tessco dot Com) > > I realy don't know. I don't know about support status. > If it is on SPD, then it is supported, but as I know, no FalconStore > is on. So it is on Falcon side to provide it. > > I do have a customers with IPstore in place and running on 8.2 > and 8.3. That's way I know for problems with 8.3 (have a case opened). > Not exactly problems you cited, but I realy strongly suggest to go with > latest FIBRE_SCSI ECOs. > > Best, Gorazd Gorazd, So you DO have ipstor running on 8.2 OVMS? What options did you have to enable to get ipstor to work successfully. Are you using a command device presented to each ovms host? Thanks, Tom Callahan ------------------------------ Date: Tue, 29 May 2007 16:05:03 +0200 From: "Gorazd Kikelj" Subject: Re: FalconStor and WAN repliaction Message-ID: Hi TOm, will need to ask customer. It was their setup that didn't involve as :-( In core you need to present volumes as HSG80 volumes and it should work. Customer will send me a procedure in a near future. Also I see on FalconStore site, that OpenVMS is listed as supported OS platform. Best, Gorazd "Tom Callahan" wrote in message news:1180446889.101851.73870@m36g2000hse.googlegroups.com... > On May 29, 9:01 am, "Gorazd Kikelj" > wrote: >> "Hal Kuff" wrote in message >> >> news:tiU6i.190961$jt2.42852@newsfe13.lga... >> >> >> >> > Hi, >> >> > OK, you sound internal to HP.... Are you (or HP) working with anyone >> > at >> >Falconstor..? We are in touch with the director of support and the >> > principle developer that writes the Fibre Channel code for OpenVMS and >> > AS400 and they seem to be fuzzy on whether OpenVMS is officially >> > supported... They have no system for test and no customer interest to >> > date.... (There was only one site using replication on record that they >> > can find and that was a POC only and the customer opted out).... We are >> > kind of curious as it looks like a good option for WAN replication >> > where >> > you have lots of bandwidth but high latency, asFalconstoruses UDP... >> >> > While we are on the subject... High latency (50ms) repliaction is >> > still >> > an issue.... what's everyone doing these days...? >> >> > --- Dave >> > --- Hal (Kuff at Tessco dot Com) >> >> I realy don't know. I don't know about support status. >> If it is on SPD, then it is supported, but as I know, no FalconStore >> is on. So it is on Falcon side to provide it. >> >> I do have a customers with IPstore in place and running on 8.2 >> and 8.3. That's way I know for problems with 8.3 (have a case opened). >> Not exactly problems you cited, but I realy strongly suggest to go with >> latest FIBRE_SCSI ECOs. >> >> Best, Gorazd > > Gorazd, > So you DO have ipstor running on 8.2 OVMS? What options did you > have to enable to get ipstor to work successfully. Are you using a > command device presented to each ovms host? > > Thanks, > Tom Callahan > ------------------------------ Date: 29 May 2007 07:40:37 -0700 From: Tom Callahan Subject: Re: FalconStor and WAN repliaction Message-ID: <1180449637.300695.182740@p77g2000hsh.googlegroups.com> On May 29, 10:05 am, "Gorazd Kikelj" wrote: > Hi TOm, > > will need to ask customer. It was their setup that didn't involve as :-( > > In core you need to present volumes as HSG80 volumes and it should work. > Customer will send me a procedure in a near future. > > Also I see on FalconStore site, that OpenVMS is listed as supported OS > platform. > > Best, Gorazd > > "Tom Callahan" wrote in message > > news:1180446889.101851.73870@m36g2000hse.googlegroups.com... > > > On May 29, 9:01 am, "Gorazd Kikelj" > > wrote: > >> "Hal Kuff" wrote in message > > >>news:tiU6i.190961$jt2.42852@newsfe13.lga... > > >> > Hi, > > >> > OK, you sound internal to HP.... Are you (or HP) working with anyone > >> > at > >> >Falconstor..? We are in touch with the director of support and the > >> > principle developer that writes the Fibre Channel code for OpenVMS and > >> > AS400 and they seem to be fuzzy on whether OpenVMS is officially > >> > supported... They have no system for test and no customer interest to > >> > date.... (There was only one site using replication on record that they > >> > can find and that was a POC only and the customer opted out).... We are > >> > kind of curious as it looks like a good option for WAN replication > >> > where > >> > you have lots of bandwidth but high latency, asFalconstoruses UDP... > > >> > While we are on the subject... High latency (50ms) repliaction is > >> > still > >> > an issue.... what's everyone doing these days...? > > >> > --- Dave > >> > --- Hal (Kuff at Tessco dot Com) > > >> I realy don't know. I don't know about support status. > >> If it is on SPD, then it is supported, but as I know, no FalconStore > >> is on. So it is on Falcon side to provide it. > > >> I do have a customers with IPstore in place and running on 8.2 > >> and 8.3. That's way I know for problems with 8.3 (have a case opened). > >> Not exactly problems you cited, but I realy strongly suggest to go with > >> latest FIBRE_SCSI ECOs. > > >> Best, Gorazd > > > Gorazd, > > So you DO have ipstor running on 8.2 OVMS? What options did you > > have to enable to get ipstor to work successfully. Are you using a > > command device presented to each ovms host? > > > Thanks, > > Tom Callahan We have set the hsg80 flags in the /proc/falconstor/falconstor device. We also created the command device as per Falconstor supports recommendations. We can see the device, but cannot access the device, we simply get a "medium offline" error. Any information you could provide would be an immense help. Thanks, Tom Callahan ------------------------------ Date: Tue, 29 May 2007 11:59:42 -0500 From: David J Dachtera Subject: Re: FalconStor and WAN repliaction Message-ID: <465C5BFE.407B8795@spam.comcast.net> Gorazd Kikelj wrote: > > Hi TOm, > > will need to ask customer. It was their setup that didn't involve as :-( > > In core you need to present volumes as HSG80 volumes and it should work. > Customer will send me a procedure in a near future. > > Also I see on FalconStore site, that OpenVMS is listed as supported OS > platform. Do you, perchance, have a URL where I could find that? The folks at work would be interested... (please: a -SPECIFIC- URL, not just http://www.falconstor.com/ ) -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Mon, 28 May 2007 11:34:43 +0200 From: "P. Sture" Subject: Re: Hobbyist Licenses for ISV Layered Products Message-ID: In article , "Richard Maher" wrote: > Hi, > > We're in the process of coming up with a Hobbyist License agreement for > Tier3/hotTIP and was wondering if someone who's already created a Hobbyist > License for a third-party product could answer a couple of questions: - > > - It appears that it is usual to piggy-back the agreement on the > Decus/Encompass/HP-VMS hobbyist license, is that correct? Is having a VMS > Hobbyist agreement normally a pre-condition for issuing a hobbyist license > for a third-party product? > > - If so then how is compliance achieved? Using LMF to check for a single > OPENVMS-HOBBYIST PAK? The Freeware 8.0 CD has a basic example of validating a PAK. http://h71000.www7.hp.com/freeware/freeware80/PAKGEN/ For a quick and dirty "see it in action", compile, link and run example2.c $ r example2 Checking for product FREEWARE MYPROGRAM license. Product sys$grant_license error! ...return status was: 0x0a948e14. End checking for product FREEWARE MYPROGRAM license. $ exit %x0a948e14 %LICENSE-F-NOLICENSE, no license is active for this software product ------ Now copy that to hobbyist_check.c, changing the following 2 lines: #define PRODUCT "MYPROGRAM" #define PRODUCER "FREEWARE" to: #define PRODUCT "OPENVMS-HOBBYIST" #define PRODUCER "DEC" compile, link and run: $ r HOBBYIST_CHECK Checking for product DEC OPENVMS-HOBBYIST license. Product is reported as licensed! End checking for product DEC OPENVMS-HOBBYIST license. I did try changing the date past the expiry date of my OPENVMS-HOBBYIST license, but it didn't fail the above check. I did say "quick and dirty" :-) > - Is the issuing of licenses coordinated? Decus/Montagar shouldering the > burden for all licenses? The OPENVMS-HOBBYIST license is included in the Hobbyist layered license file available from Montagar. > - Where can I get a copy of the exact wording of your Hobbyist License > Agreement? (I have a copy of the old "COMPAQ HOBBY LICENSE AGREEMENT For > OpenVMS" the T's & C's of which appear to be mirrored in other ISV > agreements. Is this the norm?) The full wording is included in both the base VMS (VAX, Alpha, IA64) and layered license files that Montagar e-mail you. A tip for applying those license files. They are supplied as DCL procedures, but on their journey through e-mail systems, my get wrapped. Stick a "$ SET NOON" at the top of them so that during the process of correcting line wraps, you can simply re-run them without having them exit due to "DUPREC, license is already registered" errors. > I have sent e-mails to David Cathay and Hunter Goatley and am yet to hear > back, but am happy to hear from anyone with experience in such matters. > > Cheers Richard Maher -- Paul Sture ------------------------------ Date: Mon, 28 May 2007 11:42:42 +0200 From: "P. Sture" Subject: Re: Hobbyist Licenses for ISV Layered Products Message-ID: In article <465A4FD1.2F25054E@spam.comcast.net>, David J Dachtera wrote: > Register for a Multinet Hobbyist License (you're under no obligation to > download/run Multinet, of course) and view their T&C. > > Also, if you do that, you'll note that PSC uses the hobbyist's VMS PAK > checksum to verify licensability. At a later date than PSC implemented that, the Hobbyist Layered Licenses introduced the OPENVMS-HOBBYIST PAK, which is simpler to check for. > > I'm not sure if you're American - this weekend is our Memorial Day to honor > our Veterans, past and present. So, many folks are focusing on family gatherings > and such this weekend and some may be taking a few days off. Richard is in Oz. It's Whitsuntide in Europe (Monday is a holiday here), and Spring Bank Holiday (last Monday of May) in the UK. It's raining here :-( -- Paul Sture ------------------------------ Date: Mon, 28 May 2007 20:59:23 +0930 From: Mark Daniel Subject: Re: Hobbyist Licenses for ISV Layered Products Message-ID: <135lbut54si37d0@corp.supernews.com> P. Sture wrote: > In article <465A4FD1.2F25054E@spam.comcast.net>, > David J Dachtera wrote: > > >>Register for a Multinet Hobbyist License (you're under no obligation to >>download/run Multinet, of course) and view their T&C. >> >>Also, if you do that, you'll note that PSC uses the hobbyist's VMS PAK >>checksum to verify licensability. > > > At a later date than PSC implemented that, the Hobbyist Layered Licenses > introduced the OPENVMS-HOBBYIST PAK, which is simpler to check for. > > >>I'm not sure if you're American - this weekend is our Memorial Day to honor >>our Veterans, past and present. So, many folks are focusing on family gatherings >>and such this weekend and some may be taking a few days off. I and a colleague were motoring around New Hampshire post Bootcamp 2006 and over the Memorial Day weekend last and remember being most taken by two things: 1) The main streets of all towns we drove through, small and large, bedecked in the Stars and Stripes, along with countless home windows and verandahs displaying Old Glory. Some were obviously at half-staff. It was particularly and unaccountably moving. 2) The number of motor-cyclists, mainly astride Harley-Davidsons as far as I could see, cruising the highways (and byways), along with the number of protective helmets *not* being worn by same. I took both as being instructive of the USA psyche (non-critical). (This is obviously on-topic having mentioned the Bootcamp :-) > Richard is in Oz. It's Whitsuntide in Europe (Monday is a holiday here), > and Spring Bank Holiday (last Monday of May) in the UK. > > It's raining here :-( Wish it was here. -- Chrono-synclastic infundibulum ... those places ... where all the different kinds of truths fit together. [Kurt Vonnegut; The Sirens of Titan] ------------------------------ Date: Mon, 28 May 2007 08:30:44 -0700 From: "Tom Linden" Subject: Re: Hobbyist Licenses for ISV Layered Products Message-ID: On Sun, 27 May 2007 20:25:18 -0700, Richard Maher = wrote: > Hi, > > We're in the process of coming up with a Hobbyist License agreement fo= r > Tier3/hotTIP and was wondering if someone who's already created a = > Hobbyist > License for a third-party product could answer a couple of questions: = - > > - It appears that it is usual to piggy-back the agreement on the > Decus/Encompass/HP-VMS hobbyist license, is that correct? Is having a = VMS > Hobbyist agreement normally a pre-condition for issuing a hobbyist = > license > for a third-party product? > > - If so then how is compliance achieved? Using LMF to check for a sing= le > OPENVMS-HOBBYIST PAK? > > - Is the issuing of licenses coordinated? Decus/Montagar shouldering t= he > burden for all licenses? > > - Where can I get a copy of the exact wording of your Hobbyist License= > Agreement? (I have a copy of the old "COMPAQ HOBBY LICENSE AGREEMENT F= or > OpenVMS" the T's & C's of which appear to be mirrored in other ISV > agreements. Is this the norm?) > > I have sent e-mails to David Cathay and Hunter Goatley and am yet to h= ear > back, but am happy to hear from anyone with experience in such matters= . > > Cheers Richard Maher > > Here is an exerpt of how I do it declare sys$grant_license entry ( char(*), char(*), (1:4) fixed bin(31) optional reference, any reference optional) returns (fixed bin(31)); declare sys$lookup_license entry ( any char(*), any reference, any char(*), bit(32) aligned optional truncate, fixed bin(31) optional truncate, any char(*) optional truncate) returns (fixed bin(31)) external; declare 1 license_items, 2 prod_date, 3 length fixed bin(15) initial (8), 3 item fixed bin(15) initial (lmf$_prod_date), 3 bufadr pointer initial (addr(binary_date)), 3 retlen fixed bin(31) initial (0), 2 prod_version, 3 length fixed bin(15) initial (4), 3 item fixed bin(15) initial (lmf$_prod_version), 3 bufadr pointer initial (addr(compiler_version)), 3 retlen fixed bin(31) initial (0), 2 endlist, 3 length fixed bin(15) initial (0), 3 item fixed bin(15) initial (0), 3 bufadr pointer initial (null()), 3 retlen fixed bin(31) initial (0); dcl lib$signal entry ( fixed binary(31) value, fixed binary(31) value optional truncate, any value truncate list); dcl context(1:4) fixed bin(31) init((4)0); save_status =3D sys$grant_license ('PLI', 'KEDNOS', context, license_ite= ms); if save_status ^=3D1 then do; /* no Kednos license found */ status =3D sys$grant_license ('OPENVMS-HOBBYIST', 'DEC', contex= t, = license_items); if status ^=3D 1 then do; /* no = OPENVMS-HOBBYIST license found */ call lib$signal(save_status); status =3D save_status; end; else if dec(substr(date(),1,4),4) > 0406 then do; /* found = OPENVMS-HOBBYIST license After = 30-JUN-2004 see if there is a = PLI-HOBBYIST license? */ status =3D sys$grant_license ('PLI-HOBBYIST', 'KEDNOS', con= text, = license_items); if status ^=3D 1 then do; /* no */ put skip list('A PLI Hobbyist license was not found.');= put skip list('Please contact Kednos directly for a PL/= I = Hobbyist license'); call lib$signal(save_status); status =3D save_status; end; So having OPENVMS-HOBBYIST license is a prerequisite to running with = PLI-HOBBYIST license, and that license is automatically generated and emailed to the requestor= = from our website and taht license is egistered in out hobbyist lmf database. -- = Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Mon, 28 May 2007 12:16:24 -0500 From: David J Dachtera Subject: Re: Hobbyist Licenses for ISV Layered Products Message-ID: <465B0E68.704F4B6F@spam.comcast.net> "P. Sture" wrote: > > In article <465A4FD1.2F25054E@spam.comcast.net>, > David J Dachtera wrote: > > > Register for a Multinet Hobbyist License (you're under no obligation to > > download/run Multinet, of course) and view their T&C. > > > > Also, if you do that, you'll note that PSC uses the hobbyist's VMS PAK > > checksum to verify licensability. > > At a later date than PSC implemented that, the Hobbyist Layered Licenses > introduced the OPENVMS-HOBBYIST PAK, which is simpler to check for. ...but not guaranteed to be registered as the purpose it serves is, AFAIK, undocumented and not apparent to the casual observer. Any third-party product checking for it would have to be sure and document such plainly and unmistakably. -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Mon, 28 May 2007 12:20:14 -0500 From: David J Dachtera Subject: Re: Hobbyist Licenses for ISV Layered Products Message-ID: <465B0F4E.85F3DA11@spam.comcast.net> Mark Daniel wrote: > [snip] > 2) The number of motor-cyclists, mainly astride Harley-Davidsons as far > as I could see, cruising the highways (and byways), along with the > number of protective helmets *not* being worn by same. > > I took both as being instructive of the USA psyche (non-critical). I was concerned about starting another OT sub-thread. As that's already happened, ... Helmets are good at one thing only: protecting the rider's cranium from blunt-force trauma. There is documented evidence that they greatly increase the risk of neck fracture and/or "internal decapitation" (separation of the skull from the spine). -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Tue, 29 May 2007 11:41:01 +0800 From: "Richard Maher" Subject: Re: Hobbyist Licenses for ISV Layered Products Message-ID: Wow, thanks everybody for the great response! (Both on/off line) Cheers Richard Maher "Richard Maher" wrote in message news:f3dibr$7mq$1@news-01.bur.connect.com.au... > Hi, > > We're in the process of coming up with a Hobbyist License agreement for > Tier3/hotTIP and was wondering if someone who's already created a Hobbyist > License for a third-party product could answer a couple of questions: - > > - It appears that it is usual to piggy-back the agreement on the > Decus/Encompass/HP-VMS hobbyist license, is that correct? Is having a VMS > Hobbyist agreement normally a pre-condition for issuing a hobbyist license > for a third-party product? > > - If so then how is compliance achieved? Using LMF to check for a single > OPENVMS-HOBBYIST PAK? > > - Is the issuing of licenses coordinated? Decus/Montagar shouldering the > burden for all licenses? > > - Where can I get a copy of the exact wording of your Hobbyist License > Agreement? (I have a copy of the old "COMPAQ HOBBY LICENSE AGREEMENT For > OpenVMS" the T's & C's of which appear to be mirrored in other ISV > agreements. Is this the norm?) > > I have sent e-mails to David Cathay and Hunter Goatley and am yet to hear > back, but am happy to hear from anyone with experience in such matters. > > Cheers Richard Maher > > ------------------------------ Date: Mon, 28 May 2007 07:46:28 -0400 From: JF Mezei Subject: Re: Indexed file (RMS) IRC$V_RRV question Message-ID: Is it correct to state that after the bucket split, ghost images of the now moved records would remain in the bucket until thats space is needed for new records ? After the split, the last record of the first bucket gets deleted. Because it is the last record, a "deleted" version of that record must remain in the bucket. I my case, those deleted records contain only the primary key as data. Does this truncation get done automatically by RMS ? Or could it be ALLIN1 that updated the record to its smallest possible size before deleting it ? Now, for the big question: do RRV records go just before the deleted record, just after it ? If not, where do they go ? aka: [logical end of bucket] or [logical end of bucket] And in the case where the last record in a bucket has not been deleted, where would the RRV records be located ? I ask because in my context of not having a usable bkt$w_freespace, I need to make assumptions based on finding the first instance of the deleted record. (Multiple copies of that record can exist since the data can be shifted left whenever other records in that bucket are deleted). Also, will CONVERTing the patched indexed file to a sequential file cause any problem if there are multiple copies of the same record ? (for instance, restoring a ghost copy from bucket 1 and the new current copy from bucket 2), or having multiple copies of a record at the end of a bucket (due to contents shifting to the left and thus leaving ghost images of the record). Will CONVERT complain about the index stating the highest key in bucket 1 is "AAA", but it finds a record "BBB" in that bucket ? ------------------------------ Date: 28 May 2007 09:23:38 -0700 From: Hein RMS van den Heuvel Subject: Re: Indexed file (RMS) IRC$V_RRV question Message-ID: <1180369418.690320.38870@q69g2000hsb.googlegroups.com> On May 28, 7:46 am, JF Mezei wrote: > Is it correct to state that after the bucket split, ghost images of the > now moved records would remain in the bucket until thats space is needed > for new records ? RMS makes no effort to clean up beyond 'next-free-byte' on a split. So a bucket which use to contain A, M, Z and which is split more or less in the middle of M, is likely to show a clean Z record starting at a randomiosh location. With a bit of BAD luck, that old data may be exactly adjacent to the next-free-byte. > After the split, the last record of the first bucket gets deleted. > Because it is the last record, a "deleted" version of that record must > remain in the bucket. Yes. > I my case, those deleted records contain only the primary key as data. > Does this truncation get done automatically by RMS ? Or could it be > ALLIN1 that updated the record to its smallest possible size before > deleting it ? That's RMS doing, not ALLIN1 > > Now, for the big question: > > do RRV records go just before the deleted record, just after it ? If > not, where do they go ? After all valid records. Deleted record CAN be sprinkled throughout the valid record depending on duplication being allowed and RU journalling, But normally RMS expunges them and the only deleted record visible would be the only record in the buckets, just before the RRVs. > Also, will CONVERTing the patched indexed file to a sequential file > cause any problem if there are multiple copies of the same record ? No issue reading, but converting back may need a /SORT and /EXCEP to catch duplicates. In that case teh real Z is likely to show last and end up in the exception file. > Will CONVERT complain about the index stating the highest key in bucket > 1 is "AAA", but it finds a record "BBB" in that bucket ? Well, that would not be CONVERT but plain RMS $GET code, as used by convert to read input. So a SEARCH or TYPE would show the same problem Hein. ------------------------------ Date: 28 May 2007 12:59:42 -0700 From: Hein RMS van den Heuvel Subject: Re: Indexed file (RMS) IRC$V_RRV question Message-ID: <1180382382.616755.18470@m36g2000hse.googlegroups.com> On May 28, 2:29 pm, JF Mezei wrote: > Is it correct to state that if no RRV records follow the deleted record, > then all records that had been moved to another bucket were later > deleted ? When adding records to an indexed file with a key value large then any other key, and this fills the bucket, then RMS does NOT do a bucket split, but just creates a new bucket for that highest key value. The old 'FF' index key above it gets the real high key value for the bucket and a new FF index record is created for the new 'last' bucket. Now one could delete records from that old last bucket which was never split and 'ghosts' could be seen without RRVs as a boundary. > If CONVERT walks the file sequentially from bucket to bucket via the > bkt$l_nxtbkt, would it know about what highest key value a bucket is > supposed to have ? No, it does not. It'll take whatever it gets. BTW... those 'ghosts', but key compression on could generate 'wild' key values. Hein. ------------------------------ Date: Mon, 28 May 2007 11:56:52 +0200 From: "P. Sture" Subject: Re: Is VMS losing the Financial Sector, also? Message-ID: In article , Ron Johnson wrote: > On 05/27/07 15:28, Larry Kilgallen wrote: > > In article <4cl6i.7903$gM1.5288@newsfe21.lga>, Ron Johnson > > writes: > > > >> If you are a good sysadmin and have only installed the packages > >> necessary for your application, then a single "yum update" command > >> on your test box downloads all the *relevant* patch packages and > >> installs them. When you are satisfied that the patches don't hose > >> your system and can schedule an app downtime, run "yum update" again. > > > > So what is the syntax to switch from: > > > > "all the *relevant* patch packages" > > > > to > > > > "all the *relevant* patch packages as of the prior date" > > > > so that the only patch packages installed are those that were tested ? > > Well, you're only testing the patches* for the packages installed on > that box, no? And the package manager "knows" if you've already > installed a patch and if there are or aren't any updates for a given > package. And you can always enumerate specific packages if you > don't want to install every relevant package. And since you've only > installed the bare minimum of packages on your server (who needs a > GUI on a server, anyway??), the list will always be minimal. > > Or am I misunderstanding you. > My interpretation of Larry's question is this: On 28-May-2007 at 11:53, you run yum on a test system to get the relevant patch packages. Once you have tested these (and this can take several days, even weeks in a corporate environment), you decide to implement them on your production systems. How do you specify to yum that you want *only* the patches that were available at 28-May-2007 at 11:53. There is also the subsidiary question of how yum handles packages which may be released while you are running yum. Could a package released at 11:54 be omitted from one run, but included in another, simply due to differences in network speeds on two successive runs? (here I am thinking of the logic whereby BACKUP/RECORD records the date and time of the *start* of the backup, not the time an individual file is backed up). > * Calling them patches is a fiction. They are full-blown updated > packages. Understood, but if shared libraries are being updated, that isn't entirely true for the products using them. > Give a man a fish, and he eats for a day. > Hit him with a fish, and he goes away for good! :-) -- Paul Sture ------------------------------ Date: Mon, 28 May 2007 10:26:52 -0400 From: "Main, Kerry" Subject: RE: Is VMS losing the Financial Sector, also? Message-ID: > -----Original Message----- > From: Ron Johnson [mailto:ron.l.johnson@cox.net] > Sent: May 27, 2007 10:52 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: Is VMS losing the Financial Sector, also? >=20 > On 05/27/07 19:42, Main, Kerry wrote: > >> -----Original Message----- > >> From: Ron Johnson [mailto:ron.l.johnson@cox.net] > >> Sent: May 27, 2007 5:48 PM > >> To: Info-VAX@Mvb.Saic.Com > >> Subject: Re: Is VMS losing the Financial Sector, also? > >> > > > > [snip...] > > > > Well, you have obviously been bitten by the hype and do not manage > large > > DC operations. > > > > That's ok - for you, Linux works fine and that is great. To each his > > own. > > > >> Except when it crashes due to a suck-arse TCP/IP implementation, > >> taking down 3 servers in the middle of the day! > >> > >> > > > > You seem to be not managing the OpenVMS environment the way most > > experienced OpenVMS SysAdmins would or you have some very unique >=20 > Sadly (or fortunately), I'm not the SysAdmin. >=20 > And we never had problems like this before Forte came along. >=20 > > circumstances, or it has never been tuned properly because I have > never > > heard of a TCPIP issue taking out 3 systems. As well, weekly reboots > for > > an application is a sure sign that someone does not understand the > App > > and or the way it has been tuned (system or process parameters). > > > > What is typical for most OpenVMS Customers are long time uptime > examples > > like the following: > > http://www.openvms.org/stories.php?story=3D07/04/13/5402784 > > "On 13 April 2007 at 09:35:50 GMT the Amsterdam Police Cluster > > celebrates 10 years uptime. > > > > This cluster is a multiple site cluster which has been upgraded and > > relocated during the 10 years while maintaining 100% application > > availability." > > > > As I stated, if you are happy with it and Linux works for you, then > that > > is great. >=20 > Whether I'm happy with it or not, It's Coming. The first Alphas are > rolling out and Linux/Oracle systems rolling in supposedly in 1Q08 > (but probably 2Q08). That's a small cluster that does ~1.6M > txn/day. Depending on how well it functions, and how the contracts > read for our 2 big clients (each do ~10M txn/day), VMS will be out > in 2009 and Linux/Solaris/HPUX will be in. >=20 > Any way you cut it, though, VMS is *out* and "Unix" is in. Why? > The customers (who all talk with each other) *specifically* don't > want VMS anymore. >=20 > Just like in 1998 they wanted GUI and in came Forte and fat client > machines. >=20 That's crap .. Customers do not care what platform is being used to deliver the service they are contracting for as long as it is always available, always sub-second response times and it gets magically fixed 5 minutes before it breaks. How many people do you think know or care what runs as the primary platform for Google? Saying "Customers do not want this platform or that platform" any more is really IT's way of promoting their own agenda and OS religion as a means to justify whatever their favourite program of the day is.=20 And that usually boils down to OS religion and hype of key techies in the IT organization who have Managers who are afraid to disagree as it might position them as being to much from the old school. Btw - If you have Customers (internal or external) dictating to you what the solution is vs. what their requirements are, then you are in a whole lot more trouble than just determining what is the next big platform of the day. Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20 OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Mon, 28 May 2007 09:35:03 -0500 From: Ron Johnson Subject: Re: Is VMS losing the Financial Sector, also? Message-ID: On 05/28/07 04:56, P. Sture wrote: > In article , > Ron Johnson wrote: > >> On 05/27/07 15:28, Larry Kilgallen wrote: >>> In article <4cl6i.7903$gM1.5288@newsfe21.lga>, Ron Johnson >>> writes: >>> >>>> If you are a good sysadmin and have only installed the packages >>>> necessary for your application, then a single "yum update" command >>>> on your test box downloads all the *relevant* patch packages and >>>> installs them. When you are satisfied that the patches don't hose >>>> your system and can schedule an app downtime, run "yum update" again. >>> So what is the syntax to switch from: >>> >>> "all the *relevant* patch packages" >>> >>> to >>> >>> "all the *relevant* patch packages as of the prior date" >>> >>> so that the only patch packages installed are those that were tested ? >> Well, you're only testing the patches* for the packages installed on >> that box, no? And the package manager "knows" if you've already >> installed a patch and if there are or aren't any updates for a given >> package. And you can always enumerate specific packages if you >> don't want to install every relevant package. And since you've only >> installed the bare minimum of packages on your server (who needs a >> GUI on a server, anyway??), the list will always be minimal. >> >> Or am I misunderstanding you. >> > > My interpretation of Larry's question is this: > > On 28-May-2007 at 11:53, you run yum on a test system to get the > relevant patch packages. > > Once you have tested these (and this can take several days, even weeks > in a corporate environment), you decide to implement them on your > production systems. > > How do you specify to yum that you want *only* the patches that were > available at 28-May-2007 at 11:53. Oh, ok. Well, typically, you download them into a private mini-repository on your own server and then do all your testing and upgrading from it. That also saves bandwidth so that all 100 servers don't hit the public server. > There is also the subsidiary question of how yum handles packages which > may be released while you are running yum. Could a package released at > 11:54 be omitted from one run, but included in another, simply due to > differences in network speeds on two successive runs? (here I am > thinking of the logic whereby BACKUP/RECORD records the date and time of > the *start* of the backup, not the time an individual file is backed up). The private repository should also take care of that. >> * Calling them patches is a fiction. They are full-blown updated >> packages. > > Understood, but if shared libraries are being updated, that isn't > entirely true for the products using them. I kinda see what you mean, but don't know if I've ever considered it relevant, since shared libraries are always packaged separately. Debian has a sweet little Python script named checkrestart which goes thru the list of open shared library files, looking to see if any "non-existent" files are still being used. (In Unix, even though you can "delete" a file that's being used, it doesn't disappear until it it's not used anymore.) Thus, if you upgrade a fundamental or "mid-range" library, you run this script and it tells you which apps have to be restarted. I don't know if Red Hate has anything like that, but it's shouldn't be hard to port the Debian tool. Of course, you'd do this on your test server first, and have that list ready when you go to upgrade your production machines. >> Give a man a fish, and he eats for a day. >> Hit him with a fish, and he goes away for good! > > :-) > -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Mon, 28 May 2007 10:13:32 -0500 From: Ron Johnson Subject: Re: Is VMS losing the Financial Sector, also? Message-ID: On 05/28/07 09:26, Main, Kerry wrote: [snip] >> >> Just like in 1998 they wanted GUI and in came Forte and fat client >> machines. >> > > That's crap .. Customers do not care what platform is being used to > deliver the service they are contracting for as long as it is always > available, always sub-second response times and it gets magically fixed > 5 minutes before it breaks. Geez, all I can do is shake my head. > How many people do you think know or care what runs as the primary > platform for Google? > > Saying "Customers do not want this platform or that platform" any more > is really IT's way of promoting their own agenda and OS religion as a > means to justify whatever their favourite program of the day is. > > And that usually boils down to OS religion and hype of key techies in > the IT organization who have Managers who are afraid to disagree as it > might position them as being to much from the old school. There's definitely faddism & hype, no doubt about it. If they say "terminals are old-fashioned and GUIs are hip", and we want to win that contract, I *guarantee* you that we're going to toss those VTs and bring in a GUI, no matter how expensive. If (or *when* then *did*) say, "VMS must go!", then guess God Damned what!! Out goes VMS. Why? Because we're wedded to our paychecks, not to VMS. > Btw - If you have Customers (internal or external) dictating to you what > the solution is vs. what their requirements are, then you are in a whole > lot more trouble than just determining what is the next big platform of > the day. We're *government contractors*!!! We do what we have to do to win the contract. See, government agencies either have their own SMEs, or hire consultants. Or both. They are people, and people have biases. And these SMEs let those biases seep into their reports. So, that means figuring out what those biases are, and (if they are at all vaguely reasonable) putting them in your Proposal. Because if said SMEs are *hostile* to VMS, and you're proposal includes VMS, then your chances of a favorable technical review are Very Very Low. -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Mon, 28 May 2007 11:21:20 -0400 From: "John Smith" Subject: Re: Is VMS losing the Financial Sector, also? Message-ID: <67755$465af368$cef89dd0$13551@TEKSAVVY.COM-Free> Not surprising. If you need any help on the conversion, we have some people on-staff who used to be with Forte's software engineering & consulting team and who have a bunch of Siebel/Oracle experience. Some of my customers were amongst the first non-beta V1.0 sites Forte had. Always thought it was a great product - incredibly productive to code in and flexible for the server side - unlike that 3gl write once-crash anywhere Java garbage. Ron Johnson wrote: > On 05/23/07 09:00, John Smith wrote: > [snip] >> >> Being *very* active in the financial services indusrty, I can assure >> you that Oracle, notwithstanding its mult-patches per month, and >> Linux with its patch du jour approach are in fact supplanting most >> things. >> >> There are very few customers still using VMS, and of those, some are >> fully committed to VMS whilst most are looking to migrate to >> unix...usually to Solaris or AIX. I can't recall the last time I saw >> a shop have any significant PHUX in-house. > > We have a bunch of HP-UX boxes because the old CIO got into a big > argument with Sun and replaced them all with HP-UX. (Don't know why > he didn't go with AIX, since this is also a m/f shop.) > > He's gone now, and new management is convinced that Linux is "good > enough", so any new mid-range kit is ProLinant and either Linux or > Windows (dependent on the contract, the app & the specific manager). > > The development group is busily working to port a DEC C, (hack, > spit) Forte' & Rdb app to Java, (gag, groan) Siebel & Oracle on > Linux. (Don't ask why they chose Siebel.) > > Why? Our customers (we're a government contractor) specifically do > not want their (the public's) data in an obscure format on a fading > platform. > > Ron Johnson > New Orleans LA -- OpenVMS - The never-advertised operating system with the dwindling ISV base. ------------------------------ Date: Mon, 28 May 2007 11:43:11 -0500 From: Ron Johnson Subject: Re: Is VMS losing the Financial Sector, also? Message-ID: On 05/28/07 10:21, John Smith wrote: > Not surprising. > > If you need any help on the conversion, we have some people on-staff who > used to be with Forte's software engineering & consulting team and who have > a bunch of Siebel/Oracle experience. > > Some of my customers were amongst the first non-beta V1.0 sites Forte had. > Always thought it was a great product - incredibly productive to code in and > flexible for the server side - unlike that 3gl write once-crash anywhere > Java garbage. I'm sure that the app wasn't implemented well (no one in our group had any OO experience back in 1998), and the framework that consultants set up was a disaster. Contact me off-line and I'll try to put you in touch with the right person. -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Mon, 28 May 2007 12:08:29 -0500 From: David J Dachtera Subject: Re: Is VMS losing the Financial Sector, also? Message-ID: <465B0C8D.BF2FFE9B@spam.comcast.net> "Main, Kerry" wrote: > [snip] > That's crap .. Customers do not care what platform is being used to > deliver the service they are contracting for as long as it is always > available, always sub-second response times and it gets magically fixed > 5 minutes before it breaks. Um, Kerry? Sorry to be the bearer of bad news, but that has not been true since the Alphacide. See the concurrent thread titled, "Anyone know why the Alpha market is so so quiet?". > [snip] > Saying "Customers do not want this platform or that platform" any more > is really IT's way of promoting their own agenda and OS religion as a > means to justify whatever their favourite program of the day is. Actually, I wish that were true. If so, there'd be a lot more customers outside HP's corporate offices with pitchforks, torches, tar and feathers in hand demanding EV9 and later. > And that usually boils down to OS religion and hype of key techies in > the IT organization who have Managers who are afraid to disagree as it > might position them as being to much from the old school. Try again. Our managers would love nothing better than a return to the machines that run z/OS. AIX on p-series is their second choice since Power-5 has such a tremendous performance advantage over EV7z (and, of course, EV7z systems perform better overall than I64 SuperDomes). Power-6 is, of course, "out the door" at IBM and ready for deployment, BTW. AIX-6 is now in field test, from what I hear. ...not to mention that my VMS job agent in Dice.com sent me no less than two IBM jobs for migration specialists (VMS->AIX) Saturday (restricted to the Chicago job market). > Btw - If you have Customers (internal or external) dictating to you what > the solution is vs. what their requirements are, then you are in a whole > lot more trouble than just determining what is the next big platform of > the day. True - but not because of the customers' demands. The customer is, after all, always right (the ones with the gold make the rules!). Only Micro$lop is big enough to dictate that the vendor is always right! Your bosses are blowing sunshine up your skirt again. Be careful about anyone offering you a cup of Kool-Aid... -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Mon, 28 May 2007 14:08:11 -0400 From: JF Mezei Subject: Re: Is VMS losing the Financial Sector, also? Message-ID: Main, Kerry wrote: > How many people do you think know or care what runs as the primary > platform for Google? Google is special because part of their database might go offline and the "system" would still appear to function. It just wouldn't return all hits for what you are looking for. Secondly, people do care what Google runs because they are a guide on how to setup a very large system. If they can do it succesfully with Linux on commodity boxes they build themselves, then other corporations can also rely on Linux. ------------------------------ Date: Mon, 28 May 2007 17:07:43 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: Is VMS losing the Financial Sector, also? Message-ID: <465b448c$0$90274$14726298@news.sunsite.dk> Main, Kerry wrote: > That's crap .. Customers do not care what platform is being used to > deliver the service they are contracting for as long as it is always > available, always sub-second response times and it gets magically fixed > 5 minutes before it breaks. End users: no. If you sell technology solutions to other companies: yes - they do care. > Saying "Customers do not want this platform or that platform" any more > is really IT's way of promoting their own agenda and OS religion as a > means to justify whatever their favourite program of the day is. No - it is a fact of life for all companies selling IT solution to non end users. Arne ------------------------------ Date: 29 May 2007 06:17:58 -0500 From: Kilgallen@SpamCop.net (Larry Kilgallen) Subject: Re: Is VMS losing the Financial Sector, also? Message-ID: In article , Ron Johnson writes: > On 05/27/07 15:28, Larry Kilgallen wrote: >> In article <4cl6i.7903$gM1.5288@newsfe21.lga>, Ron Johnson writes: >> >>> If you are a good sysadmin and have only installed the packages >>> necessary for your application, then a single "yum update" command >>> on your test box downloads all the *relevant* patch packages and >>> installs them. When you are satisfied that the patches don't hose >>> your system and can schedule an app downtime, run "yum update" again. >> >> So what is the syntax to switch from: >> >> "all the *relevant* patch packages" >> >> to >> >> "all the *relevant* patch packages as of the prior date" >> >> so that the only patch packages installed are those that were tested ? > > Well, you're only testing the patches* for the packages installed on > that box, no? And the package manager "knows" if you've already > installed a patch and if there are or aren't any updates for a given > package. No, you are testing those patches on a sacrificial test machine to ensure they do not harm operations on the production machine. But when it comes time to install on the production machines, you want only patches that have been tested on the sacrificial test machine. > And you can always enumerate specific packages if you > don't want to install every relevant package. Thereby removing any advantage of the simple syntax you touted. That syntax is not appropriate for a production environment where effective change control is in place. > * Calling them patches is a fiction. They are full-blown updated > packages. My use is "patch" as in "afterthough due to earlier inadequate design or testing". To say that it is bigger just says that a longer test period is appropriate. ------------------------------ Date: 29 May 2007 06:19:39 -0500 From: Kilgallen@SpamCop.net (Larry Kilgallen) Subject: Re: Is VMS losing the Financial Sector, also? Message-ID: In article , Ron Johnson writes: > See the last paragraph in this post, and ask yourself if NASDAQ > would accept that. > Except when it crashes due to a suck-arse TCP/IP implementation, > taking down 3 servers in the middle of the day! You have a choice of multiple TCP/IP implementations on VMS. ------------------------------ Date: Tue, 29 May 2007 12:36:39 -0400 From: "Main, Kerry" Subject: RE: Is VMS losing the Financial Sector, also? Message-ID: > -----Original Message----- > From: Ron Johnson [mailto:ron.l.johnson@cox.net] > Sent: May 28, 2007 11:14 AM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: Is VMS losing the Financial Sector, also? >=20 > On 05/28/07 09:26, Main, Kerry wrote: > [snip] > >> > >> Just like in 1998 they wanted GUI and in came Forte and fat client > >> machines. > >> > > > > That's crap .. Customers do not care what platform is being used to > > deliver the service they are contracting for as long as it is always > > available, always sub-second response times and it gets magically > fixed > > 5 minutes before it breaks. >=20 > Geez, all I can do is shake my head. >=20 My statement referred to internal IT environment. I do agree that a contractors strategy is typically to win the business and worry about the implementation later. Its why many Customers now dictate fixed price contracts as they understand the old game of ``low balling up front proposals and then up sell with change requests later` game. However, most contractors know that the OS platform NBT (next big thing) changes every 3-4 years. So it is always amazing to hear Customers who still want to chase the golden rainbow every where that it goes. :-) Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20 OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: 28 May 2007 01:23:29 -0700 From: "tomarsin2015@comcast.net" Subject: Looking for a H7868 power supply Message-ID: <1180340609.043141.142600@n15g2000prd.googlegroups.com> If anybody has a spare laying around and wants to sell the item let me know. tks phil ------------------------------ Date: Mon, 28 May 2007 08:49:12 -0500 From: Ron Johnson Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: On 05/28/07 00:41, JF Mezei wrote: > BBC now reports that the owner of the Dubai stock exchange is expected > to make a counterbid for OMX. > > OMX operates in 17 countries. > > BBC's reporter said something akin to ; "OMX also operates a succesful > software business that many people would love to get their hands onto". > > > > Now, if Dubai gets OMX, perhaps they will notice VMS and buy it from HP > to rescue it and then spend whatever money is needed to market and grow it. They're rolling in petro-dollars, so they could definitely afford it. But would it even *occur* to a businessman? -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Mon, 28 May 2007 12:41:00 +0200 From: "P. Sture" Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: In article <00A683FD.3F803566@SendSpamHere.ORG>, VAXman- @SendSpamHere.ORG wrote: > In article , "P. Sture" > writes: > > > > > >In article <00A68303.33AAAAC4@SendSpamHere.ORG>, > > VAXman- @SendSpamHere.ORG wrote: > > > >> In article , "P. > >> Sture" > >> writes: > >> > > >> > > >> >In article , > >> > helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to > >> > reply) wrote: > >> > > >> >> I would rather have no VMS at all (or move into a non-computing job) > >> >> than have some watered-down VMS. > >> > > >> >Something about that statement reminds me of NT... :-( > >> > >> NT is completely submerged, not just watered-down. It's inundated in > >> the mirk and mire that is the WEENDOZE mentality. > > > >I'm seriously wondering how many rats will finally realise with the > >weight of VISTA that that the ship is sinking. > > Nearly everyone I have spoken with that has purchased a new PeeCee with > Vista has installed XP on it. That is what I am hearing too. Various hardware manufacturers haven't got suitable hardware drivers ready yet (and as history repeats itself, some drivers simply won't get updated, hp being a notable offender here). I don't expect corporate clients to be upgrading any time soon. It takes an enormous amount of effort to test current applications and deploy a new platform. M$ recognize this with the availability of "downgrade licenses" A smaller business using QuickBooks? "Should Microsoft be blamed if independent software vendors screw up? Some people are blaming Microsoft for the fact that Intuit ignored programming guidelines and requirements for 6 years. Are they out in left field or do you agree with them?" I'd lay the blame firmly on Intuit here, as they obviously didn't learn from their cock up with QuickBooks help files on NT 10 years ago. (IIRC the installation program had a test for Win 3.n or Win95 - cf conditional compilation directive changes discussed here when VMS went from 2 platforms to 3). Apportioning blame is a bit irrelevant - if something as important as your accounting system breaks, you ain't going there. -- Paul Sture ------------------------------ Date: Mon, 28 May 2007 08:03:19 -0700 From: "Tom Linden" Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: On Mon, 28 May 2007 03:41:00 -0700, P. Sture = wrote: > In article <00A683FD.3F803566@SendSpamHere.ORG>, > VAXman- @SendSpamHere.ORG wrote: > >> In article , "P= . = >> Sture" >> writes: >> > >> > >> >In article <00A68303.33AAAAC4@SendSpamHere.ORG>, >> > VAXman- @SendSpamHere.ORG wrote: >> > >> >> In article ,= = >> "P. >> >> Sture" >> >> writes: >> >> > >> >> > >> >> >In article , >> >> > helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES= to >> >> > reply) wrote: >> >> > >> >> >> I would rather have no VMS at all (or move into a non-computing= = >> job) >> >> >> than have some watered-down VMS. >> >> > >> >> >Something about that statement reminds me of NT... :-( >> >> >> >> NT is completely submerged, not just watered-down. It's inundated= in >> >> the mirk and mire that is the WEENDOZE mentality. >> > >> >I'm seriously wondering how many rats will finally realise with the >> >weight of VISTA that that the ship is sinking. >> >> Nearly everyone I have spoken with that has purchased a new PeeCee wi= th >> Vista has installed XP on it. There are some memory leaks in XP, if I have too many windows open in = Outlook it looses contact with my smtp server MX on VMS and I have to reboot XP.= Simply restarting Outlook doen't do it. > > That is what I am hearing too. > > Various hardware manufacturers haven't got suitable hardware drivers > ready yet (and as history repeats itself, some drivers simply won't ge= t > updated, hp being a notable offender here). > > I don't expect corporate clients to be upgrading any time soon. It tak= es > an enormous amount of effort to test current applications and deploy a= > new platform. M$ recognize this with the availability of "downgrade > licenses" > > A smaller business using QuickBooks? > > &start=3D0> > > "Should Microsoft be blamed if independent software vendors screw up? > > Some people are blaming Microsoft for the fact that Intuit ignored > programming guidelines and requirements for 6 years. Are they out in > left field or do you agree with them?" > > I'd lay the blame firmly on Intuit here, as they obviously didn't lear= n > from their cock up with QuickBooks help files on NT 10 years ago. (IIR= C > the installation program had a test for Win 3.n or Win95 - cf > conditional compilation directive changes discussed here when VMS went= > from 2 platforms to 3). > > Apportioning blame is a bit irrelevant - if something as important as > your accounting system breaks, you ain't going there. > -- = Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Mon, 28 May 2007 16:37:24 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465b3d70$0$90269$14726298@news.sunsite.dk> JF Mezei wrote: > Arne Vajhøj wrote: >> The chairman confessed that in a private conversion with you ? > NASDAQ and OM/OMX don't have much overlap since they don't operate in > the same country. Claims of massive savings are for the gullible only, > unless NASDAQ accepts to use the OM software, which doesn't appear too > lightly (but I remain open to a pleasant surprise). Up can not see cost saving in maintaining one system instead of two systems ???? Arne ------------------------------ Date: Mon, 28 May 2007 16:41:50 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465b3e7c$0$90269$14726298@news.sunsite.dk> Main, Kerry wrote: > Always is in any merger. Some things overlap, some do not. If there > is a great deal of overlap, then what is the added value in the > merger to begin with? ???? It is when the systems overlap that the biggest cost savings occur. > In most mergers, the big carrot is to > merge/acquire a company that has *complimentary* technologies, not > overlapping. That is true for technology "producing" companies, but for technology "consuming" companies it is the opposite. >> The "save a few head count" is something you invented. > > IT Staffing is 60-70% of most IT budgets today. Excatly. Which is why your "save a few head count" is completely wrong. >> In short they plan on saving 100M$ a year primarily on IT. > > Yeah, but it's a whole lot more complicated than simply proclaiming > "make it so .." Very true. But we should assume that it is the plan when they say so. Arne ------------------------------ Date: Mon, 28 May 2007 16:45:02 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465b3f3a$0$90269$14726298@news.sunsite.dk> JF Mezei wrote: > NASDAQ buying OMX is a pride move. It isn't a business move. Which is your completely unsubstantiated hypothesis. Arne ------------------------------ Date: Mon, 28 May 2007 22:49:35 +0200 From: "Dr. Dweeb" Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465b4061$0$21933$157c6196@dreader1.cybercity.dk> "Ron Johnson" wrote in message news:Ebr6i.353475$2Q1.103981@newsfe16.lga... > On 05/27/07 17:22, Dr. Dweeb wrote: > [snip] >> >> But, why do I even have to purchase ANAY OS on the PC? Let's face it, it >> is a non-trivial, in fact more or less impossible task, to get a laptop >> without > > Laptops are a bit of a special case, since they are such highly integrated > beasties. Still... > > http://system76.com/index.php/cPath/1?osCsid=534805a0aa7cf984b154bd2701cbbb94 > http://www.dell.com/content/topics/segtopic.aspx/linux_3x?c=us&cs=19&l=en&s=dhs > http://configure.us.dell.com/dellstore/config.aspx?c=us&cs=19&kc=6V440&l=en&oc=DNCWEL1&s=dhs > http://www.emperorlinux.com/ > >> an OS - or even getting one with XP instead of Wishfor. Same goes for >> cabinet machines. So much for the convicted monopolist changing its >> spots - > > "Cabinet" machine? It that the same as rack-mounted, or pedestal? > > http://www.dell.com/content/topics/segtopic.aspx/e510_nseries?c=us&cs=19&l=en&s=dhs > >> the HW suppliers are still under the thumb and know it. > > Dell has a contractual obligation to MSFT to *not* sell "naked" computers. I thought that was illegal and there was a court case about that. Nothing seems to have changed I guess Ubuntu counts as "free" though I did not check the relative prices - a start I guess. > So, some clever person got the idea to install FreeDOS. This is for those > people who already have Windows site licenses or want to install > Linux/*BSD/Solaris from scratch. > > -- > Ron Johnson, Jr. > Jefferson LA USA > > Give a man a fish, and he eats for a day. > Hit him with a fish, and he goes away for good! ------------------------------ Date: Mon, 28 May 2007 22:54:16 +0200 From: "Dr. Dweeb" Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465b417a$0$21926$157c6196@dreader1.cybercity.dk> "Arne Vajhøj" wrote in message news:465a159f$0$90271$14726298@news.sunsite.dk... > Dr. Dweeb wrote: >> "Arne Vajhøj" wrote in message >> news:4659f22c$0$90266$14726298@news.sunsite.dk... >>> Dr. Dweeb wrote: >>>> AFAIK the OM software is not capable of supporting a NASDAQ sized >>>> exchange, but I may be wrong. >>> Because ? >> >> There is more or less only one criterion - transaction volume. > > I has guessed that "... not capable ... sized ..." was related > to transaction volume. > > But what specifically is the problem ? Can they not buy > a sufficient big Alpha/I64 box that RDB can cope ? Or is > the bottleneck in the application tier ? Or ? > I do not know the specifics, since I did not join the company. I merely heard this during discussions with OM. Dweeb > Arne ------------------------------ Date: Mon, 28 May 2007 22:55:18 +0200 From: "Dr. Dweeb" Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465b41b8$0$21932$157c6196@dreader1.cybercity.dk> "JF Mezei" wrote in message news:8c2ff$465a033f$cef8887a$32578@TEKSAVVY.COM... > Arne Vajhøj wrote: >>> http://www.nasdaq.com/newsroom/news/newsroomnewsStory.aspx?textpath=pr2007\ACQPMZ200705250249PRIMZONEFULLFEED120316.htm&cdtime=05%2f25%2f2007%20+2%3a49AM >> (SEK342 million) estimated revenue synergies. Cost synergies >> will be realized through the rationalization of IT systems >> and data centres, rationalization of non-IT functions, and >> reduced capital and procurement expenditure. Revenue >> synergies will be achieved through the creation of deeper >> liquidity pools, increased cross-border trading, increased >> international listings, packaged data products and enhanced >> technology sales. > > > This is just meaningless PR gobledeegook made by the chairman after having > had a power lunch with a PR executive who found a positive spin on the > economics of this transaction. > > They can rationalise IT and find they lose a lot of revenus because IT is > a significant source of revenu for OM. > > There is also danger that hosting european stocks on USA based systems > will go against european strict data privacy laws. They may also find that > > The real goal isn't cost synergies. It is just NASDAQ stroking its ego by > being able to claim a european presence. > They have been trying for ages. NASDAQ Europe, NASDAQ Germany etc. etc > The cost savings are just made up excuses to try to put a positive > economic spin to this. ------------------------------ Date: Mon, 28 May 2007 23:01:43 +0200 From: "Dr. Dweeb" Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465b4339$0$21927$157c6196@dreader1.cybercity.dk> "Main, Kerry" wrote in message news:FA60F2C4B72A584DBFC6091F6A2B8684023E7DCC@tayexc19.americas.cpqcorp.net... > -----Original Message----- > From: Dr. Dweeb [mailto:spam@dweeb.net] > Sent: May 27, 2007 6:25 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: OM Group acquired by Nasdaq - VMS probably out > > > "Arne Vajhøj" wrote in message > news:4659f22c$0$90266$14726298@news.sunsite.dk... > > Dr. Dweeb wrote: > >> AFAIK the OM software is not capable of supporting a NASDAQ sized > >> exchange, but I may be wrong. > > > > Because ? > > > > Arne > > There is more or less only one criterion - transaction volume. > > Dweeb > Well, fwiw, there has never been a case I am aware of that OpenVMS could not handle when it comes to transaction volumes. A few years ago an internal benchmark was done on some Tru64 and OpenVMS servers using Oracle Enterprise Server. OpenVMS was within +10 and -15% of the Tru64 UNIX numbers which even the competition would acknowledge was one of the fastest UNIX's around. And as many folks here have discovered, OpenVMS keeps getting better from an overall performance perspective with each new release. And Rdb is certainly very capable of very high transaction rates as well . Reference: http://h71000.www7.hp.com/openvms/journal/v3/oracle_rdb.html http://tinyurl.com/yw9t9z (Row cache) Bottom line is that OpenVMS today handles very high transaction rates with numerous stock exchanges around the globe. (Begin - Something wrong with your wrapping) So what is the biggest OMX exchange, and what is it's volume compared to NASDAQ ? I did not say that VMS or Rdb were the specific bottlenecks (I said "OM software"), as these two technologies are clearly state of the art. I suspect the issues are at the application architecture and implementation level (where they ALWAYS are). (End) Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT) OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Mon, 28 May 2007 23:25:01 +0200 From: "Dr. Dweeb" Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465b48b1$0$21933$157c6196@dreader1.cybercity.dk> "Arne Vajhøj" wrote in message news:465b3f3a$0$90269$14726298@news.sunsite.dk... > JF Mezei wrote: >> NASDAQ buying OMX is a pride move. It isn't a business move. > > Which is your completely unsubstantiated hypothesis. > Arne, have you been watching NASDAQ fumbling around in Europe for the past 10 years with extremely limited success? JF is wrong about it not being business, but it certainly is about "big dicks". NASDAQ likes to think of themselves as the "big swinging dick" of exchanges, and their failures in Europe have doubtless damaged the internal corporate image - thus they continue to persue expansion in Europe. They have failed miserably on their own merits, so their current solution is to "buy" market share. I guess it's one way to do it, but I suspect several years from now we will be reading case studies of a technological integration gone awry. I still think openning a book on the longevity of the OM CEO in the new structure is warranted. As for the culture clash between OM (and sweden) and NASDAQ (and USA) - I think storms are ahead. > Arne ------------------------------ Date: Mon, 28 May 2007 17:41:34 -0400 From: JF Mezei Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <57482$465b4cb3$cef8887a$8605@TEKSAVVY.COM> Arne Vajhøj wrote: > Up can not see cost saving in maintaining one system instead of two > systems ???? OM operates the IT systems for 8 stock exhanges ( and sells sotware more many more). It already has a few data centres to do that. Do you really think that NASDAQ could consolidate them into its own systems ? If NASDAQ wants to buy OMX (or the LSE before that), it isn't because it can get cost savings. It is because it has decided, as a strategy, to get into the european markets. OMX's software sideline is not part of the NASDAQ strategy, it is a side effect. The statement about cost synergies, cost savings, staff reductions, greater purchasing power are just anciliary statements to help justify the strategy. But the core reason is the strategy. ------------------------------ Date: Mon, 28 May 2007 16:57:07 -0500 From: Ron Johnson Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: On 05/28/07 15:49, Dr. Dweeb wrote: [snip] >>> the HW suppliers are still under the thumb and know it. >> Dell has a contractual obligation to MSFT to *not* sell "naked" computers. > > I thought that was illegal and there was a court case about that. Probably the "Microsoft tax" is illegal. "Not naked" is probably legal, since there are other options besides MSFT products. > Nothing > seems to have changed Dell is selling end-user PCs w/o Windows and systems w/o Intel. Pigs will soon commence flying out of Urectum[0]. [0] http://en.wikipedia.org/wiki/Urectum -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Mon, 28 May 2007 18:01:12 -0400 From: JF Mezei Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: from: http://ir.nasdaq.com/releasedetail.cfm?ReleaseID=245848 * WORLD EXCHANGE TECHNOLOGY LEADER: OMX has been a pioneer in creating a truly integrated cross-border stock market. OMX also has created a world-renowned technology customer base of equity, debt, and derivatives exchanges with 60 clients in 50 countries worldwide, including Hong Kong, Singapore, Australia, and the US. NASDAQ pioneered electronic trading, and has continued to innovate over the last thirty years and now has the fastest, most efficient trading platform in the US. Together, the Combined Group will provide the technology for the world's increasingly competitive and demanding capital markets; ----------- Notes: OMX "has" been a pionner (past tense) Also , OMX has created a CUSTOMER BASE. NASDAQ pioneered electronic trading and has continuyed to innovate... has the fastest and most efficient trading platform in the USA. So, Nasdaq wants to buy the OM software customer base and sell its own software to them. IBM and Sun have great opportunities coming up. The second NASDAQ sends the message that the OM software platform is to be stopped and customers asked to migrate to the NASDAQ platform, it will be open season for IBM and Sun to acquire those customers who will be faced with a painfful conversion. Unfortunatly, if NASDAQ wants to move customer froms VMS to Tandem, HP's higher management aren't going to lift a finger because they will think that 100% of customers will remain with HP. Lets hope the Dubai stock exchange gets to save OMX. (and indirectly save VMS). ------------------------------ Date: Mon, 28 May 2007 20:16:02 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465b70b0$0$90273$14726298@news.sunsite.dk> JF Mezei wrote: > Arne Vajhøj wrote: >> Up can not see cost saving in maintaining one system instead of two >> systems ???? > > OM operates the IT systems for 8 stock exhanges ( and sells sotware more > many more). It already has a few data centres to do that. Do you really > think that NASDAQ could consolidate them into its own systems ? Maybe. I don't know. I will get back to the main point later in the post. > If NASDAQ wants to buy OMX (or the LSE before that), it isn't because it > can get cost savings. It is because it has decided, as a strategy, to > get into the european markets. OMX's software sideline is not part of > the NASDAQ strategy, it is a side effect. > > The statement about cost synergies, cost savings, staff reductions, > greater purchasing power are just anciliary statements to help justify > the strategy. But the core reason is the strategy. It is amazing that you think you are able to analyze the NASDAQ managements thinking considering that you can not even read my postings as they were intended. "maintaining one system" is not about data center operation but about software development. Arne ------------------------------ Date: Mon, 28 May 2007 20:23:13 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465b725f$0$90273$14726298@news.sunsite.dk> Dr. Dweeb wrote: > "Arne Vajhøj" wrote in message > news:465b3f3a$0$90269$14726298@news.sunsite.dk... >> JF Mezei wrote: >>> NASDAQ buying OMX is a pride move. It isn't a business move. >> Which is your completely unsubstantiated hypothesis. > > Arne, have you been watching NASDAQ fumbling around in Europe for the past > 10 years with extremely limited success? > > JF is wrong about it not being business, but it certainly is about "big > dicks". NASDAQ likes to think of themselves as the "big swinging dick" of > exchanges, and their failures in Europe have doubtless damaged the internal > corporate image - thus they continue to persue expansion in Europe. They > have failed miserably on their own merits, so their current solution is to > "buy" market share. I guess it's one way to do it, but I suspect several > years from now we will be reading case studies of a technological > integration gone awry. You don't think that the CxO's meet once in a while and make fun of IT staff "And then my IT guy wanted to upgrade all the database servers tape system - he claimed it was because the backup could no longer be on one tape, but I know that it was just because his friend got those new tape drives and he want them too" ? I am pretty sure that there has been a very detailed analysis of costs, value and various scenarios. And the fact that mergers is a very difficult art does not indicate that it is just "mine is biggest" that is driving the decision. > I still think openning a book on the longevity of the OM CEO in the new > structure is warranted. > > As for the culture clash between OM (and sweden) and NASDAQ (and USA) - I > think storms are ahead. They will have to learn or leave. Arne ------------------------------ Date: Mon, 28 May 2007 20:26:11 -0400 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465b7310$0$90273$14726298@news.sunsite.dk> JF Mezei wrote: > from: http://ir.nasdaq.com/releasedetail.cfm?ReleaseID=245848 > > * WORLD EXCHANGE TECHNOLOGY LEADER: OMX has been a pioneer in > creating a truly integrated cross-border stock market. OMX > also has created a world-renowned technology customer base > of equity, debt, and derivatives exchanges with 60 clients > in 50 countries worldwide, including Hong Kong, Singapore, > Australia, and the US. NASDAQ pioneered electronic trading, > and has continued to innovate over the last thirty years and > now has the fastest, most efficient trading platform in > the US. Together, the Combined Group will provide the > technology for the world's increasingly competitive and > demanding capital markets; > Also , OMX has created a CUSTOMER BASE. NASDAQ pioneered electronic > trading and has continuyed to innovate... has the fastest and most > efficient trading platform in the USA. I think the keyword is efficient. > So, Nasdaq wants to buy the OM software customer base and sell its own > software to them. Undoubtedly. > IBM and Sun have great opportunities coming up. The second NASDAQ sends > the message that the OM software platform is to be stopped and customers > asked to migrate to the NASDAQ platform, it will be open season for IBM > and Sun to acquire those customers who will be faced with a painfful > conversion. No. Neither IBM or SUN has a stock exchange platform as far as I know. And I am sure the customers will be told that the change will be transparent to them. Arne ------------------------------ Date: Mon, 28 May 2007 22:11:48 -0400 From: JF Mezei Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: Arne Vajhøj wrote: > And the fact that mergers is a very difficult art does not indicate > that it is just "mine is biggest" that is driving the decision. You greatly overestimate new york bankers. There is also a mentality of "if you don't acquire, you will be acquired" which drives those CEOs to find any way to justify some acquisition so they are seen as a predator instead of prey. The irony is that it is shareholders of the prey which end up winning by being given a huge premium for their shares, whereas shareholders of the predator are stuck with higher debts and a few yesr of underperforming while the predator east the prey and discards all the fat it never needed anyways. Note how HP shareholders weren't too keen on Carly's folly. Carly had to buy Deutshe Bank with promises of deals with them in order to convince them to go against their judgement and vote in favour of this folly. ------------------------------ Date: Tue, 29 May 2007 13:14:51 +0800 From: "Richard Maher" Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: Hi Dweeb, > I think the first thing to be phased out will be Magnus Böcker. Shall we > open a book on how long he lasts? Let's make it a two horse race between him and the guy at the ASX that sold the idea of throwing out their home grown SEATS and moving to OM Click :-) That's outsourcing for you. Cheers Richard Maher "Dr. Dweeb" wrote in message news:46575fde$0$21925$157c6196@dreader1.cybercity.dk... > John Smith wrote: > > http://online.wsj.com/article/SB118007353287814521.html?mod=home_whats_news_ > > us > > > > In the article, Nasdaq's Mr. Greifeld described at a conference in > > New York on Monday what he called "the magic of exchange > > consolidation." > > > > "It's about getting to a single platform, sucking the costs out, > > providing a better economic experience for your customer and a > > superior return to your investors," he said. > > > > > > According to what I hear this morning, VMS will be phased out. > > I think the first thing to be phased out will be Magnus Böcker. Shall we > open a book on how long he lasts? > > Dr. Dweeb > > ------------------------------ Date: Tue, 29 May 2007 02:28:22 -0500 From: Ron Johnson Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: On 05/28/07 17:01, JF Mezei wrote: [snip] > Notes: > > OMX "has" been a pionner (past tense) > > Also , OMX has created a CUSTOMER BASE. NASDAQ pioneered electronic > trading and has continuyed to innovate... has the fastest and most > efficient trading platform in the USA. > > So, Nasdaq wants to buy the OM software customer base and sell its own > software to them. One of the few things that Big Larry should have learned from the Oracle purchase of Rdb is that things don't always go as planned. -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Tue, 29 May 2007 08:53:30 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: In article <465b48b1$0$21933$157c6196@dreader1.cybercity.dk>, "Dr. Dweeb" writes: > As for the culture clash between OM (and sweden) and NASDAQ (and USA) - I > think storms are ahead. As someone familiar with both countries (but living in neither at the moment), there is certainly some possibility of culture clash. On the other hand, how big a role that will play in a corporate environment is another question. Of course, there are lots of takeovers, mergers, buyouts, collaborations etc between stock exchanges at the moment, and lots of interviews with executives discussing their strategies (or at least what they want the reader to perceive as being their strategies). Interestingly, I recently read an interview with an OM executive who said he would prefer a merger or whatever with an "anglo-saxon" company to one with a European company. ------------------------------ Date: Tue, 29 May 2007 13:36:45 +0200 From: "Dr. Dweeb" Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465c104f$0$21924$157c6196@dreader1.cybercity.dk> JF Mezei wrote: > Arne Vajhøj wrote: >> And the fact that mergers is a very difficult art does not indicate >> that it is just "mine is biggest" that is driving the decision. > > You greatly overestimate new york bankers. > > There is also a mentality of "if you don't acquire, you will be > acquired" which drives those CEOs to find any way to justify some > acquisition so they are seen as a predator instead of prey. > > The irony is that it is shareholders of the prey which end up winning > by being given a huge premium for their shares, whereas shareholders > of the predator are stuck with higher debts and a few yesr of > underperforming while the predator east the prey and discards all the > fat it never needed anyways. > > > Note how HP shareholders weren't too keen on Carly's folly. Carly had > to buy Deutshe Bank with promises of deals with them in order to > convince them to go against their judgement and vote in favour of > this folly. Having spent time "on the street" and in corporate USA (measured in years and at different companies) I am, sadly, in agreement with JF here. The drive to "grow" at any cost is a built-in corporate instinct. As I mentioned before, NASDAQs efforts have thusfar failed, so they have changed strategy - not desire or goal. Unless you have actually worked on the street, it is hard to actually understand the "big swinging dick" mentality and how it pervades so much activity and decision making. NASDAQ is a "brash" culture. These things are not necessarily bad, but they are oh so real! All that being said, it's still business and the business case will still have to make sense. However, sometimes the business vs. pride conflict results in the wrong victor. Time will tell. Dr. Dweeb. ------------------------------ Date: Tue, 29 May 2007 13:38:15 +0200 From: "Dr. Dweeb" Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <465c10a9$0$21924$157c6196@dreader1.cybercity.dk> Ron Johnson wrote: > On 05/28/07 17:01, JF Mezei wrote: > [snip] >> Notes: >> >> OMX "has" been a pionner (past tense) >> >> Also , OMX has created a CUSTOMER BASE. NASDAQ pioneered electronic >> trading and has continuyed to innovate... has the fastest and most >> efficient trading platform in the USA. >> >> So, Nasdaq wants to buy the OM software customer base and sell its >> own software to them. > > One of the few things that Big Larry should have learned from the > Oracle purchase of Rdb is that things don't always go as planned. Luckily for Larry as it turned out. He thought it "had" gone away until the Peoplesoft deal needed justifying, and Rdb's continued existence seved the day - lol. If it weren't serious it would be funny. Dr. Dweeb ------------------------------ Date: Tue, 29 May 2007 11:56:32 -0400 From: "John Smith" Subject: Re: OM Group acquired by Nasdaq - VMS probably out Message-ID: <413c7$465c4d28$cef89dd0$15517@TEKSAVVY.COM-Free> Arne Vajhøj wrote: > ..... > I am pretty sure that there has been a very detailed analysis of > costs, value and various scenarios. Frankly I'd be amazed if there was *anything* exhaustive analyzed knowing how M&A deals are often done. The analysis won't be done until the money changes hands and the first thing they'll do is fire the wrong staff. -- OpenVMS - The never-advertised operating system with the dwindling ISV base. ------------------------------ Date: 28 May 2007 18:20:13 -0700 From: Sue Subject: Re: OpenVMS 2007 Bootcamp Message-ID: <1180401613.369548.256460@q75g2000hsh.googlegroups.com> On May 27, 12:31 pm, David J Dachtera wrote: > s...@obanion.us wrote: > > > This may line wrap... > > >http://picasaweb.google.com/SeanOBanionPictures/OpenVMSBootcamp2007 > > I wish the picture had included the urinal - would have been more convincing > and, hence, more humorous. > Vaxman, You were missed - this year even the bar keep asked where you were?? not to mention many of the attendees. We had a 19 year old from Norway that you would have enjoyed sparing with. You would have loved the SRO General Sessions, not to mention the engineer petting zoo with equipment. Sue > -- > David J Dachtera > dba DJE Systemshttp://www.djesys.com/ > > Unofficial OpenVMS Marketing Home Pagehttp://www.djesys.com/vms/market/ > > Unofficial Affordable OpenVMS Home Page:http://www.djesys.com/vms/soho/ > > Unofficial OpenVMS-IA32 Home Page:http://www.djesys.com/vms/ia32/ > > Unofficial OpenVMS Hobbyist Support Page:http://www.djesys.com/vms/support/ ------------------------------ Date: Tue, 29 May 2007 10:52:35 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: OpenVMS 2007 Bootcamp Message-ID: <00A68546.6735F066@SendSpamHere.ORG> In article <1180401613.369548.256460@q75g2000hsh.googlegroups.com>, Sue writes: > > >On May 27, 12:31 pm, David J Dachtera >wrote: >> s...@obanion.us wrote: >> >> > This may line wrap... >> >> >http://picasaweb.google.com/SeanOBanionPictures/OpenVMSBootcamp2007 >> >> I wish the picture had included the urinal - would have been more convincing >> and, hence, more humorous. >> Vaxman, >You were missed - this year even the bar keep asked where you were?? I'll bet his sales of Guinness were at an all time low. >not to mention many of the attendees. We had a 19 year old from >Norway that you would have enjoyed sparing with. You would have loved Please elaborate. >the SRO General >Sessions, not to mention the engineer petting zoo with equipment. As you know, too many things happening in my life. I'm hoping that Friday will see an end to at least one of the happening things. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: 29 May 2007 07:41:40 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: OpenVMS 2007 Bootcamp Message-ID: In article <1180102785.034506.109100@m36g2000hse.googlegroups.com>, Galen writes: > >> >> >Still dreaming about VMS on your TRS-80 or Sinclair, eh? ;-) >> >> No... but if it would run on my old trusty slide rule I'd pull mine out from >> the mothballs. > > > How about on an abacus? (It would have to be a very large one > indeed. :-) No, that would be too slow. But I'm still waiting for FreeVMS to be completely implemented and soley in C so I can port it to my Pro 350. ------------------------------ Date: 29 May 2007 02:45:21 -0700 From: wim.de-boer@corusgroup.com Subject: openvms pascal rdb "IN" predicate parameterparsing Message-ID: <1180431921.669494.235700@p47g2000hsd.googlegroups.com> Hello PASCAL / RDB-friends, Working on openVMS 8.2 on Alpha DS15 using Pascal V5.9-95 and Oracle Rdb V7.1-451. I'm trying to pass a varying amount of strings to an IN-predicate, please see the example -- start example MODULE Sql_plts_mgp_adv_curs LANGUAGE Pascal ALIAS Rdb$dbhandle PARAMETER COLONS DECLARE ALIAS FOR FILENAME BLITS_DB DECLARE Bezette_vakken_in_rij READ ONLY CURSOR FOR SELECT matr_park_lok_x, matr_park_lok_y FROM materiaal_eenheid WHERE matr_park_id = :Park AND matr_proces_sts = 'IN' AND matr_park_lok_x IN (:Strng) GROUP BY matr_park_lok_x, matr_park_lok_y PROCEDURE Open_bezette_vakken_in_rij Sqlcode :Park RECORD FROM 'CDD$DEFAULT.DOM_PARK_INST_ID' END RECORD :Strng varchar(80); OPEN Bezette_vakken_in_rij; PROCEDURE Fetch_bezette_vakken_in_rij Sqlcode :Lok_X RECORD FROM 'CDD$DEFAULT.DOM_X_LOK' END RECORD :Lok_Y RECORD FROM 'CDD$DEFAULT.DOM_Y_LOK' END RECORD; FETCH Bezette_vakken_in_rij INTO :Lok_X,:Lok_Y; PROCEDURE Close_bezette_vakken_in_rij Sqlcode; CLOSE Bezette_vakken_in_rij; -- end example Pascal definitions in a module ext_sqlmod environment('rdb$build:ext_sqlmod')] module ext_sqlmod; type $quad = [hidden,quad] record l0 : unsigned; l1 : integer; end; [external] procedure Open_bezette_vakken_in_rij (var sqlcode : integer; Park : t_park_inst_id; Strng : varying [u] of char ); external; [external] procedure Fetch_bezette_vakken_in_rij ( var Sqlcode : integer; var Lok_x : t_x_lok; var Lok_y : t_y_lok ); external; [external] procedure Close_bezette_vakken_in_rij (var sqlcode : integer); external; {tenslotte de deklaratie van de gegenereerde sql procedures : } %include 'rdb$build:build$_ext_sqlmod.pin' end. {module ext_sqlmod} part of the code using this sql-modules FUNCTION Zoek_voorkeur_rij : BOOLEAN; VAR vervoer_str : VARYING [80] OF CHAR; intv : INTEGER; str1 : VARYING [8] OF CHAR; eerste : BOOLEAN; lv_sqlcode : INTEGER; lv_drgn : BOOLEAN; vk_vak : Voorkeur_record; bezet_str : VARYING [800] OF CHAR; plaatsen : plek; i : INTEGER; c : CHAR; lv_x : T_X_lok; lv_y : T_Y_lok; x_str : varying [8] OF CHAR; lv_result : BOOLEAN; aant_vrij_per_rij : vrij_per_rij; meeste_vrij, rij_met_meeste_vrij : INTEGER; vrij_vak_gevonden : BOOLEAN; BEGIN {Zoek_voorkeur_rij} Zoek_voorkeur_rij := FALSE; vervoer_str := ''; bezet_str := ''; aant_vrij_per_rij := vrij_per_rij (UPPER(verv) OF 0); FOR i := 1 TO UPPER(verv) DO BEGIN WRITEV(x_str,i:0,ERROR:=CONTINUE); Alg$skip_lead_spaces(x_str); Alg$skip_trail_spaces(x_str); lv_x := x_str; FOR c := 'A' TO 'Z' DO BEGIN IF c IN BC_Geldig_Vakletters THEN BEGIN lv_y := c + ' '; lv_result := Blits$_check_parkgrenzen(huidig_park,lv_x,lv_y); IF lv_result THEN BEGIN plaatsen[i,c] := 'V'; {mogelijk vrij} aant_vrij_per_rij[i] := aant_vrij_per_rij[i] + 1; END ELSE plaatsen[i,c] := 'X'; {geen geldige XY-lok in dit park} END ELSE plaatsen[i,c] := 'X'; {geen geldige XY-lok in dit park} END; END; eerste := TRUE; FOR intv := LOWER(vervoer_rij) TO UPPER(vervoer_rij) DO BEGIN IF (intv IN vervoer_rij) THEN BEGIN str1 := ''; WRITEV(str1,intv:0,ERROR:=CONTINUE); IF eerste THEN BEGIN vervoer_str := '''' + str1 + ''''; eerste := false; END ELSE vervoer_str := vervoer_str + ',''' + str1 + ''''; END; END; vervoer_str := '''' + '98' + ''''; { vervoer_str := '98,114' ;} { vervoer_str := '98' ;} vervoer_str := pad(vervoer_str,' ',80); lv_drgn := TRUE; Open_bezette_vakken_in_rij (lv_sqlcode, Huidig_park,vervoer_str); {leest de bezette vakken / rijen in de voorkeurrijen} IF (lv_sqlcode <> Sqlcode_Success) THEN BEGIN lv_drgn := FALSE; END; IF lv_drgn THEN BEGIN REPEAT vk_vak := ZERO; WITH vk_vak DO Fetch_bezette_vakken_in_rij(lv_sqlcode, Park_Lok_X,Park_Lok_Y); IF (lv_sqlcode = Sqlcode_EOS) THEN BEGIN lv_drgn := FALSE; END; IF lv_sqlcode = Sqlcode_Success THEN BEGIN READV(vk_vak.Park_Lok_X,i, ERROR:= CONTINUE); c := vk_vak.Park_Lok_Y[1]; IF i < UPPER(verv) THEN BEGIN plaatsen[i,c] := 'B'; {bezet} aant_vrij_per_rij[i] := aant_vrij_per_rij[i] - 1; END; END ELSE BEGIN IF lv_sqlcode <> Sqlcode_EOS THEN BEGIN lv_drgn := FALSE; END; END; UNTIL (lv_sqlcode <> Sqlcode_Success); Close_bezette_vakken_in_rij (lv_sqlcode); END; { de bezette vakken / rijen hebben we } { tevens het aantal vrije vakken per rij } { zoek rij met meeste ruimte } IF lv_drgn THEN BEGIN meeste_vrij := 0; FOR i := 1 TO UPPER(verv) DO BEGIN IF aant_vrij_per_rij[i] > meeste_vrij THEN rij_met_meeste_vrij := i; END; vrij_vak_gevonden := FALSE; c := 'A'; REPEAT IF plaatsen[rij_met_meeste_vrij,c] = 'V' THEN vrij_vak_gevonden := TRUE; c := CHR (ORD(c) + 1); UNTIL ((vrij_vak_gevonden) OR_ELSE (NOT(c IN ['A'..'Z']))); IF vrij_vak_gevonden THEN BEGIN WRITEV(x_str,rij_met_meeste_vrij,ERROR:=CONTINUE); REPEAT IF plaatsen[rij_met_meeste_vrij,c] = 'V' THEN vrij_vak_gevonden := TRUE; c := CHR (ORD(c) + 1); UNTIL ((vrij_vak_gevonden) OR_ELSE (NOT(c IN ['A'..'Z']))); IF vrij_vak_gevonden THEN BEGIN WRITEV(x_str,rij_met_meeste_vrij,ERROR:=CONTINUE); Alg$skip_lead_spaces(x_str); Alg$skip_trail_spaces(x_str); Adv_x := x_str; Adv_y := c; Zoek_voorkeur_rij := TRUE; END; END; END; {Zoek_voorkeur_rij} Problem: The Fetch nether ends successfull when passing more then one string (like this: '98','113') Using SQL directly this do workdev_bso_nt> ty rijmin.sql sel m1.matr_park_lok_x, m1.matr_park_lok_y from materiaal_eenheid m1 where m1.matr_park_id = 'W_P' and m1.matr_proces_sts = 'IN' and m1.matr_park_lok_x IN ('98','114') group by m1.matr_park_lok_x, m1.matr_park_lok_y ; commit; dev_bso_nt> Please is there someone in the wide world who can help me??? With regard's to all off you Wim de Boer ------------------------------ Date: Tue, 29 May 2007 07:06:28 -0500 From: Ron Johnson Subject: Re: openvms pascal rdb "IN" predicate parameterparsing Message-ID: <9HU6i.88851$vE1.19866@newsfe24.lga> On 05/29/07 04:45, wim.de-boer@corusgroup.com wrote: > Hello PASCAL / RDB-friends, > Working on openVMS 8.2 on Alpha DS15 using Pascal V5.9-95 and Oracle > Rdb V7.1-451. > > I'm trying to pass a varying amount of strings to an IN-predicate, > please see the example Ah, I don't think you can do that. I would suggest using a global temporary table. At the SQL> prompt, CREATE GLOBAL TEMPORARY TABLE GTT_FOOBAR ( VAL CHAR(10)) ON COMMIT PRESERVE ROWS; Then, in your program: DECLARE Bezette_vakken_in_rij READ ONLY CURSOR FOR SELECT matr_park_lok_x, matr_park_lok_y FROM materiaal_eenheid WHERE matr_park_id = :Park AND matr_proces_sts = 'IN' AND matr_park_lok_x IN (SELECT SOME_VAL FROM GTT_FOOBAR) GROUP BY matr_park_lok_x, matr_park_lok_y Before you call Open_bezette_vakken_in_rij, delete from GTT_FOOBAR, then make a loop that inserts all the necessary records into GTT_FOOBAR; -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Mon, 28 May 2007 10:05:07 -0700 From: "Tom Linden" Subject: OT Helmets: Hobbyist Licenses for ISV Layered Products Message-ID: A little discipline is a good thing. On Mon, 28 May 2007 10:20:14 -0700, David J Dachtera wrote: > Mark Daniel wrote: >> [snip] >> 2) The number of motor-cyclists, mainly astride Harley-Davidsons as far >> as I could see, cruising the highways (and byways), along with the >> number of protective helmets *not* being worn by same. >> >> I took both as being instructive of the USA psyche (non-critical). > > I was concerned about starting another OT sub-thread. As that's already > happened, ... > > Helmets are good at one thing only: protecting the rider's cranium from > blunt-force trauma. > > There is documented evidence that they greatly increase the risk of neck > fracture and/or "internal decapitation" (separation of the skull from the > spine). > -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Mon, 28 May 2007 23:08:52 GMT From: winston@SSRL.SLAC.STANFORD.EDU (Alan Winston - SSRL Central Computing) Subject: PCSI problems on 8.3 - what to do? Message-ID: <00A684CA.F27AF107@SSRL.SLAC.STANFORD.EDU> VMSers -- DS20E VMS 8.3-1 with update 100 installed. Hadn't used PRODUCT INSTALL for several months. Now, I tried it and was told that a bunch of .pcsi$database files were missing. (Curiously, these were all for long-since-supplanted products; old versions of JAVA, CSWS, etc.) I have no idea how long ago these disappeared, but they aren't on my most recent backups of my system disk. I don't know how they disappeared. I'm not going to reinstall the old versions, and I haven't saved the kits of stuff that's many years and many versions back. Is there some way to tell PCSI to just forget about those? (I tried creating dummy files with the appropriate names for the missing databases, and a $ PRODUCT ANALYZE PDB is now happy, but if I do $ PRODUCT INSTALL * it aborts with an access violation.) Any recommendations? (Maybe something other than "you should have kept system disk backups forever".) On a related topic, when I download the relevant patches for update 200 from ITRC via PC and then saved them to VMS disk using Samba, I had problems with the big zip file. My unZIP thought it was corrupt. Anybody else seeing this kind of problem? Thanks, -- Alan ------------------------------ Date: 29 May 2007 02:12:58 -0700 From: IanMiller Subject: Re: PCSI problems on 8.3 - what to do? Message-ID: <1180429978.830693.169410@p77g2000hsh.googlegroups.com> what version of unzip ? ------------------------------ Date: Tue, 29 May 2007 09:08:27 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: PCSI, disk space, UNDO, unseen dangers etc Message-ID: I have about 1½ million blocks (of about 8 million total) on the system disk in PCSI$UNDO_%%% directories. I specify /SAVE with the product command just to be on the safe side. (I also break the system-disk shadow set (cleanly via a reboot) so I have a certainly bootable backup disk in case things get so hosed that I can't even remove patches. I have never had to use either, though.) I think this is wasting too much space. I believe it gets deleted when upgrading VMS, but I don't want to wait that long. I probably can't just delete stuff in those directories, or delete the entire contents of a directory. I generally add the backup member of the system-disk shadow set back in after a few days (a better long-term solution would be to bring it up to three members first before removing a backup member), but would like to have the ability to remove patches for a while longer, say weeks to months. Yesterday, I installed the latest update, the patches (9) which have appeared since then and aren't in the update (this is 7.3-2) and the latest MOTIF and TCPIP (5.4 ECO 6) patches (taking care to use a couple of (identical) shareable images from ECO 5 to circumvent a known problem). I can't imagine any situation where I would need to revert to a state older than these updates yesterday. So, is there a simple command to get rid of all the PCSI$UNDO stuff except that needed to revert to the state at a certain time? How stable is PCSI? Is it worth delving into the documentation and understanding how it works, or (since it might be changing under the hood), it is preferable to regard it as a black box? ------------------------------ Date: 29 May 2007 02:11:48 -0700 From: IanMiller Subject: Re: PCSI, disk space, UNDO, unseen dangers etc Message-ID: <1180429907.973154.94730@k79g2000hse.googlegroups.com> See $ HELP PRODUCT DELETE RECOVERY_DATA ------------------------------ Date: Tue, 29 May 2007 11:16:09 +0200 From: JOUKJ Subject: Re: PCSI, disk space, UNDO, unseen dangers etc Message-ID: <465BEF59.2010005@hrem.nano.tudelft.nl> Phillip Helbig---remove CLOTHES to reply wrote: > I have about 1½ million blocks (of about 8 million total) on the system > disk in PCSI$UNDO_%%% directories. I specify /SAVE with the product > command just to be on the safe side. (I also break the system-disk > shadow set (cleanly via a reboot) so I have a certainly bootable backup > disk in case things get so hosed that I can't even remove patches. I > have never had to use either, though.) I think this is wasting too much > space. I believe it gets deleted when upgrading VMS, but I don't want > to wait that long. > > I probably can't just delete stuff in those directories, or delete the > entire contents of a directory. $ PRODUCT DELETE RECOVERY_DATA [/qualifiers] > > I generally add the backup member of the system-disk shadow set back in > after a few days (a better long-term solution would be to bring it up to > three members first before removing a backup member), but would like to > have the ability to remove patches for a while longer, say weeks to > months. > > Yesterday, I installed the latest update, the patches (9) which have > appeared since then and aren't in the update (this is 7.3-2) and the > latest MOTIF and TCPIP (5.4 ECO 6) patches (taking care to use a couple > of (identical) shareable images from ECO 5 to circumvent a known > problem). I can't imagine any situation where I would need to revert to > a state older than these updates yesterday. So, is there a simple > command to get rid of all the PCSI$UNDO stuff except that needed to > revert to the state at a certain time? yes see above and look at the /BEFORE qualifier Jouk ------------------------------ Date: Tue, 29 May 2007 09:21:10 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: PCSI, disk space, UNDO, unseen dangers etc Message-ID: In article <1180429907.973154.94730@k79g2000hse.googlegroups.com>, IanMiller writes: > See > $ HELP PRODUCT DELETE RECOVERY_DATA That's it, thanks. What is the relation between patches and the PCSI$UNDO directories? It's not one-to-one since my patch action yesterday modified some older directories. (I don't know what it's doing under the hood.) In other words, does a "recovery-data set" refer to a patch or to a PCSI$UNDO directory? ------------------------------ Date: Tue, 29 May 2007 14:31:23 +0200 From: JOUKJ Subject: Re: PCSI, disk space, UNDO, unseen dangers etc Message-ID: <465C1D1B.6050607@hrem.nano.tudelft.nl> Phillip Helbig---remove CLOTHES to reply wrote: > In article <1180429907.973154.94730@k79g2000hse.googlegroups.com>, > IanMiller writes: > > >>See >>$ HELP PRODUCT DELETE RECOVERY_DATA > > > That's it, thanks. What is the relation between patches and the > PCSI$UNDO directories? It's not one-to-one since my patch action > yesterday modified some older directories. (I don't know what it's > doing under the hood.) In other words, does a "recovery-data set" refer > to a patch or to a PCSI$UNDO directory? > refers to a directory. You probably installed several patches by one PRODUCT command. If you want them per patch give the PRODUCT INSTALL/SAVE command for each patch seperately. Jouk ------------------------------ Date: Tue, 29 May 2007 14:31:42 +0200 From: JOUKJ Subject: Re: PCSI, disk space, UNDO, unseen dangers etc Message-ID: <465C1D2E.8050405@hrem.nano.tudelft.nl> Phillip Helbig---remove CLOTHES to reply wrote: > In article <1180429907.973154.94730@k79g2000hse.googlegroups.com>, > IanMiller writes: > > >>See >>$ HELP PRODUCT DELETE RECOVERY_DATA > > > That's it, thanks. What is the relation between patches and the > PCSI$UNDO directories? It's not one-to-one since my patch action > yesterday modified some older directories. (I don't know what it's > doing under the hood.) In other words, does a "recovery-data set" refer > to a patch or to a PCSI$UNDO directory? > refers to a directory. You probably installed several patches by one PRODUCT command. If you want them per patch give the PRODUCT INSTALL/SAVE command for each patch seperately. Jouk ------------------------------ Date: Tue, 29 May 2007 14:31:52 +0200 From: JOUKJ Subject: Re: PCSI, disk space, UNDO, unseen dangers etc Message-ID: <465C1D38.4070609@hrem.nano.tudelft.nl> Phillip Helbig---remove CLOTHES to reply wrote: > In article <1180429907.973154.94730@k79g2000hse.googlegroups.com>, > IanMiller writes: > > >>See >>$ HELP PRODUCT DELETE RECOVERY_DATA > > > That's it, thanks. What is the relation between patches and the > PCSI$UNDO directories? It's not one-to-one since my patch action > yesterday modified some older directories. (I don't know what it's > doing under the hood.) In other words, does a "recovery-data set" refer > to a patch or to a PCSI$UNDO directory? > refers to a directory. You probably installed several patches by one PRODUCT command. If you want them per patch give the PRODUCT INSTALL/SAVE command for each patch seperately. Jouk ------------------------------ Date: Tue, 29 May 2007 12:47:26 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: PCSI, disk space, UNDO, unseen dangers etc Message-ID: In article <465C1D1B.6050607@hrem.nano.tudelft.nl>, JOUKJ writes: > refers to a directory. You probably installed several patches by one > PRODUCT command. If you want them per patch give the PRODUCT > INSTALL/SAVE command for each patch seperately. No, I always install one patch per PRODUCT command. ------------------------------ Date: Mon, 28 May 2007 16:01:57 GMT From: "Jerry Alan Braga" Subject: php error from command line Message-ID: when I run php from command line I get this $php PHP Warning: Unknown(): Invalid library (maybe not a PHP library) 'php_mysql.exe' in Unknown on line 0 I am using the php_mysql.exe that was provided by jfp kit. If I rename that file and go back to original file installed with php kit the error goes away. Is this a problem or not ? ------------------------------ Date: Tue, 29 May 2007 15:41:34 +1000 From: "Gremlin" Subject: Problem with NFS client connecting to Windows 2003 server Message-ID: <135nf8i6dheh60a@corp.supernews.com> Hi All [SYSMGR] > tcpip sho ver HP TCP/IP Services for OpenVMS Alpha Version V5.5 - ECO 1 on an AlphaServer DS10L 466 MHz running OpenVMS V8.2 I have created an NFS share on a Windows2003 server running Windows Services for Unix v3.5. The share is OK, there are no error messages in the Windows log file, I have mounted the share using: TCPIP> mount dnsf0: /host="DL380" /path="NFSdata" %TCPIP$DNFSMOUNT-S-MOUNTED, NFSdata mounted on _DNFS7:[000000] then [SYSMGR] > sho dev d Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt DNFS0: Online 0 DNFS7: Mounted 0 DL380$NFSdat 418661312 1 1 VAX$DQA0: Offline 1 VAX$DQA1: Mounted 0 ALPHASYS 7059032 435 1 VAX$DQB0: Online 0 VAX$DQB1: Offline 1 VAX$DVA0: Online 0 [SYSMGR] > dir dnfs7:[000000] %DIRECT-E-OPENIN, error opening DNFS7:[000000]*.*;* as input -RMS-E-DNF, directory not found -SYSTEM-F-TIMEOUT, device timeout Can't see it, get to it, etc. Any thoughts appreciated - cheers ------------------------------ Date: Tue, 29 May 2007 12:00:05 +0200 From: "P. Sture" Subject: Re: Problem with NFS client connecting to Windows 2003 server Message-ID: In article <135nf8i6dheh60a@corp.supernews.com>, "Gremlin" wrote: > Hi All > > [SYSMGR] > tcpip sho ver > > HP TCP/IP Services for OpenVMS Alpha Version V5.5 - ECO 1 > on an AlphaServer DS10L 466 MHz running OpenVMS V8.2 > > I have created an NFS share on a Windows2003 server running Windows Services > for Unix v3.5. The share is OK, there are no error messages in the Windows > log file, I have mounted the share using: > > TCPIP> mount dnsf0: /host="DL380" /path="NFSdata" > %TCPIP$DNFSMOUNT-S-MOUNTED, NFSdata mounted on _DNFS7:[000000] > Differences from my mount command for an NFS share on an OS X system: $ tcpip mount dnfs2:[paul] /host="mac" - /path="/Volumes/VMS_Backup"/system 1) I specify a non-zero unit number for the DNFS device - DNFS2 above will be the name of the device I can subsequently access it with, where yours appears to end up with a random (round robin?) unit number. 2) I specify a directory other than [000000]. A test shows that [000000] also works, but it can be any (valid syntax) name you wish. 3) My command uses /SYSTEM > then > > [SYSMGR] > sho dev d > > Device Device Error Volume Free Trans > Mnt > Name Status Count Label Blocks Count > Cnt > DNFS0: Online 0 > DNFS7: Mounted 0 DL380$NFSdat 418661312 1 > 1 > VAX$DQA0: Offline 1 > VAX$DQA1: Mounted 0 ALPHASYS 7059032 435 > 1 > VAX$DQB0: Online 0 > VAX$DQB1: Offline 1 > VAX$DVA0: Online 0 > [SYSMGR] > dir dnfs7:[000000] > %DIRECT-E-OPENIN, error opening DNFS7:[000000]*.*;* as input > -RMS-E-DNF, directory not found > -SYSTEM-F-TIMEOUT, device timeout > > Can't see it, get to it, etc. > > Any thoughts appreciated - cheers -- Paul Sture ------------------------------ Date: Tue, 29 May 2007 21:38:30 +1000 From: "Gremlin" Subject: Re: Problem with NFS client connecting to Windows 2003 server Message-ID: <135o45oqaljd82c@corp.supernews.com> Hi Paul Tried as follows..... TCPIP> mount dnsf1:[webdata] /host="dl380" /path="NFSdata" /system %TCPIP$DNFSMOUNT-S-MOUNTED, NFSdata mounted on _DNFS1:[WEBDATA] TCPIP> Exit [SYSMGR] > sho dev d Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt DNFS0: Online 0 DNFS1: Mounted 0 dl380$NFSdat 415081224 1 1 VAX$DQA0: Offline 1 VAX$DQA1: Mounted 0 ALPHASYS 7051090 436 1 VAX$DQB0: Online 0 VAX$DQB1: Offline 1 VAX$DVA0: Online 0 [SYSMGR] > dir dnsf1:[webdata] %DIRECT-E-OPENIN, error opening DNSF1:[WEBDATA]*.*;* as input -RMS-F-DEV, error in device name or inappropriate device type for operation [SYSMGR] > dir dnsf1:[000000...] %DIRECT-E-OPENIN, error opening DNSF1:[000000...]*.*;* as input -RMS-F-DEV, error in device name or inappropriate device type for operation [SYSMGR] > dir dnsf1:[*...] %DIRECT-E-OPENIN, error opening DNSF1:[*...]*.*;* as input -RMS-F-DEV, error in device name or inappropriate device type for operation [SYSMGR] > Something I'm missing here, not sure what though.... "P. Sture" wrote in message news:paul.sture.nospam-232AC5.12000529052007@mac.sture.ch... > In article <135nf8i6dheh60a@corp.supernews.com>, > "Gremlin" wrote: > >> Hi All >> >> [SYSMGR] > tcpip sho ver >> >> HP TCP/IP Services for OpenVMS Alpha Version V5.5 - ECO 1 >> on an AlphaServer DS10L 466 MHz running OpenVMS V8.2 >> >> I have created an NFS share on a Windows2003 server running Windows >> Services >> for Unix v3.5. The share is OK, there are no error messages in the >> Windows >> log file, I have mounted the share using: >> >> TCPIP> mount dnsf0: /host="DL380" /path="NFSdata" >> %TCPIP$DNFSMOUNT-S-MOUNTED, NFSdata mounted on _DNFS7:[000000] >> > > Differences from my mount command for an NFS share on an OS X system: > > $ tcpip mount dnfs2:[paul] /host="mac" - > /path="/Volumes/VMS_Backup"/system > > 1) I specify a non-zero unit number for the DNFS device - DNFS2 above > will be the name of the device I can subsequently access it with, where > yours appears to end up with a random (round robin?) unit number. > > 2) I specify a directory other than [000000]. A test shows that [000000] > also works, but it can be any (valid syntax) name you wish. > > 3) My command uses /SYSTEM > >> then >> >> [SYSMGR] > sho dev d >> >> Device Device Error Volume Free >> Trans >> Mnt >> Name Status Count Label Blocks >> Count >> Cnt >> DNFS0: Online 0 >> DNFS7: Mounted 0 DL380$NFSdat 418661312 >> 1 >> 1 >> VAX$DQA0: Offline 1 >> VAX$DQA1: Mounted 0 ALPHASYS 7059032 >> 435 >> 1 >> VAX$DQB0: Online 0 >> VAX$DQB1: Offline 1 >> VAX$DVA0: Online 0 >> [SYSMGR] > dir dnfs7:[000000] >> %DIRECT-E-OPENIN, error opening DNFS7:[000000]*.*;* as input >> -RMS-E-DNF, directory not found >> -SYSTEM-F-TIMEOUT, device timeout >> >> Can't see it, get to it, etc. >> >> Any thoughts appreciated - cheers > > -- > Paul Sture ------------------------------ Date: Tue, 29 May 2007 15:26:35 +0200 From: "Gorazd Kikelj" Subject: Re: Problem with NFS client connecting to Windows 2003 server Message-ID: "Gremlin" wrote in message news:135nf8i6dheh60a@corp.supernews.com... > Hi All > > [SYSMGR] > tcpip sho ver > > HP TCP/IP Services for OpenVMS Alpha Version V5.5 - ECO 1 > on an AlphaServer DS10L 466 MHz running OpenVMS V8.2 > > I have created an NFS share on a Windows2003 server running Windows > Services for Unix v3.5. The share is OK, there are no error messages in > the Windows log file, I have mounted the share using: > > TCPIP> mount dnsf0: /host="DL380" /path="NFSdata" > %TCPIP$DNFSMOUNT-S-MOUNTED, NFSdata mounted on _DNFS7:[000000] > ... > [SYSMGR] > dir dnfs7:[000000] > %DIRECT-E-OPENIN, error opening DNFS7:[000000]*.*;* as input > -RMS-E-DNF, directory not found > -SYSTEM-F-TIMEOUT, device timeout > > Can't see it, get to it, etc. > > Any thoughts appreciated - cheers > This is usualy missing username mappings and proxy settings, hence security. So create a valid username mapping on SFU (Windows) side and then also create right proxy records to reflect win mappings. And you are done :-) So create a mapping in SFU for example: uid=100, gid=100 -> windows\Administrator and proxy in tcpip uid=100, gid=100, vmsuser that should do the trick. Best, Gorazd ------------------------------ Date: Tue, 29 May 2007 15:36:37 +0200 From: "Gorazd Kikelj" Subject: Re: Problem with NFS client connecting to Windows 2003 server Message-ID: Ok, I reread the question. This is an old SFU bug that Microsoft don't want to resolve. (not enough pressure from customers to do it. It was reported to MS several times). Use /NOADF in mount and your nfs share will work OK. Best, Gorazd "Gorazd Kikelj" wrote in message news:f3h9mc$g0u$1@usenet01.boi.hp.com... > > "Gremlin" wrote in message > news:135nf8i6dheh60a@corp.supernews.com... >> Hi All >> >> [SYSMGR] > tcpip sho ver >> >> HP TCP/IP Services for OpenVMS Alpha Version V5.5 - ECO 1 >> on an AlphaServer DS10L 466 MHz running OpenVMS V8.2 >> >> I have created an NFS share on a Windows2003 server running Windows >> Services for Unix v3.5. The share is OK, there are no error messages in >> the Windows log file, I have mounted the share using: >> >> TCPIP> mount dnsf0: /host="DL380" /path="NFSdata" >> %TCPIP$DNFSMOUNT-S-MOUNTED, NFSdata mounted on _DNFS7:[000000] >> > ... >> [SYSMGR] > dir dnfs7:[000000] >> %DIRECT-E-OPENIN, error opening DNFS7:[000000]*.*;* as input >> -RMS-E-DNF, directory not found >> -SYSTEM-F-TIMEOUT, device timeout >> >> Can't see it, get to it, etc. >> >> Any thoughts appreciated - cheers >> > > This is usualy missing username mappings and proxy settings, hence > security. > So create a valid username mapping on SFU (Windows) side and then also > create right proxy records to reflect win mappings. And you are done :-) > > > So create a mapping in SFU for example: > > uid=100, gid=100 -> windows\Administrator > > and proxy in tcpip > > uid=100, gid=100, vmsuser > > that should do the trick. > > Best, Gorazd > ------------------------------ Date: Tue, 29 May 2007 19:20:26 +0200 From: "P. Sture" Subject: Re: Problem with NFS client connecting to Windows 2003 server Message-ID: In article <135o45oqaljd82c@corp.supernews.com>, "Gremlin" wrote: > Hi Paul > > Tried as follows..... > > TCPIP> mount dnsf1:[webdata] /host="dl380" /path="NFSdata" /system > %TCPIP$DNFSMOUNT-S-MOUNTED, NFSdata mounted on _DNFS1:[WEBDATA] > TCPIP> Exit > [SYSMGR] > dir dnsf1:[webdata] > %DIRECT-E-OPENIN, error opening DNSF1:[WEBDATA]*.*;* as input > -RMS-F-DEV, error in device name or inappropriate device type for operation > [SYSMGR] > dir dnsf1:[000000...] > %DIRECT-E-OPENIN, error opening DNSF1:[000000...]*.*;* as input > -RMS-F-DEV, error in device name or inappropriate device type for operation > > Something I'm missing here, not sure what though.... OK. Gorazd's answer looks hopeful. > > Paul Sture -- Paul Sture ------------------------------ Date: Tue, 29 May 2007 12:10:24 -0500 From: David J Dachtera Subject: PURGE Enhancement Suggestion (was Re: AUTOGEN: %SYSTEM-W-HEADERFULL, file header Message-ID: <465C5E80.F0ECB26C@spam.comcast.net> Phillip Helbig---remove CLOTHES to reply wrote: > > In article <214ed$465b21c6$cef8887a$28611@TEKSAVVY.COM>, JF Mezei > writes: > > > Phillip Helbig---remove CLOTHES to reply wrote: > > > SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_RECV_RUN.LOG gets up to about ;4000 > > > every day (spam!). They get purged automatically, and every night I > > > renumber them, so in itself not a problem, but it is a lot of turnover. > > > > $SET DIRECTORY SYS$SPECIFIC:[TCPIP$SMTP]/VERSION=10 > > $PURGE SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_RECV_RUN.LOG > > $set file SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_RECV_RUN.LOG;*/version=10 > > This keeps the number of files low, but there is still the turnover. > And you still have to rename when you get to ;32767. It's probably been suggested before: $ PURGE/RESEQUENCE ...to "renumber" the existing versions, starting at ;1. I do this in DCL with a pair of wild-carded RENAMEs. Not as efficient as it could be. -- David J Dachtera dba DJE Systems http://www.djesys.com/ Unofficial OpenVMS Marketing Home Page http://www.djesys.com/vms/market/ Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/ Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/ Unofficial OpenVMS Hobbyist Support Page: http://www.djesys.com/vms/support/ ------------------------------ Date: Tue, 29 May 2007 16:22:48 +0400 From: "Ruslan R. Laishev" Subject: RDB turorial Message-ID: Hello, Hal! Hal Kuff wrote: > There are quite a few (three) GREAT books from the digital press... read > Them!!!! In fact Lilian Hobbs, the creator or one of the creators of rdb > wrote them. Thanks! I'll try to find it. > Also, how the hell are you these days? Still doing kerberos? Kerberos ... RADIUS ... teaching young VMS students at Ford Russia, and so on... > > > "Ruslan R. Laishev" > wrote in message news:3995DF2D429C1314ED2D60760705A2C1@NNTP.DeltaTel.RU... >> Hello, All! >> >> >> I looking for RDB tutorial "for beginners". >> >> >> Thanks for any pointers. >> >> >> -- >> + WBR, OpenVMS [Sys|Net] HardWorker ............. Skype: SysMan-One + >> Delta Telecom JSC, IMT-MC-450(CDMA2000) cellular operator >> Russia,191119,St.Petersburg,Transportny per. 3 Cel: +7 (812) 716-3222 >> +http://starlet.deltatelecom.ru ............. Frying on OpenVMS only + > > > -- + WBR, OpenVMS [Sys|Net] HardWorker ............. Skype: SysMan-One + Delta Telecom JSC, IMT-MC-450(CDMA2000) cellular operator Russia,191119,St.Petersburg,Transportny per. 3 Cel: +7 (812) 716-3222 +http://starlet.deltatelecom.ru ............. Frying on OpenVMS only + ------------------------------ Date: Tue, 29 May 2007 16:28:00 +0400 From: "Ruslan R. Laishev" Subject: RDB Tutorial Message-ID: <26E14AF24CF263A19D2D9E5F5342EF9F@NNTP.DeltaTel.RU> Hello, Dr. Dweeb! Dr. Dweeb wrote: > Ruslan R. Laishev wrote: >> Hello, All! >> >> >> I looking for RDB tutorial "for beginners". >> >> >> Thanks for any pointers. > > Do you mean OracleRdb as a specific product, or "rdb" as a generic reference > to relational databases. > > Read Chris Date's book for the latter and I guess the Hobbs book for the > former. > Amazon search is your friend here. I looking for any guide which describe for beginners several ways for developing/programming apps under/on RDB: - SQL Mods - SQL Preprocs - Stored Procedures, Packages (?) - Creating and using constants in the stored procedures/packages (as is it in the ORACLE PL/SQL stuff). > > Dweeb > > > -- + WBR, OpenVMS [Sys|Net] HardWorker ............. Skype: SysMan-One + Delta Telecom JSC, IMT-MC-450(CDMA2000) cellular operator Russia,191119,St.Petersburg,Transportny per. 3 Cel: +7 (812) 716-3222 +http://starlet.deltatelecom.ru ............. Frying on OpenVMS only + ------------------------------ Date: 28 May 2007 12:48:20 +0200 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: Recommended Graphic Adapter for Hobbyist Itanic? Message-ID: <465acf94$1@news.langstoeger.at> I've an offer for a Itanic zx6000 with an ATI Fire GL Z1 (AGP, 128MB, 2xDVI) and I'm wondering if I could use it for VMS (OpenVMS I64 V8.3 and up). AFAIK, the zx6000 is a rx2600 with another (AGP) packpanel and I could run VMS unsupported. But what about the graphic adapter? I don't think this AGP card would run with VMS DECwindows-MOTIF/CDE. Right? What graphic adapter would fit in a zx6000 and could be used with VMS MOTIF? Many TIA -EPLAN PS: I've also an offer for a rx2600 (with onboard graphic), but the price is more than double of the zx (ok, rx2600 has 2x1500MHz instead of 2x1300MHz and has a RAID 5302 - would that run with VMS? - and is rack mount, while the zx6000 is - hopefully office friendly aka slient - floor stand packaged) -- Peter "EPLAN" LANGSTOEGER Network and OpenVMS system specialist E-mail peter@langstoeger.at A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist ------------------------------ Date: Mon, 28 May 2007 07:19:05 -0400 From: "FredK" Subject: Re: Recommended Graphic Adapter for Hobbyist Itanic? Message-ID: You need an ATI Radeon 7500. VMS will work. Someone reported that an AGP R7500 worked (but I don't know how - since we never did AGP support). "Peter 'EPLAN' LANGSTOeGER" wrote in message news:465acf94$1@news.langstoeger.at... > I've an offer for a Itanic zx6000 with an ATI Fire GL Z1 (AGP, 128MB, > 2xDVI) > and I'm wondering if I could use it for VMS (OpenVMS I64 V8.3 and up). > > AFAIK, the zx6000 is a rx2600 with another (AGP) packpanel and I could > run VMS unsupported. But what about the graphic adapter? I don't think > this AGP card would run with VMS DECwindows-MOTIF/CDE. Right? > > What graphic adapter would fit in a zx6000 and could be used with VMS > MOTIF? > > Many TIA > > -EPLAN > > PS: I've also an offer for a rx2600 (with onboard graphic), but the price > is more than double of the zx (ok, rx2600 has 2x1500MHz instead of > 2x1300MHz > and has a RAID 5302 - would that run with VMS? - and is rack mount, while > the zx6000 is - hopefully office friendly aka slient - floor stand > packaged) > -- > Peter "EPLAN" LANGSTOEGER > Network and OpenVMS system specialist > E-mail peter@langstoeger.at > A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist ------------------------------ Date: 28 May 2007 15:16:48 +0200 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: Re: Recommended Graphic Adapter for Hobbyist Itanic? Message-ID: <465af260$1@news.langstoeger.at> In article , "FredK" writes: >You need an ATI Radeon 7500. Thanks (but do you mean the All-In-Wonder ATI Radeon 7500 or the Mobility ATI RADEON 7500 or what else ATI Radeon 7500?). Please be a little bit more verbose (eg. what computer bus is your ATI Radeon 7500? PCI or PCI-X or PCI-Express or what else and what speed and how much memory and so on)? As a non-AGP ATI 7500 is quite uncommon here (no entries in stores and not one at Ebay - at least so far for me) are there any alternatives? >VMS will work. Thanks for confirming (that OpenVMS I64 V8.3 runs on HP zx6000 workstation). > Someone reported that an AGP >R7500 worked (but I don't know how - since we never did AGP support). Why not try it? You seem the one that has the resources to at least trying such a configuration and prove/disprove such an statement ;-) TIA -- Peter "EPLAN" LANGSTOEGER Network and OpenVMS system specialist E-mail peter@langstoeger.at A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist ------------------------------ Date: Mon, 28 May 2007 16:08:43 +0200 From: Jur van der Burg <"lddriver at digiater dot nl"> Subject: Re: Recommended Graphic Adapter for Hobbyist Itanic? Message-ID: <465ae26c$0$322$e4fe514c@news.xs4all.nl> I've got a zx6000 and it works for sure with VMS (8.0 - 8.3). It's a floor stand model and makes little noise. I personally don't use graphics (the system is located in another room far from my workspace), but an X11 emulator on a pc works fine. Jur. Peter 'EPLAN' LANGSTOeGER wrote: > In article , "FredK" writes: >> You need an ATI Radeon 7500. > > Thanks (but do you mean the All-In-Wonder ATI Radeon 7500 or the Mobility > ATI RADEON 7500 or what else ATI Radeon 7500?). Please be a little bit > more verbose (eg. what computer bus is your ATI Radeon 7500? PCI or PCI-X > or PCI-Express or what else and what speed and how much memory and so on)? > > As a non-AGP ATI 7500 is quite uncommon here (no entries in stores and > not one at Ebay - at least so far for me) are there any alternatives? > >> VMS will work. > > Thanks for confirming (that OpenVMS I64 V8.3 runs on HP zx6000 workstation). > >> Someone reported that an AGP >> R7500 worked (but I don't know how - since we never did AGP support). > > Why not try it? You seem the one that has the resources to at least trying > such a configuration and prove/disprove such an statement ;-) > > TIA > ------------------------------ Date: Mon, 28 May 2007 20:51:35 -0400 From: "FredK" Subject: Re: Recommended Graphic Adapter for Hobbyist Itanic? Message-ID: "Peter 'EPLAN' LANGSTOeGER" wrote in message news:465af260$1@news.langstoeger.at... > In article , "FredK" > writes: >>You need an ATI Radeon 7500. > > Thanks (but do you mean the All-In-Wonder ATI Radeon 7500 Yes. Get one from Island. > or the Mobility > ATI RADEON 7500 or what else ATI Radeon 7500?). Please be a little bit > more verbose (eg. what computer bus is your ATI Radeon 7500? PCI or PCI-X > or PCI-Express or what else and what speed and how much memory and so on)? > The card is PCI, the bus is PCI-X. > As a non-AGP ATI 7500 is quite uncommon here (no entries in stores and > not one at Ebay - at least so far for me) are there any alternatives? > Island has them. I believe the HP part number is AB551A > > Why not try it? You seem the one that has the resources to at least trying > such a configuration and prove/disprove such an statement ;-) > I do not have a zx6000. Someone in COV reported it. Search google. ------------------------------ Date: Tue, 29 May 2007 13:29:19 GMT From: rdeininger@mindspringdot.com (Robert Deininger) Subject: Re: Recommended Graphic Adapter for Hobbyist Itanic? Message-ID: In article <465acf94$1@news.langstoeger.at>, peter@langstoeger.at wrote: >I've an offer for a Itanic zx6000 with an ATI Fire GL Z1 (AGP, 128MB, 2xDVI) >and I'm wondering if I could use it for VMS (OpenVMS I64 V8.3 and up). > >AFAIK, the zx6000 is a rx2600 with another (AGP) packpanel and I could >run VMS unsupported. But what about the graphic adapter? I don't think >this AGP card would run with VMS DECwindows-MOTIF/CDE. Right? > >What graphic adapter would fit in a zx6000 and could be used with VMS MOTIF? Does the system have the optional Management Processor? If not, you should be able to add one. The graphics adapter in the MP works with VMS. >PS: I've also an offer for a rx2600 (with onboard graphic), but the price >is more than double of the zx (ok, rx2600 has 2x1500MHz instead of 2x1300MHz >and has a RAID 5302 - would that run with VMS? - and is rack mount, while >the zx6000 is - hopefully office friendly aka slient - floor stand packaged) The SmartArray 5302 is supported on Alpha, but it wasn't qualified on Itanium because it was at or near end-of-life at the time. ------------------------------ Date: 29 May 2007 07:18:24 -0700 From: Peter Weaver Subject: RMS_TUNE_CHECK - ROOT: Primary key index root level is high Message-ID: <1180448304.797793.8280@q75g2000hsh.googlegroups.com> For the past few months I have been working with a client who had severe performance problems on a 4 node/14 CPU Alpha cluster. At the start I could put one of the machines into high Interrupt and MP Sync by doing a simple DIR/DATE on four files, each file took 6 to 8 seconds to display. We have performance at an acceptable level now but now I am trying to get the machines to show what they can really do. To help keep things under control I wrote up a little DCL to run Hein's RMS_TUNE_CHECK on a few key files and mail off a HTML formatted message showing the results so the programmers would know when to do converts. The customer asked me to help them interpret the various messages. But I am not sure what to do with "ROOT: Primary key index root level is high: 3 (goal=2)." Are there any settings in the FDL I can play with that will fix this? Peter Weaver www.weaverconsulting.ca CHARON-VAX CHARON-AXP DataStream Reflection PreciseMail HP Commercial Hardware ------------------------------ Date: Mon, 28 May 2007 02:45:00 -0400 From: JF Mezei Subject: Suggestion: SET PROMPT/POST="string" Message-ID: <1dc4e$465a7a8d$cef8887a$9447@TEKSAVVY.COM> Suggestion for SET PROMPT command: Add the capability to display a string AFTER the user has terminated the command (pressed return etc). Goal: To enable the SET PROMPT to set a colour, display the prompt, allow user to enter the command in that colour, and after pressing return, another string is sent to the terminal to reset the colour so that the output would be in the normal colour. (commands would therefore stand out from the output). (change "colour" with "display attributes" or anything else. example: SET PROMPT="[43m$ "/POST="[0m" You could also have a SET PROMPT="Enter Command: "/POST="WORKING" (so the user would know that a command that takes a while to execute without any output is being processed). One could also insert scrolling region escape sequences so that commands are entered in one region of the VT terminal, and the command output would be switched to another scrolling region. The list of possibilities is endless :-) ------------------------------ Date: 28 May 2007 03:36:24 -0700 From: Neil Rieck Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: <1180348584.087758.8840@q66g2000hsg.googlegroups.com> Here's another web site's take on the IBM announcement: http://www.dailytech.com/IBM+Unleashes+47+GHz+POWER6+Microprocessor/article7384.htm ### IBM claims to be launching the world's fastest chip for UNIX servers IBM just launched the dual-core 64-bit POWER6 processor running at 4.7 GHz, which doubles the speed of the previous generation POWER5 while using nearly the same amount of electricity to run and cool it. The POWER6 processor is a result of a five-year R&D period, is composed of 790 million transistors and is built using IBM's 65nm process technology. IBM scientists targeted the way instructions are executed inside the chip to improve performance. For example, in the POWER6, the number of pipeline stages - the chunks of operations that must be completed in a single cycle of clock time - are kept static, but each stage is made faster, removing unnecessary work and doing more in parallel. As a result, execution time is reduced. Earlier this year, IBM hinted that its new POWER6 architecture may hit frequencies higher than 5 GHz. The POWER6 chip has a total cache size of 8 MB per chip - four times the POWER5 chip - to keep pace with the processor bandwidth. With 300 GB/s on tap, IBM boasts that its processor has so much bandwidth that the POWER6 chip could download the entire iTunes catalog in about 60 seconds. IBM believes that it has designed the POWER6 chip with a balanced amount of bandwidth and processing power. "Like the victory of IBM's Deep Blue chess-playing supercomputer 10 years ago this month, the debut of POWER6 processor-based systems proves that relentless innovation brings 'impossible' goals within reach," said Bill Zeitler, senior vice president, IBM Systems and Technology Group. "The POWER6 processor forges blazing performance and energy conservation technologies into a single piece of silicon, driving unprecedented business value for our customers." To facilitate the lower energy demands of the new chip, the POWER6 designers separated circuits that can't support low voltage operation onto their own power supply "rails," allowing IBM to dramatically reduce power for the rest of the chip. IBM engineers also used a new method of chip design that enables POWER6 to operate at low voltages, allowing the same chip to be used in low power blade environments as well as large, high-performance symmetric multiprocessing machines. In another design to reduce energy consumption and heat production, processor clocks can be dynamically turned off when there is no useful work to be done and turned back on when there are instructions to be executed. Also, the chip has configurable bandwidth, enabling customers to choose maximum performance or minimal cost. Parts of the memory not being utilized are dynamically turned off and then turned back on when needed. In cases where an over-temperature condition is detected, the POWER6 chip can reduce the rate of instruction execution to remain within an acceptable, user-defined temperature envelope. The chip is fast too, as a server built by IBM using the POWER6 architecture is the first ever to hold all four major benchmark speed records for business and technical performance. IBM says that its new 2- to 16-core server is multiple times faster than the HP Superdome or Itanium machines. The POWER6 chip is also aimed as being a midrange consolidation machine, containing special hardware and software that allows it to create many virtual servers on a single box. IBM calculates that 30 SunFire v890s can be consolidated into a single rack of the new IBM machine, saving more than $100,000 per year on energy costs. IBM plans to introduce the POWER6 chip throughout the System p and System i server lines. The POWER6 chip in the new IBM System p 570 server is the first UNIX microprocessor able to calculate decimal floating point arithmetic in hardware. Until now, calculations involving decimal numbers with floating decimal points were done using software. The built-in decimal floating point capability gives tremendous advantage to enterprises running complex tax, financial and ERP programs. Neil Rieck Kitchener/Waterloo/Cambridge, Ontario, Canada. http://www3.sympatico.ca/n.rieck/ ------------------------------ Date: Mon, 28 May 2007 09:15:13 -0500 From: Ron Johnson Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: On 05/28/07 05:36, Neil Rieck wrote: [snip] > The POWER6 processor is a result of a five-year R&D period, is > composed of 790 million transistors That's just a stunning number. Even at 65nm, it must be a *huge* chip. [snip] > The POWER6 chip has a total cache size of 8 MB per chip - four times I've always faulted Intel for "cheating" by adding a huge cache, thus hiding inherent weaknesses in your design, but which are exposed in highly multi-processing situation where you get lots of cache faulting. This seems like the same situation. But I guess that if you throw enough cache at a problem, it goes away. [snip] > > "Like the victory of IBM's Deep Blue chess-playing supercomputer 10 > years ago this month, Sure, if you change the program between each game! [snip] > > To facilitate the lower energy demands of the new chip, the POWER6 > designers separated circuits that can't support low voltage operation > onto their own power supply "rails," allowing IBM to dramatically > reduce power for the rest of the chip. IBM engineers also used a new Now that's a clever idea. [snip] > > The chip is fast too, as a server built by IBM using the POWER6 > architecture is the first ever to hold all four major benchmark speed > records for business and technical performance. IBM says that its new > 2- to 16-core server is multiple times faster than the HP Superdome or > Itanium machines. Hmmm. [snip] > software. The built-in decimal floating point capability gives > tremendous advantage to enterprises running complex tax, financial and > ERP programs. Floating point to compute financials???????? -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Mon, 28 May 2007 08:35:59 -0700 From: "Tom Linden" Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: On Mon, 28 May 2007 07:15:13 -0700, Ron Johnson wrote: > On 05/28/07 05:36, Neil Rieck wrote: > [snip] >> The POWER6 processor is a result of a five-year R&D period, is >> composed of 790 million transistors > > That's just a stunning number. Even at 65nm, it must be a *huge* chip. > > [snip] >> The POWER6 chip has a total cache size of 8 MB per chip - four times > > I've always faulted Intel for "cheating" by adding a huge cache, thus > hiding inherent weaknesses in your design, but which are exposed in > highly multi-processing situation where you get lots of cache faulting. > This seems like the same situation. > > But I guess that if you throw enough cache at a problem, it goes away. > > [snip] >> "Like the victory of IBM's Deep Blue chess-playing supercomputer 10 >> years ago this month, > > Sure, if you change the program between each game! > > [snip] >> To facilitate the lower energy demands of the new chip, the POWER6 >> designers separated circuits that can't support low voltage operation >> onto their own power supply "rails," allowing IBM to dramatically >> reduce power for the rest of the chip. IBM engineers also used a new > > Now that's a clever idea. > > [snip] >> The chip is fast too, as a server built by IBM using the POWER6 >> architecture is the first ever to hold all four major benchmark speed >> records for business and technical performance. IBM says that its new >> 2- to 16-core server is multiple times faster than the HP Superdome or >> Itanium machines. > > Hmmm. > > [snip] >> software. The built-in decimal floating point capability gives >> tremendous advantage to enterprises running complex tax, financial and >> ERP programs. > > Floating point to compute financials???????? Decimal, not binary. > -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Mon, 28 May 2007 11:49:44 -0500 From: Ron Johnson Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: On 05/28/07 10:35, Tom Linden wrote: [snip] >> >> [snip] >>> software. The built-in decimal floating point capability gives >>> tremendous advantage to enterprises running complex tax, financial and >>> ERP programs. >> >> Floating point to compute financials???????? > > Decimal, not binary. Hmm, ok. I guess the question is: what is "decimal *floating* point"? Is that a non-techie mangling of BCD/packed-decimal? -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Mon, 28 May 2007 10:02:19 -0700 From: "Tom Linden" Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: On Mon, 28 May 2007 09:49:44 -0700, Ron Johnson wrote: > On 05/28/07 10:35, Tom Linden wrote: > [snip] >>> >>> [snip] >>>> software. The built-in decimal floating point capability gives >>>> tremendous advantage to enterprises running complex tax, financial and >>>> ERP programs. >>> >>> Floating point to compute financials???????? >> Decimal, not binary. > > Hmm, ok. I guess the question is: what is "decimal *floating* point"? > Is that a non-techie mangling of BCD/packed-decimal? > Not sure what you mean by non-techie mangling, but it is essentially the same as scaled fixed decimal (which is stored in BCD, unlike picture which is stored as zoned decimal) with a two digit signed exponent -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Mon, 28 May 2007 12:31:22 -0500 From: Ron Johnson Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: On 05/28/07 12:02, Tom Linden wrote: > On Mon, 28 May 2007 09:49:44 -0700, Ron Johnson > wrote: > >> On 05/28/07 10:35, Tom Linden wrote: >> [snip] >>>> >>>> [snip] >>>>> software. The built-in decimal floating point capability gives >>>>> tremendous advantage to enterprises running complex tax, financial and >>>>> ERP programs. >>>> >>>> Floating point to compute financials???????? >>> Decimal, not binary. >> >> Hmm, ok. I guess the question is: what is "decimal *floating* >> point"? Is that a non-techie mangling of BCD/packed-decimal? >> > Not sure what you mean by non-techie mangling, but it is essentially the > same > as scaled fixed decimal (which is stored in BCD, unlike picture which is > stored > as zoned decimal) with a two digit signed exponent OK. Well then, heck, that's sweet, and a perfect example of a company using intelligence and forethought. I mean, when was the last time that a non-mainframe optimized BCD instructions? (Yes, I realize that POWER6 will be used in z/ systems, but presumably AICHES compilers will also get the benefits.) -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Mon, 28 May 2007 18:08:24 GMT From: Rob Brown Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: On Mon, 28 May 2007, Tom Linden wrote: > On Mon, 28 May 2007 07:15:13 -0700, Ron Johnson > wrote: > >> On 05/28/07 05:36, Neil Rieck wrote: >> [snip] >>> software. The built-in decimal floating point capability gives >>> tremendous advantage to enterprises running complex tax, financial >>> and ERP programs. >> >> Floating point to compute financials???????? > > Decimal, not binary. I can't see the use of it. When would the financial guys like to use floating point? When do they ever accept x+y=x for y<>0? -- Rob Brown b r o w n a t g m c l d o t c o m G. Michaels Consulting Ltd. (780)438-9343 (voice) Edmonton (780)437-3367 (FAX) http://gmcl.com/ ------------------------------ Date: 28 May 2007 12:54:39 -0700 From: Andrew Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: <1180382079.153328.32190@u30g2000hsc.googlegroups.com> On 28 May, 15:15, Ron Johnson wrote: > On 05/28/07 05:36, Neil Rieck wrote: > [snip] > > > The POWER6 processor is a result of a five-year R&D period, is > > composed of 790 million transistors > > That's just a stunning number. Even at 65nm, it must be a *huge* chip. It is big but nothing like the size of Montecito which comes in at 1.72 billion transistors, of course most of this is cache as it is in the Power 6. > > [snip] > > > The POWER6 chip has a total cache size of 8 MB per chip - four times > > I've always faulted Intel for "cheating" by adding a huge cache, > thus hiding inherent weaknesses in your design, but which are > exposed in highly multi-processing situation where you get lots of > cache faulting. This seems like the same situation. > > But I guess that if you throw enough cache at a problem, it goes away. No you just make the problem tolerable. Increasing cache sizes is an inevitable consequence of the increasing gap between CPU cycle time and main memory access time. There are a number of other strategies that try to reduce the impact having to stall while waiting for a block load and increasing cache sizes is one of them. BTW Power 6 doesn't have that large a Cache, Montecito for example goes up to 24 MB to 8 MB is relatively modest. regards Andrew Harrison ------------------------------ Date: Mon, 28 May 2007 16:48:48 -0500 From: Ron Johnson Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: <47I6i.88778$vE1.41903@newsfe24.lga> On 05/28/07 14:54, Andrew wrote: > On 28 May, 15:15, Ron Johnson wrote: >> On 05/28/07 05:36, Neil Rieck wrote: >> [snip] >> >>> The POWER6 processor is a result of a five-year R&D period, is >>> composed of 790 million transistors >> That's just a stunning number. Even at 65nm, it must be a *huge* chip. > > It is big but nothing like the size of Montecito which comes in at > 1.72 billion transistors, of course most of this is cache as it is in > the Power 6. > >> [snip] [snip] > > BTW Power 6 doesn't have that large a Cache, Montecito for example > goes up to 24 MB to 8 MB is relatively modest. Ok, I'm officially humbled. -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Mon, 28 May 2007 18:20:11 -0400 From: Bill Todd Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: Andrew wrote: > On 28 May, 15:15, Ron Johnson wrote: ... >> I've always faulted Intel for "cheating" by adding a huge cache, >> thus hiding inherent weaknesses in your design, but which are >> exposed in highly multi-processing situation where you get lots of >> cache faulting. This seems like the same situation. >> >> But I guess that if you throw enough cache at a problem, it goes away. > > No you just make the problem tolerable. Increasing cache sizes is an > inevitable consequence of the increasing gap between CPU cycle time > and main memory access time. It's also a consequence of the inability to cool very many cores running at high speed: if you're at your core-count single-chip cooling limit but could economically produce a larger chip (because chip area is only one cost component - other costs, e.g., packaging - aren't proportional to chip area), throwing in more cache can be almost free (especially since more cache doesn't increase defective chip percentages when you set aside a few extra blocks to use to repair defects). Given that a POWER6 chip has fewer than half as many transistors as Montecito (and apparently splits their use at least somewhat similarly between logic and cache), and is one process generation newer, one might expect it to be roughly 1/4 the size of a Montecito chip (which IIRC would make it around 150 sq. mm., or perhaps a bit more if Intel has noticeably denser cache cells). Since commodity PC processors were around this size not long ago at all, it may well be in the size range where any further size reduction (by decreasing the amount of cache) would not produce noticeable cost savings. - bill ------------------------------ Date: Mon, 28 May 2007 19:06:09 -0700 From: "Tom Linden" Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: On Mon, 28 May 2007 10:31:22 -0700, Ron Johnson wrote: > On 05/28/07 12:02, Tom Linden wrote: >> On Mon, 28 May 2007 09:49:44 -0700, Ron Johnson >> wrote: >> >>> On 05/28/07 10:35, Tom Linden wrote: >>> [snip] >>>>> >>>>> [snip] >>>>>> software. The built-in decimal floating point capability gives >>>>>> tremendous advantage to enterprises running complex tax, financial >>>>>> and >>>>>> ERP programs. >>>>> >>>>> Floating point to compute financials???????? >>>> Decimal, not binary. >>> >>> Hmm, ok. I guess the question is: what is "decimal *floating* >>> point"? Is that a non-techie mangling of BCD/packed-decimal? >>> >> Not sure what you mean by non-techie mangling, but it is essentially >> the same >> as scaled fixed decimal (which is stored in BCD, unlike picture which >> is stored >> as zoned decimal) with a two digit signed exponent > > OK. Well then, heck, that's sweet, and a perfect example of a company > using intelligence and forethought. I mean, when was the last time that > a non-mainframe optimized BCD instructions? VAX had bcd support. Alpha was an example of design based on lack of market understanding. > > (Yes, I realize that POWER6 will be used in z/ systems, but presumably > AICHES compilers will also get the benefits.) > -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Tue, 29 May 2007 02:13:42 -0500 From: Ron Johnson Subject: Re: World's Fastest Processor? For Now, Anyway Message-ID: On 05/28/07 21:06, Tom Linden wrote: > On Mon, 28 May 2007 10:31:22 -0700, Ron Johnson > wrote: > >> On 05/28/07 12:02, Tom Linden wrote: >>> On Mon, 28 May 2007 09:49:44 -0700, Ron Johnson >>> wrote: >>> >>>> On 05/28/07 10:35, Tom Linden wrote: >>>> [snip] >>>>>> >>>>>> [snip] >>>>>>> software. The built-in decimal floating point capability gives >>>>>>> tremendous advantage to enterprises running complex tax, >>>>>>> financial and >>>>>>> ERP programs. >>>>>> >>>>>> Floating point to compute financials???????? >>>>> Decimal, not binary. >>>> >>>> Hmm, ok. I guess the question is: what is "decimal *floating* >>>> point"? Is that a non-techie mangling of BCD/packed-decimal? >>>> >>> Not sure what you mean by non-techie mangling, but it is essentially >>> the same >>> as scaled fixed decimal (which is stored in BCD, unlike picture which >>> is stored >>> as zoned decimal) with a two digit signed exponent >> >> OK. Well then, heck, that's sweet, and a perfect example of a company >> using intelligence and forethought. I mean, when was the last time >> that a non-mainframe optimized BCD instructions? > > VAX had bcd support. As did the 808[68] in microcode. Presumably those opcodes haven't yet been deprecated. But typical compilers don't use them. > Alpha was an example of design based on lack of > market understanding. It was a "Unix" market understanding. -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Tue, 29 May 2007 09:05:52 -0400 From: Chip Coldwell Subject: Re: Xpdf 3.02 for OpenVMS/Itanium Message-ID: On Thu, 24 May 2007, Peter 'EPLAN' LANGSTOeGER wrote: > In article , Chip Coldwell writes: > >OK, thanks for that information. I have updated the package and its > >dependencies. For the record, the dependencies are only > > > > software dec i64vms dwmotif version minimum v1.2 ; > > software freeware i64vms ghostscript version minimum v6.50 ; > > Why is XPDF dependant on GHOSTSCRIPT? It gets the required 13 type 1 fonts from ghostscript. Chip -- Charles M. "Chip" Coldwell "Turn on, log in, tune out" ------------------------------ End of INFO-VAX 2007.292 ************************