INFO-VAX Tue, 30 Jan 2007 Volume 2007 : Issue 60 Contents: Re: An opening for VMS: licence management Re: An opening for VMS: licence management Re: An opening for VMS: licence management Re: An opening for VMS: licence management Re: An opening for VMS: licence management Re: An opening for VMS: licence management Re: An opening for VMS: licence management Bill Hancock died on Jan 1, 2007 cannot log in at console Re: cannot log in at console Cannot ping ADSL from external Re: Cannot ping ADSL from external Re: Cannot ping ADSL from external Re: Cannot ping ADSL from external Re: Cannot ping ADSL from external Re: cannot restore SYSUAF Re: cannot restore SYSUAF Re: cannot restore SYSUAF Re: cannot restore SYSUAF Re: Delayed telnet connection Disk space problem ECP and TDC on Integrity, VMS v8.2-1 GNV for Alpha 8.2 Help needed with MONITOR /SUMMARY Monitor rms issue on Vax Re: PL/I for Itanium Re: PL/I for Itanium Re: Power down to remove SCSI tape drive? Re: Power down to remove SCSI tape drive? Re: Power down to remove SCSI tape drive? Re: Power down to remove SCSI tape drive? Re: Power down to remove SCSI tape drive? Re: Power down to remove SCSI tape drive? Re: Power down to remove SCSI tape drive? Re: PPPD, TCPIP Services and incomplete negotiations PRODUCT INSTALL question Re: Purveyor CGI mailbox capacity [bit long winded] Re: VMS in the HP hierarchy ---------------------------------------------------------------------- Date: 30 Jan 2007 02:10:59 -0800 From: etmsreec@yahoo.co.uk Subject: Re: An opening for VMS: licence management Message-ID: <1170151859.310891.4350@k78g2000cwa.googlegroups.com> You'd be surprised JF. Just because someone has a VMS license or layered product license loaded doesn't mean that they're *meant* to have it loaded. I've heard stories in the past of people having backdoor licenses from Digital and loading licenses that they've already traded in. On 30 Jan, 05:10, JF Mezei wrote: > News that Microsoft , in the UK, will be demanding that certain customers > submit to a Microsoft software audit of all their systems to find illegal > software and force customer to pay for those licences. If they do not agree > to the audit, MS will take them to british court. > > If HP were not a subsidiary of Microsoft, it would take a hige advantage of > this and market VMS as a reliable operating system with a built in licence > management system that doesn't require audits, and with a vendor that is > not out to get people. > > But sine marketing of VMS is not permitted by HP, it is likely that Linux > will be the one reapin the benefits of this since Linux customers do not > have to fear software audits. ------------------------------ Date: Tue, 30 Jan 2007 05:46:21 -0500 From: JF Mezei Subject: Re: An opening for VMS: licence management Message-ID: <8c159$45bf2201$cef8887a$13298@TEKSAVVY.COM> etmsreec@yahoo.co.uk wrote: > You'd be surprised JF. Just because someone has a VMS license or > layered product license loaded doesn't mean that they're *meant* to > have it loaded. I've heard stories in the past of people having > backdoor licenses from Digital and loading licenses that they've > already traded in. Definitely. But the point is that Microsoft is now embarking on a witch hunt wantijng to audit business customer sites and threathen to charge for any illegal license. VMS could be marketed as "we have a half decent licensing scheme and don't need to audit our customers". ------------------------------ Date: 30 Jan 2007 07:32:09 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: An opening for VMS: licence management Message-ID: <01U6iaCD7vQY@eisner.encompasserve.org> In article <45bed366$0$8947$c3e8da3@news.astraweb.com>, JF Mezei writes: > News that Microsoft , in the UK, will be demanding that certain customers > submit to a Microsoft software audit of all their systems to find illegal > software and force customer to pay for those licences. If they do not agree > to the audit, MS will take them to british court. Ah, yes Micrsoft, learning from SCO are they? "Buy our software or we'll take you to court." First Microsoft gives SCO money, now they emulate thier sales strategy, could be the begining of a beautifull new world. ------------------------------ Date: 30 Jan 2007 07:08:45 -0800 From: davidc@montagar.com Subject: Re: An opening for VMS: licence management Message-ID: <1170169725.823196.228250@s48g2000cws.googlegroups.com> On Jan 30, 4:59 am, JF Mezei wrote: > Oh, and if VMS ran on industrat standard machines, HP could tells its UK > Windows customers to migrate to VMS before Microsoft comes knocking at > their door :-) :-) :-) ;-) Actually, this story made it on Slashdot. For machines not running MS (i.e. Linux), they could force MS to pay for the audit since it's not governed by the EULA. This may end up increasing Linux adoption (see the story about Ernie Ball at http://news.com.com/ 2008-1082_3-5065859.html?tag=lh ) since MS will irritate their customers. This does provide an OpenVMS opportunity (as well as Linux and Mac, too), if you have the software on OpenVMS to fill the need. For now, the best opportunity is in the server area: Web services, Mail, Samba file servers. The market is Microsoft's to lose, and they appearently feel the need to squeeze the customer base in order to continue their "growth" in a monopolistic (convicted on 2 continents, isn't it?) market. Now, it's time for the non-Microsoft side to start their FUD engines for a change... ------------------------------ Date: Tue, 30 Jan 2007 15:39:12 GMT From: Tad Winters Subject: Re: An opening for VMS: licence management Message-ID: JF Mezei wrote in news:33e05$45bf2524$cef8887a$13669@TEKSAVVY.COM: > Oh, and if VMS ran on industrat standard machines, HP could tells its > UK Windows customers to migrate to VMS before Microsoft comes knocking > at their door :-) :-) :-) ;-) "industry standard"? Does that just mean the most common CPU? :-v) ------------------------------ Date: Tue, 30 Jan 2007 17:08:48 +0100 From: Paul Sture Subject: Re: An opening for VMS: licence management Message-ID: In article <01U6iaCD7vQY@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote: > In article <45bed366$0$8947$c3e8da3@news.astraweb.com>, JF Mezei > writes: > > > News that Microsoft , in the UK, will be demanding that certain customers > > submit to a Microsoft software audit of all their systems to find illegal > > software and force customer to pay for those licences. If they do not agree > > to the audit, MS will take them to british court. > > Ah, yes Micrsoft, learning from SCO are they? "Buy our software or > we'll take you to court." > > First Microsoft gives SCO money, now they emulate thier sales > strategy, could be the begining of a beautifull new world. This is nothing new. http://en.wikipedia.org/wiki/Business_Software_Alliance -- Paul Sture ------------------------------ Date: 30 Jan 2007 10:12:47 -0800 From: davidc@montagar.com Subject: Re: An opening for VMS: licence management Message-ID: <1170180767.854178.293120@p10g2000cwp.googlegroups.com> On Jan 30, 7:32 am, koeh...@eisner.nospam.encompasserve.org (Bob Koehler) wrote: > First Microsoft gives SCO money, now they emulate thier sales > strategy, could be the begining of a beautifull new world. I think it's already started. People are already talking about "Vista fatigue" since they are tired of talking about a product that's been delayed so long. IE's market share has dropped to under 80%. And I just heard on the radio about a local CompUSA that had a dozen or so folks lined up early for the new Vista release. A dozen or so? That's all? Absolutely nothing like the Windows95 release, which was remeniscient of crazed Mom's fighting over a new shipment of Cabbage Patch dolls. ------------------------------ Date: 30 Jan 2007 10:35:54 -0800 From: twnews@kittles.com Subject: Bill Hancock died on Jan 1, 2007 Message-ID: <1170182151.211276.129940@a75g2000cwd.googlegroups.com> I just did some VMS SIG list catching up and read that Bill Hancock died on Jan 1, 2007. I am sure many of you saw this from the SIG list. I am surprised that no one posted it here. Bill was an icon at DECUS Symposia. He would fill the largest ball rooms talking about networking. I never got to know Bill on a personal level. He was just this industry giant that seemed to know everything about networks (especially in a VMS environment) and about entertaining a room full of techies. Every year at HP World or HPTF or what ever the name of the year is I will hear some one say, "What ever happen to Bill Hancock?" This year I will have a new answer. He died. He will be missed. Thomas Wirt ------------------------------ Date: 30 Jan 2007 08:29:01 -0500 From: colonel@monmouth.com (Hoary Hairy Hoax) Subject: cannot log in at console Message-ID: We have a Digital Alpha running VMS 7.3-2. Its only user is SYSTEM. I can log in as SYSTEM from another host with ssh. I cannot log in at the console. The window interface accepts my user name and password, brings up a Motif hourglass, and reverts to the login screen. Any suggestions? -:- J Darkened chamber, dismal howls-- Is it tigers, wolves, or owls? Now I spy him through the gloom: Dr. JACKAL, I presume! --The Roguelet's ABC -- Col. G. L. Sicherman home: colonel@mail.monmouth.com work: George.Sicherman@Netrics.com web: ------------------------------ Date: 30 Jan 2007 12:01:29 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: cannot log in at console Message-ID: In article , colonel@monmouth.com (Hoary Hairy Hoax) writes: > We have a Digital Alpha running VMS 7.3-2. Its only > user is SYSTEM. I can log in as SYSTEM from another > host with ssh. I cannot log in at the console. The > window interface accepts my user name and password, > brings up a Motif hourglass, and reverts to the login > screen. Any suggestions? Sounds like the CDE desktop is configure not to allow SYSTEM to log in (may be called root in the configuration files). You can find and change this configuration, change the window manager from CDE to traditioanl DECwindows, or add another user, all over the SSH connection. Or you can use a minimum startup and make the changes from the console in text mode. ------------------------------ Date: 30 Jan 2007 03:58:18 -0800 From: mikeuk@ntlworld.com Subject: Cannot ping ADSL from external Message-ID: <1170158298.510033.226940@k78g2000cwa.googlegroups.com> I have been battling with this problem for a few days now and would really appreciate any help you could provide. I have setup my ADSL as follows. The provider is Plusnet. They have given me a static IP address of 212.159.66.255. My phone line has a microfilter plugged into it then a 30m cable going to my Netgear DG834 router. This then connects via one of it's LAN ports to my Windows 2003 server. I know the web server is working on windows as when it was connected previously to an NTl broadband connection it worked fine. when I type my static IP address into a browser on an external PC, I expect the default website to come up - but it says page cannot be displayed. When I do a trace route, it gets to the ISP then times out. Even when I enable allow pinging on the WAN port it still times out. Another thing, when I enable remote configuring of the router, it says from an external PC, ti will be port 8080 - so I type in http:// 212.159.66.255:8080 as it instructs and nothing comes up. The ISP think its something wrong with the router. Please help - this is driving me mad now. ------------------------------ Date: Tue, 30 Jan 2007 13:39:08 +0100 From: Paul Sture Subject: Re: Cannot ping ADSL from external Message-ID: In article <1170158298.510033.226940@k78g2000cwa.googlegroups.com>, mikeuk@ntlworld.com wrote: > I have been battling with this problem for a few days now and would > really appreciate any help you could provide. > > I have setup my ADSL as follows. The provider is Plusnet. They have > given me a static IP address of 212.159.66.255. My phone line has a > microfilter plugged into it then a 30m cable going to my Netgear DG834 > router. This then connects via one of it's LAN ports to my Windows > 2003 server. I know the web server is working on windows as when it > was connected previously to an NTl broadband connection it worked > fine. Can you connect to the webserver from you Windows system by entering localhost or 127.0.0.1 into a browser there? A few minutes ago I was getting "server not responding" messages, but now I am seeing some response, although you appear to have a blank index.html. -- Paul Sture ------------------------------ Date: Tue, 30 Jan 2007 13:52:05 +0000 (UTC) From: moroney@world.std.spaamtrap.com (Michael Moroney) Subject: Re: Cannot ping ADSL from external Message-ID: mikeuk@ntlworld.com writes: >I have been battling with this problem for a few days now and would >really appreciate any help you could provide. >I have setup my ADSL as follows. The provider is Plusnet. They have >given me a static IP address of 212.159.66.255. My phone line has a >microfilter plugged into it then a 30m cable going to my Netgear DG834 >router. This then connects via one of it's LAN ports to my Windows >2003 server. I know the web server is working on windows as when it >was connected previously to an NTl broadband connection it worked >fine. I don't know specifically about the Netgear router, but here's two points: The router may have to be told what to do with an incoming ping. It can ignore it, respond to it itself or pass it on to one of the IPs beind it. It may be set to ignore by default. (some security reason) It may route pings to what they call the "DMZ" address. They often say don't use it since it doesn't protect from external attacks. But VMS is secure enough so it's not a problem setting a VMS system as the DMZ. However, expect to see thousands of breakin attempts against SSH or FTP, if you don't use them, route those ports to an unused internal address. The router may or may not know to route an internal connection to its external address as if it is an external request. You may have to access any internal webpages or whatever by the 192.168.0.x address. ------------------------------ Date: Tue, 30 Jan 2007 10:58:38 -0500 From: "Richard B. gilbert" Subject: Re: Cannot ping ADSL from external Message-ID: <45BF6B2E.40101@comcast.net> mikeuk@ntlworld.com wrote: > I have been battling with this problem for a few days now and would > really appreciate any help you could provide. > > I have setup my ADSL as follows. The provider is Plusnet. They have > given me a static IP address of 212.159.66.255. My phone line has a > microfilter plugged into it then a 30m cable going to my Netgear DG834 > router. This then connects via one of it's LAN ports to my Windows > 2003 server. I know the web server is working on windows as when it > was connected previously to an NTl broadband connection it worked > fine. > > when I type my static IP address into a browser on an external PC, I > expect the default website to come up - but it says page cannot be > displayed. When I do a trace route, it gets to the ISP then times out. > Even when I enable allow pinging on the WAN port it still times out. > Another thing, when I enable remote configuring of the router, it says > from an external PC, ti will be port 8080 - so I type in http:// > 212.159.66.255:8080 as it instructs and nothing comes up. > > The ISP think its something wrong with the router. > > Please help - this is driving me mad now. > The "home" routers such as Netgear, Linksys, and DLink are "NAT" routers. They work fairly well for outgoing connections. Incoming connections are something else! If your internal addresses are RFC-1918 private addresses; e.g. 192.168.1.*, or 172. or 10. then your router is doing NAT. The router should be able to map the external address/port combination to an internal address/port if you configure it properly. The home routers also act as "stateful packet inspection" firewalls which means that incoming traffic is blocked by default unless the connection was initiated from inside the firewall. You may be able to configure such a router to do what you want. RTFM or pay for a "real" router; e.g. Cisco. ------------------------------ Date: 30 Jan 2007 09:57:39 -0800 From: elementyl@hotmail.com Subject: Re: Cannot ping ADSL from external Message-ID: <1170179859.892924.315310@a34g2000cwb.googlegroups.com> On Jan 30, 3:58 am, mik...@ntlworld.com wrote: > They have given me a static IP address of 212.159.66.255. This is a broadcast address, so you should double check the static IP address. ------------------------------ Date: Tue, 30 Jan 2007 15:56:07 GMT From: Tad Winters Subject: Re: cannot restore SYSUAF Message-ID: colonel@monmouth.com (World's Largest Leprechaun) wrote in news:epnmv4 $65k$1@shell.monmouth.com: > I have booted an Alpha running VMS 7.3-2 from CD-ROM > and am trying to restore the user authorization file > on the hard drive. Yesterday I increased SYSTEM's > process limit from 10 to 100, and since then I have > not been able to log in normally. > > $$$ dir sysuaf* > SYSUAF.DAT;1 SYSUAF.TEMPLATE;1 > $$$ copy sysuaf.template sysuaf.dat > %CLI-F-SYNTAX, error parsing 'BLOCK_SIZE' > -CLI-E-ENTNF, specified entity not found in command tables It would appear you're copying the template file to its original directory. Since it's a read-only device, that's not going to happen. You're going to need to specify the disk device, along with the directory (e.g. DKA100:[SYS0.SYSEXE]SYSUAF.DAT) > $$$ > > Quoting the arguments to COPY does not help. What am I doing wrong? > > -:- > "To what do I owe the honour of this unexpected visit, Lord > Ruthven . . . alias Lyford Pemberton!" > > H.-C. Artmann, "Tom Parker, International Detective" ------------------------------ Date: 30 Jan 2007 10:12:15 -0600 From: briggs@encompasserve.org Subject: Re: cannot restore SYSUAF Message-ID: In article , colonel@monmouth.com (World's Largest Leprechaun) writes: > I have booted an Alpha running VMS 7.3-2 from CD-ROM > and am trying to restore the user authorization file > on the hard drive. Yesterday I increased SYSTEM's > process limit from 10 to 100, and since then I have > not been able to log in normally. > > $$$ dir sysuaf* > SYSUAF.DAT;1 SYSUAF.TEMPLATE;1 > $$$ copy sysuaf.template sysuaf.dat > %CLI-F-SYNTAX, error parsing 'BLOCK_SIZE' > -CLI-E-ENTNF, specified entity not found in command tables > $$$ The "COPY" command that you are using (SYS$SYSTEM:COPY.EXE would be usual) is attempting to parse the command line. The command line syntax is described in your process command tables (SYS$SHARE:DCLTABLES would be usual). The version of DCLTABLES that your process is using does not match the version of DCLTABLES that COPY.EXE expects. Your COPY.EXE expects that the syntax for COPY includes a BLOCK_SIZE qualifier. But your DCLTABLES does not think that the COPY syntax includes a BLOCK_SIZE qualifier. Since you say that you booted from CDROM, it seems that this inconsistency is buried in the CDROM. One obvious possibility would be to use BACKUP rather than COPY. $ BACKUP SYSUAF.TEMPLATE SYSUAF.DAT /NEW_VERSION /LOG The /NEW_VERSION ensures that you don't overwrite the existing SYSUAF.DAT. BACKUP will, by default, refuse to copy a file if the version already exists. The /LOG ensures that you see exactly what BACKUP does. I include it out of a healthy sense of paranoia. BACKUP should work -- it's the usual purpose of a boot from CDROM. NOTE 1: Rather than overwriting your SYSUAF, my inclination would be to do a conversational boot, and run AUTHORIZE to change the PRCLM back. NOTE 2: Changing PRCLM should not have caused any such problem. So I'd be using that conversational boot to try to figure out what really went wrong. And I'd be posting symptoms to the newsgroup rather than asking for ways to fix something that might not even be the problem. ------------------------------ Date: Tue, 30 Jan 2007 16:15:33 -0000 From: "Richard Brodie" Subject: Re: cannot restore SYSUAF Message-ID: "World's Largest Leprechaun" wrote in message news:epnmv4$65k$1@shell.monmouth.com... > $$$ copy sysuaf.template sysuaf.dat > %CLI-F-SYNTAX, error parsing 'BLOCK_SIZE' > -CLI-E-ENTNF, specified entity not found in command tables > $$$ You are using a version of COPY program that is out of step with the command line definition in DCLTABLES. > Quoting the arguments to COPY does not help. What am I doing wrong? I'd use BACKUP to do a simple file COPY if COPY was having a bad day. ------------------------------ Date: 30 Jan 2007 12:07:12 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: cannot restore SYSUAF Message-ID: In article , colonel@monmouth.com (World's Largest Leprechaun) writes: > > $$$ dir sysuaf* > SYSUAF.DAT;1 SYSUAF.TEMPLATE;1 > $$$ copy sysuaf.template sysuaf.dat > %CLI-F-SYNTAX, error parsing 'BLOCK_SIZE' > -CLI-E-ENTNF, specified entity not found in command tables > $$$ If $$$ is the standalone BACKUP prompt, then only the BACKUP command will work. But I thought you said you were using an Alpha, which means you at least have some minimal version of VMS booted. In that case either COPY is defined as a symbol somewhere with a BLOCK_SIZE parameter, or your DCLTABLES.EXE is out of sync with other parts of the operating system. (COPY should not have a BLOCK_SIZE qualifier.) Alternatively, if you're following the instructions for recovering a system when the SYSTEM password is unknown, you've missed a step. ------------------------------ Date: 30 Jan 2007 05:37:42 -0800 From: "Dave Gullen" Subject: Re: Delayed telnet connection Message-ID: <1170164262.050024.300550@a75g2000cwd.googlegroups.com> Something else that can cause similar behaviour to this is inconsistent settings between the server NIC and the the network port it is connecting to. It's worth checking that both are set to 100/ Full or whatever your network supports. MCR LANCP LANCP> SET DEV ewa0 /FULL/SPEED=100 LANCP> DEF DEV ewa0 /FULL/SPEED=100 or you can do it from the console >>> prompt. I remember some Alpha models had cards that didn't autonegotiate well, so these days I tend not to use it. D. ------------------------------ Date: 30 Jan 2007 07:40:52 -0800 From: "tadamsmar" Subject: Disk space problem Message-ID: <1170171652.426528.307070@a75g2000cwd.googlegroups.com> I was updating my VMS 7.3.2 systems with the UPDATE 9 patch. But I pulled an old AXP 150 out of storage and found that it has too little disk space for UPDATE 9. We tried to upgrade the disks (it has a shadowset) a while back, but it seems to be finicky about the disks it will use. (We have put big disks on all our other Alphas) PRODUCT INSTALL says I need an extra 216,901 blocks. My dump file is has already been reduced to almost nothing. My page file is at 270,000. swap file at 17,500. PRODUCT INSTALL says that UPDATE 9 takes about 480,000 blocks. I guess I could just install the TDF patch and it's dependencies and give up on running the same patches on all my Alpha boxes. Any other ideas? ------------------------------ Date: 30 Jan 2007 09:40:49 -0800 From: etmsreec@yahoo.co.uk Subject: ECP and TDC on Integrity, VMS v8.2-1 Message-ID: <1170178849.731039.270980@a34g2000cwb.googlegroups.com> Hiya, I have a client running an Integrity server with VMS v8.2-1. They're running TDC v2.2-108 and, although it seems to be populating its output files with something (a typical day's .TDC$DAT output file being over 133000 blocks), the ECP analyzer tool on my Alpha doesn't appear to think that there are any data in the file. The ECP Analyzer is v5.5A-25 and is on VMS v8.2 on the Alpha. Anybody know what the incantation or versions are that will allow me to see what the Integrity server is doing with itself? Thanks in advance Steve ------------------------------ Date: 30 Jan 2007 08:54:20 -0800 From: "issinoho" Subject: GNV for Alpha 8.2 Message-ID: <1170176060.480938.194180@s48g2000cws.googlegroups.com> Can anyone tell me where to download a version of GNV suitable for use on OpenVMS Alpha 8.2 The HP website only seems to link to v2.1 which refuses to install on anything before 8.3 Thanks. ------------------------------ Date: Tue, 30 Jan 2007 12:05:20 -0500 From: "Syltrem" Subject: Help needed with MONITOR /SUMMARY Message-ID: <12ruulv4j3sc48e@corp.supernews.com> Hi I recorded a day's long of activity using $ MONITOR SYSTEM,PROC/TOPCPU /RECORD=X.X Now I need to be able to tell what is the max usage of the CPUs during that whole time. And also the average. Or it can be shown as user time vs idle time and system/kernel time I need to show people that their machine is not being used much. They have a DS20 that I see is no more than 0% to 5% (on 200%) busy 99% of the time, and they are complaining it is too slow and they want to buy a faster machine (Itanium) I tried /summary=X.LIS /display but am not sure how to interpret all the numbers there Thanks! Syltrem ------------------------------ Date: 30 Jan 2007 07:21:58 -0800 From: magalettac@hotmail.com Subject: Monitor rms issue on Vax Message-ID: <1170170518.144836.173970@m58g2000cwm.googlegroups.com> Hi , I am not able to do the following on Vax 7.3 system... $ monitor rms/file=z.z %MONITOR-E-COLLERR, error during data collection -SYSTEM-F-ACCVIO, access violation, reason mask=D0, virtual address=03C00000, PC =00000000, PSL=200000A0 All other monitor switches seem to run fine. The same test runs fine on any of my Alphas, anyone have any ideas ? Thanks, Carmine ------------------------------ Date: 30 Jan 2007 07:35:30 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: PL/I for Itanium Message-ID: In article , glen herrmannsfeldt writes: > He didn't say faster, he said better. It might be easier to write, > debug, and otherwise get working. It might be that the code is enough > better than the Fortran code to be faster. PL/I does have more overhead > than Fortran 66. Static arrays are pretty fast, but they can't change > size when you need them to. Now that Fortran has dynamic arrays it > is probably the same speed as equivalent PL/I. As Fortran has grown, some attention has been paid to keeping its strenght as the best language for high speed number crunching. Generally if you don't use the fancy new features the old optimizations are still possible and still in the compilers. Static arrays can be optimized in ways that dynamic arrays and pointers can't. Ever since pointers were introduced to DEC Fortran some of those optimizations just don't take place if a pointer is seen. ------------------------------ Date: Tue, 30 Jan 2007 06:34:46 -0800 From: "Tom Linden" Subject: Re: PL/I for Itanium Message-ID: On Tue, 30 Jan 2007 05:35:30 -0800, Bob Koehler wrote: > In article , glen > herrmannsfeldt writes: > >> He didn't say faster, he said better. It might be easier to write, >> debug, and otherwise get working. It might be that the code is enough >> better than the Fortran code to be faster. PL/I does have more overhead >> than Fortran 66. Static arrays are pretty fast, but they can't change >> size when you need them to. Now that Fortran has dynamic arrays it >> is probably the same speed as equivalent PL/I. > > As Fortran has grown, some attention has been paid to keeping its > strenght as the best language for high speed number crunching. > Generally if you don't use the fancy new features the old > optimizations are still possible and still in the compilers. > > Static arrays can be optimized in ways that dynamic arrays and > pointers can't. Ever since pointers were introduced to DEC Fortran > some of those optimizations just don't take place if a pointer is > seen. > It is aliasing that makes optimization more of a challenge. -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ ------------------------------ Date: Tue, 30 Jan 2007 01:31:58 -0800 From: glen herrmannsfeldt Subject: Re: Power down to remove SCSI tape drive? Message-ID: tadamsmar wrote: > I need to make temporary use of a external SCSI connected TLZ06. > Is it better to power down the Alphastation 400 before removing TLZ06 > from the SCSI cable? > I would just as well not power down, but I don't want to damage > something. I believe SCSI uses open collector drivers, which normally are safe against damage from hot swap. Data safety is a different question. If you do it during a transfer, it might be possible to lose data. In the old Sun days one would go into the ROM monitor to stop the OS while changing things, then continue. I don't know how many machines now can do that. There are stories of Apple directly driving the SCSI bus without buffers when they would normally be required. That made it a little more sensitive than it otherwise should have been. I usually power down. Most systems need to reboot to detect the changes anyway. -- glen ------------------------------ Date: 30 Jan 2007 02:08:06 -0800 From: etmsreec@yahoo.co.uk Subject: Re: Power down to remove SCSI tape drive? Message-ID: <1170151685.925196.322030@k78g2000cwa.googlegroups.com> Assuming it's VMS since the message was in c.o.v... Safer to power down, less risk of damage from powering down first, but if you want to do it hot then it should survive. When adding the drive you'll need to do MC SYSMAN IO AUTO [/LOG] (the /LOG is optional, but it tells you what additional hardware the system's found). When you disconnect the drive again you'll still get the "ghost" device until VMS is rebooted. The system will still show an MK device for the tape drive, but it won't be usable (since you'll have disconnected the drive). Steve On 29 Jan, 19:21, "tadamsmar" wrote: > I need to make temporary use of a external SCSI connected TLZ06. > > Is it better to power down the Alphastation 400 before removing TLZ06 > from the SCSI cable? > > I would just as well not power down, but I don't want to damage > something. ------------------------------ Date: Tue, 30 Jan 2007 11:02:10 +0000 From: "R.A.Omond" Subject: Re: Power down to remove SCSI tape drive? Message-ID: glen herrmannsfeldt wrote: > [...snip...] > In the old Sun days one would go into the ROM monitor to stop the > OS while changing things, then continue. I don't know how many > machines now can do that. The equivalent on VMS (standalone!) would be to hit Control-P (or Break) to get to the chevrons, doing the "stuff" and then Continuing. ------------------------------ Date: 30 Jan 2007 05:18:58 -0800 From: "tadamsmar" Subject: Re: Power down to remove SCSI tape drive? Message-ID: <1170163137.989409.230510@a34g2000cwb.googlegroups.com> On Jan 29, 4:33 pm, koeh...@eisner.nospam.encompasserve.org (Bob Koehler) wrote: > In article <1170098508.052327.164...@p10g2000cwp.googlegroups.com>, "tadamsmar" writes: > > > I need to make temporary use of a external SCSI connected TLZ06. > > > Is it better to power down the Alphastation 400 before removing TLZ06 > > from the SCSI cable? > > > I would just as well not power down, but I don't want to damage > > something. SCSI by itself is not designed to be hot swappable, although people > have added hot swappable connections to SCSI based IO systems. > Vanilla AlphaStation 400 and TLZ06 do not come with such additional > layers. > > Experts will tell you that if the bus is quiet you can get away with > it. I never do it, and won't think of doing it if I have a disk on > the bus with data I might want to use someday. How about if I shutdown to the >>> prompt? You seem to be saying that I need to turn off the power switch. ------------------------------ Date: 30 Jan 2007 07:24:16 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Power down to remove SCSI tape drive? Message-ID: In article <1170163137.989409.230510@a34g2000cwb.googlegroups.com>, "tadamsmar" writes: > > How about if I shutdown to the >>> prompt? That should certainly make the bus quiet. > You seem to be saying that I need to turn off the power switch. > Once your shutdown to >>>, is there a big issue cycling power? Or are you just temporarily halting a non-clustered system, as suggested in earlier posts? I used to do the latter on my 11/780. Had to hurry or DECnet connections would fail. Typically I was the only one using DECnet, anyhow. ------------------------------ Date: 30 Jan 2007 07:15:16 -0800 From: "tadamsmar" Subject: Re: Power down to remove SCSI tape drive? Message-ID: <1170170116.764020.70990@k78g2000cwa.googlegroups.com> On Jan 30, 8:24 am, koeh...@eisner.nospam.encompasserve.org (Bob Koehler) wrote: > In article <1170163137.989409.230...@a34g2000cwb.googlegroups.com>, "tadamsmar" writes: > > > > > How about if I shutdown to the >>> prompt? That should certainly make the bus quiet. > > > You seem to be saying that I need to turn off the power switch. Once your shutdown to >>>, is there a big issue cycling power? > Or are you just temporarily halting a non-clustered system, as > suggested in earlier posts? > > I used to do the latter on my 11/780. Had to hurry or DECnet > connections would fail. Typically I was the only one using DECnet, > anyhow. The system is non-clustered. The only issue with power-down is hardware preservation. One of my AlphaStation 400s has shown some unreliability with not restarting after a powerdown. I have UPSes on both of them. Just trying for maximum life extension. But, maybe I am obsessing about the boxes that have not shown restart unreliability. Just trying out the idea of never powering them down. ------------------------------ Date: Tue, 30 Jan 2007 15:58:56 GMT From: Tad Winters Subject: Re: Power down to remove SCSI tape drive? Message-ID: "tadamsmar" wrote in news:1170170116.764020.70990@k78g2000cwa.googlegroups.com: > The system is non-clustered. The only issue with power-down is > hardware preservation. One of my AlphaStation 400s has shown some > unreliability with not restarting after a powerdown. I have UPSes on > both of them. Just trying for maximum life extension. But, maybe I > am obsessing about the boxes that have not shown restart > unreliability. Just trying out the idea of never powering them down. Personally, I've just shutdown to the console prompt and make the hardware change. Then issue the INITIALIZE command at the console prompt. That should give you the desired affect without complete power down. ------------------------------ Date: Tue, 30 Jan 2007 07:54:58 -0500 From: JF Mezei Subject: Re: PPPD, TCPIP Services and incomplete negotiations Message-ID: <399c8$45bf4026$cef8887a$12631@TEKSAVVY.COM> Oh. The way I free the terminal device is by STOP/ID the disconnected process that issued the PPPD CONNECT command. ------------------------------ Date: 30 Jan 2007 07:50:43 -0800 From: "tadamsmar" Subject: PRODUCT INSTALL question Message-ID: <1170172242.963938.161170@v45g2000cwv.googlegroups.com> I have the PCSI$COMPRESSED file for the 7.3.2 UPDATE V9 patch. Is there any way to extract specific patches from that? For instance, can I pull out the TDF V3 patch and just apply that one? Or do I have to go get the specific patch from the vms web site? I can't find a way to do only a specific patch using product install. My Alphas are not on the WWW so I have to go through some rigamarole to get the patch in place. ------------------------------ Date: 30 Jan 2007 10:13:22 -0800 From: bob@instantwhip.com Subject: Re: Purveyor CGI mailbox capacity [bit long winded] Message-ID: <1170180802.457065.9690@q2g2000cwa.googlegroups.com> poll about what? We want to use it and obviously someone else wants too also ... are you saying that yahmail also truncated in Purveyor? Just an idea ... it says you can avoid vms mailboxes and restrictions by writing a .DLL in "c" to handle this ... what about writing one? ------------------------------ Date: 30 Jan 2007 02:03:48 -0800 From: etmsreec@yahoo.co.uk Subject: Re: VMS in the HP hierarchy Message-ID: <1170151428.760760.122190@a34g2000cwb.googlegroups.com> Maybe Ann McQuaid's job is to convince the "layers who are not=20 interested in VMS" that they should be? Marketing has to be internal=20 as well as external. Engineering teams know each other are there and worked together in=20 porting VMS to Integrity. There's no reason to suspect that the other=20 teams don't do the same. Steve On 30 Jan, 03:49, JF Mezei wrote: > Arne Vajh=F8j wrote: > > If the head of VMS is to report directly to the CEO, then I > > guess VMS would need to deliver something in the 10-25 B$ range.I didn'= t suggest that McQuaid should report directly to Hurd. But one has > to make sure she is close enough that Hurd woulod get her point of view > unfiltered by layers who are not interested in VMS. ------------------------------ End of INFO-VAX 2007.060 ************************