INFO-VAX Sun, 15 Apr 2007 Volume 2007 : Issue 208 Contents: Dead clock, NTP and 1000 seconds Re: Dead clock, NTP and 1000 seconds Jetty6 for OpenVMS Re: Problems with TCPIP startup on boot, was Re: Supported devices Re: Problems with TCPIP startup on boot, was Re: Supported devices Re: Sun shows Rock first silicon Re: Sun shows Rock first silicon TCPIP SMTP: suggestion Re: TCPIP SMTP: suggestion ---------------------------------------------------------------------- Date: 15 Apr 2007 03:00:06 -0700 From: "Rambo" Subject: Dead clock, NTP and 1000 seconds Message-ID: <1176631205.995392.180460@n76g2000hsh.googlegroups.com> I noticed I had dead battery in my DEC2000 (and it's the 4v cube, that is "plugged" to motherboard with a cable, so the replacement won't be arriving son). I've enabled successfully the NTP services, but it keeps dying because of sanity drift limit of 1000 secs- I have to set date anyway. My question is: Is there a way to disable that limit? Rambo ------------------------------ Date: Sun, 15 Apr 2007 06:49:54 -0400 From: John Sauter Subject: Re: Dead clock, NTP and 1000 seconds Message-ID: On 15 Apr 2007 03:00:06 -0700, "Rambo" wrote: >I noticed I had dead battery in my DEC2000 (and it's the 4v cube, that >is "plugged" to motherboard with a cable, so the replacement won't be >arriving son). > >I've enabled successfully the NTP services, but it keeps dying because >of sanity drift limit of 1000 secs- I have to set date anyway. My >question is: Is there a way to disable that limit? > >Rambo Add tinker panic 0 to your ntp.conf file. See for background and more details. John Sauter (J_Sauter@Empire.Net) ------------------------------ Date: 15 Apr 2007 09:39:50 -0700 From: thierry.uso@wanadoo.fr Subject: Jetty6 for OpenVMS Message-ID: <1176655190.326810.34210@n59g2000hsh.googlegroups.com> Jetty is a Web server (HTTP server and servlet container) for static and dynamic content written in Java under Apache 2.0 license. Jetty6 is a new implementation of Jetty which supports Servlet 2.5 and JSP 2.1. Moreover, Jetty6 meets the needs of AJAX applications and has a better scalability than the previous versions. Jetty 6.1.1 has been ported on OpenVMS. Porting means testing all the functions, writing startup and shutdown procedures and creating a PCSI kit. Jetty6 for OpenVMS needs Java 1.4.2 or later. It MUST be installed on an ODS5 volume. Jetty6 has been tested on Alpha and Itanium. Download link: http://vmsfree.free.fr/freen/index.php?s=jetty6 ------------------------------ Date: 15 Apr 2007 09:52:11 +0200 From: eplan@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) Subject: Re: Problems with TCPIP startup on boot, was Re: Supported devices Message-ID: <4621f5cb$1@news.langstoeger.at> In article <1176600770.146411.201540@n59g2000hsh.googlegroups.com>, "Rambo" writes: >Bottom line: DECwindows complains about too little free GBLSECTIONS (235 >instead of 280), and wants to run SYSGEN- yet it doesn't help, after disabling >TCPIP everything gets back to normal again (so it's likely hardware limit). Yup. AUTOGEN doesn't increase GBLSECTIONS high enough to start successfully. Known problem. I entered a line to my SYS$SYSTEM:MODPARAMS.DAT at these days: MIN_GBLSECTIONS = 1024 ! help VMS to make DECwindows run btw: Why not running VMS V8.3 (and TCPIP V5.6 and DECwindows-MOTIF V1.6)? -- 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: 15 Apr 2007 02:48:06 -0700 From: "Rambo" Subject: Re: Problems with TCPIP startup on boot, was Re: Supported devices Message-ID: <1176630486.588031.159160@n76g2000hsh.googlegroups.com> > MIN_GBLSECTIONS = 1024 ! help VMS to make DECwindows run Added, AUTOGENned, works :) Thanks. And that's double thanks, because increasing that value made Mozilla work! :) (on EV4/150! and it FLIES!) > btw: Why not running VMS V8.3 (and TCPIP V5.6 and DECwindows-MOTIF V1.6)? Silly reason, while AS600 would even be supported, Jensen is not (7.3-1 is possibly the last to have support for DEC2000). There's also "availability" of media- I'm pretty much on deserted island here :-) Rambo ------------------------------ Date: Sun, 15 Apr 2007 08:05:58 -0400 From: "Neil Rieck" Subject: Re: Sun shows Rock first silicon Message-ID: <4622086e$0$16392$88260bb3@free.teranews.com> "Andrew" wrote in message news:1176470355.483306.114870@d57g2000hsg.googlegroups.com... > > After quietly announcing 1.95Ghz and 2.1Ghz Dual Core UltraSPARC IV+ > modules Sun have show pictures of the first Rock Silicon along with a > claim that these are working chips. > > http://www.theregister.co.uk/2007/04/10/sun_rock_schwartz/ > > These announcements come at a good time for Sun after recent reports > of delays to the release of Power 6 from IBM though an expected early > or on time delivery of Rock may weaken sales of the servers Sun has > developed with Fujitsu which are due for release next week. > > Regards > Andrew Harrison > Yikes! This thing has 16 cores. Here is an excerpt from Jonathon's Blog (which also appears in the article)... ### "Rock is 16 cores - we haven't said how many threads per core. Nor have we said why this chip heralds the golden age of effortless parallel programming, or how it brings fault tolerance to the masses. But stay tuned, I think we're planning on talking up both in the next few weeks." In the past, we've disclosed that Rock will run two threads per core, giving it 32 threads per chip. In addition, the 256TB of memory described by Schwartz would come via a 8-socket box that holds 512 DIMMs, depicted here. Our sources have indicated that Sun stopped work on systems any larger than the Platinum box. ### Like it or not, this is another proof of why it was dumb to kill Alpha and then defer that problem to Intel ("hey, we want to be more like Dell"). At the Canadian seminar I attended in February-2007 the HP people said that Itanium won't go quad-core until 2009 (citing the lack of proper chip sets). So now our favourite OS is running on a chip that won't go quad core until 2009 while the chip's vendor has plans to improve x86-64 which HP has decided not to port OpenVMS to (catch-22). I'm no marketing person but the only way I can see HP getting out of this dilemma is to sell SMP based systems for a price equivalent to the multi-core systems. This may mean that they loose money on each CPU (which would put them right back to where they started when then told us that they lost money on every Alpha) but what else can they do after cutting loose so many OpenVMS engineers? Neil Rieck Kitchener/Waterloo/Cambridge, Ontario, Canada. http://www3.sympatico.ca/n.rieck/ -- Posted via a free Usenet account from http://www.teranews.com ------------------------------ Date: Sun, 15 Apr 2007 08:46:10 -0400 From: "Neil Rieck" Subject: Re: Sun shows Rock first silicon Message-ID: <462211db$0$6888$88260bb3@free.teranews.com> "Larry Kilgallen" wrote in message news:e9AoFMDCdKbF@eisner.encompasserve.org... > [...] > > That is not the issue. The point is, who claims this is a VMS matter ? > It has long ago been acknowledged there are chips on which VMS does > not run. This is not the newsgroup for discussing them. > > Perhaps instead of splitting the newsgroup we need to make it moderated > since people have so little self-control. > I see no problem with discussing announcements from the competition. 1) first off, while "some" HP employees "post here" I have been told that many others are just lurkers. It just might be possible that some dissent found here will filter back to HP (although Compaq/HP decisions over the years make me question that thought) 2) occasionally I get dragged into meetings where someone else is flogging new technology from another vendor. In these situations things seem to go better for me if I've already heard of the stuff and am also aware of a few potential warts. p.s. If I just want the rose-colored glasses view I can always check out the HP site. NSR -- Posted via a free Usenet account from http://www.teranews.com ------------------------------ Date: Sun, 15 Apr 2007 10:07:52 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: TCPIP SMTP: suggestion Message-ID: With the anti-spam features, sometimes a message is in the log file (apparently a username but has no account), sometimes it is an OPCOM (matched RBL, relay is suspected spam, in bad-clients list). At first I thought that some spam might be recognised before the receiver log file is generated, but that can't be the case, since after setting up an RBL, I saw an OPCOM saying that the IP address matched RBL list, but then the same address was in the log file saying that the IP address is unbacktranslatable (I don't reject mail based on this, but the message always appears if it is unbacktranslatable). So, that means that even though the address matched an RBL, a receiver log file was generated. I think it would make sense if all messages were everywhere, or at least if one could configure them to be in the log file of the receiver, as an OPCOM or both. Perhaps this could be done via the SPAM-ACTION flag. For now, I set up just one RBL: ! sbl-xbl: Combined zone to reduce queries. Includes both SBL and XBL. ! zen.spamhaus.org: sbl-xbl.spamhouse.org + pbl.spamhouse.org ! pbl.spamhouse.org is recommended, but might block legitimate email ! from dynamic IP addresses. I will wait on this until I can determine ! a) if I need it (i.e. I am still getting too much spam) and b) if I ! want to block some legitimate mail as a side-effect. RBLs: sbl-xbl.spamhaus.org Some internet research a few weeks ago resulted in a list of RBLs which I thought might make sense for me. There were several in sorbs.net and, in addition, bl.spamcop.net, list.dsbl.org and psbl.surriel.com. How large is the overlap between these and sbl-xbl.spamhaus.org? More lists mean more blockage, but also more DNS lookup time. ------------------------------ Date: Sun, 15 Apr 2007 13:07:19 +0000 (UTC) From: moroney@world.std.spaamtrap.com (Michael Moroney) Subject: Re: TCPIP SMTP: suggestion Message-ID: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: >For now, I set up just one RBL: >! sbl-xbl: Combined zone to reduce queries. Includes both SBL and XBL. >! zen.spamhaus.org: sbl-xbl.spamhouse.org + pbl.spamhouse.org >! pbl.spamhouse.org is recommended, but might block legitimate email >! from dynamic IP addresses. I will wait on this until I can determine >! a) if I need it (i.e. I am still getting too much spam) and b) if I >! want to block some legitimate mail as a side-effect. >RBLs: sbl-xbl.spamhaus.org >Some internet research a few weeks ago resulted in a list of RBLs which >I thought might make sense for me. There were several in sorbs.net and, >in addition, bl.spamcop.net, list.dsbl.org and psbl.surriel.com. How >large is the overlap between these and sbl-xbl.spamhaus.org? More lists >mean more blockage, but also more DNS lookup time. A good one is zen.spamhaus.org. It includes sbl-xbl plus their pbl, which is a list of addresses that shouldn't be sending email directly. This catches many 'zombies', and 'zombies' are the current favorite way for spammers to send spam. ------------------------------ End of INFO-VAX 2007.208 ************************