INFO-VAX Fri, 14 Jan 2005 Volume 2005 : Issue 27 Contents: Re: An interesting little SSD ANN: New and updated freeware from Lyle W. West (NOPE, VIEWFILE, etc) etc)etc) DVE and maximum file count JVM much slower on Itanium box (rx2600) Re: JVM much slower on Itanium box (rx2600) Mass mailing from Gorham Re: Mass mailing from Gorham Re: Mass mailing from Gorham Re: Mass mailing from Gorham Re: Mass mailing from Gorham Re: Mass mailing from Gorham Re: Mass mailing from Gorham Re: Microvax II questions Re: Multiple-site Fibre Channel configurations New VMS TCO study due to be posted Re: New VMS TCO study due to be posted OpenVMS 8.2 on Itanium... any emulators? Re: OpenVMS 8.2 on Itanium... any emulators? Re: Q: How to have SEARCH match a "null" line Re: Q: How to have SEARCH match a "null" line Re: Q: How to have SEARCH match a "null" line Re: TCPIP SSH failure logging Re: VTfm new release 2.3-2 ---------------------------------------------------------------------- Date: 13 Jan 2005 21:44:22 -0600 From: young_r@encompasserve.org (Rob Young) Subject: Re: An interesting little SSD Message-ID: <8$vl7ZkLTybu@eisner.encompasserve.org> In article , "John Smith" writes: > Could something like this work on an Integrity server with VMS, perhaps > something similar in concept if not this exact device? > > Manufacturer > http://www.acceleratedlogic.com/index.htm > > http://www.hyperos2002.com/ > (more info ATA-100 version - scroll to bottom) > > http://www.hyperos2002.com/07042003/img/HD2zoom.gif > (large image) > The markets for these products are very narrow, especially in the VMS world. Host cache, write-back cache, large storage caches (blech), RAID, LUNs that sit on many physical disks - have marginalized these products to the extreme needs realm. The SSD wars are occasionally played out over in comp.arch.storage. Bottom line, the SSD's of the world are a dying breed. Imperial Tech (one of the more famous, larger SSD vendors) is gone: http://tinyurl.com/64sfk Maybe it is a marketing problem? ;-) LOL Rob ------------------------------ Date: Thu, 13 Jan 2005 21:17:24 -0600 From: Hunter Goatley Subject: ANN: New and updated freeware from Lyle W. West (NOPE, VIEWFILE, etc) etc)etc) Message-ID: <41E739C4.6040306@goatley.com> Thanks to Lyle W. West, the following software has been added to and updated in my VMS freeware archives. o LWW-REVERT V1.2-2 - Changes file version numbers of files in a directory (update) o LWW-UTC V1.2 - A Motif UTC clock (new!) o LWW-VIEWXBM V1.2-8 - A Motif-based Bitmap viewer (new) o LWW-XMEASURE V1.2-2 - Measure Motif-based windows (new) o NOPE V1.2-1 - Simulate a PURGE to see what would be deleted (NEW!) o ViewFile V1.2-2 -- Motif-based file viewer/browser (update) Of particular note is the NOPE utility, which I think looks especially useful. http://www.process.com/openvms/ ftp://ftp.process.com/vms-freeware/fileserv/ http://vms.process.com/ftp/vms-freeware/fileserv/ ftp://ftp.tmk.com/vms-freeware/fileserv/ http://www.tmk.com/ftp/vms-freeware/fileserv/ And on the other mirrors soon. Hunter ------ Hunter Goatley, Process Software, http://www.process.com/ http://www.goatley.com/hunter/ ------------------------------ Date: 13 Jan 2005 14:09:22 -0800 From: "Ed Wilts" Subject: DVE and maximum file count Message-ID: <1105654162.328837.289790@z14g2000cwz.googlegroups.com> With the welcome introduction of dynamic volume expansion, I expect that eventually we'll find ourselves with enough disk space but not enough file headers. What are the current plans, if any, on allowing customers to increase this without a volume re-init? Is this even technically do-able? Even an option like mounting the volume privately for a few minutes would be acceptable (like enabling DVE on an existing volume). Thanks, .../Ed ------------------------------ Date: 13 Jan 2005 18:12:14 -0800 From: Don.Zong@gmail.com (Jeff Rogers) Subject: JVM much slower on Itanium box (rx2600) Message-ID: <4da1c6a.0501131812.54c59f27@posting.google.com> The same java program, runs much slower on rx2600 than on ds25. The program is a simple one like. for (int i=1; i<1000000000;i++){ int j = i;} It took 3 seconds on the alpha box but 80 seconds on rx2600. We have a servlet program which usually takes about half seconds on the alpha box to finish but took 35 seconds on the itanium one. On rx2600 two 1.4G IA64 CPUs, 4G memory openvms8.2 field test version, jvm 1.4.2-beta 2 On ds25 2G memory, openvms 7.3-1, jvms 1.4.2 Followed the jvm's user guide on performance tips. Don't see any IO, quota issues on both boxes. Any thoughts? ------------------------------ Date: Fri, 14 Jan 2005 07:09:13 +0100 From: "julien courtemanche" Subject: Re: JVM much slower on Itanium box (rx2600) Message-ID: <41e76209$0$25781$8fcfb975@news.wanadoo.fr> according to me, the field test java version is not a HP fast VM version but a classic one. so, I'm agree for the moment it's very very slow..... "Jeff Rogers" a écrit dans le message de news: 4da1c6a.0501131812.54c59f27@posting.google.com... > The same java program, runs much slower on rx2600 than on ds25. > > The program is a simple one like. > > for (int i=1; i<1000000000;i++){ int j = i;} > > It took 3 seconds on the alpha box but 80 seconds on rx2600. > > We have a servlet program which usually takes about half seconds on > the alpha box to finish but took 35 seconds on the itanium one. > > On rx2600 two 1.4G IA64 CPUs, 4G memory openvms8.2 field test version, > jvm 1.4.2-beta 2 > > On ds25 2G memory, openvms 7.3-1, jvms 1.4.2 > > Followed the jvm's user guide on performance tips. Don't see any IO, > quota issues on both boxes. > > Any thoughts? ------------------------------ Date: Thu, 13 Jan 2005 15:02:25 -0500 From: "John Smith" Subject: Mass mailing from Gorham Message-ID: Dear valued OpenVMS customer, On January 18, 2005, HP will announce the availability of OpenVMS version 8.2 for Integrity servers based on the Intel® Itanium® 2 processor and AlphaServer systems. Please join us for this watershed announcement as part of HP's major Enterprise Computing Web Event. I encourage you to participate by registering at http://www.hp.com/go/agility. OpenVMS Version 8.2 is currently orderable for HP Integrity rx4640-8, rx2620-2, rx2600, rx1620-2 and rx1600 servers as well as the entire family of AlphaServer systems. OpenVMS will be available on the full line of Integrity servers, including the HP Superdome Server, by the first half of 2006. You can now take full advantage of the availability, scalability, and rock-solid reliability of OpenVMS on the industry standard Integrity server platform. If you currently run OpenVMS, you can be assured that your investments are protected. You can increase your returns on IT by adding OpenVMS Integrity servers to your existing AlphaServer environment. Legendary OpenVMS clustering is now extended to support both AlphaServer systems and Integrity servers as well as mixed architecture clusters that can include both Alpha and Integrity servers. Application investments are protected because the vast majority of Alpha-based OpenVMS applications need only be recompiled, relinked, and tested in order to run on Integrity servers. HP believes that the timing is also excellent for new OpenVMS customers because they can have the well-recognized functionality of OpenVMS as well as affordability, flexibility, adaptability, and low total cost of ownership. Ann Livermore, Executive Vice President for HP's Technology Solutions Group states, "OpenVMS continues to be a strategic platform for HP. This is demonstrated by the new release of OpenVMS Version 8.2 for both Integrity servers and AlphaServer systems. With OpenVMS now supporting Integrity Servers, we have expanded our multi-OS capability. As well, we are delivering another key offering as part of our Adaptive Enterprise strategy." Over 100 customers have field tested OpenVMS v8.2 and hundreds more have been experiencing it for months in its evaluation release. HP has the programs and support to enable you to seamlessly add Integrity servers into your existing IT environment. Our list of leading software partners porting their applications to Integrity servers continues to grow daily. In a business environment driven by relentless, unpredictable, and accelerating change, OpenVMS will continue to improve on its well-known attributes of availability, scalability, and rock-solid reliability on which customers, worldwide, continue to depend. "Our customers and channel partners are enormously enthusiastic about OpenVMS on HP Integrity servers" says Jack Novia, senior vice president, Customer Solutions Group, and managing director, HP Americas. "They know that we at HP are strongly committed to supporting their current IT investments, while providing a solid path forward for continuous improvement enabling them to capitalize on change". For more information, visit: http://www.hp.com/go/openvms The web event will be at: http://www.hp.com/go/agility. Very truly yours, Mark Gorham, VP OpenVMS Systems Division ----------------- You don't hear that carly(tm) thinks OpenVMS is strategic. I've got money riding on it that January 18th doesn't change that. ------------------------------ Date: Thu, 13 Jan 2005 21:04:31 GMT From: "John Vottero" Subject: Re: Mass mailing from Gorham Message-ID: "John Smith" wrote in message news:kJmdnQDkCZhaTnvcRVn-iA@igs.net... > > You don't hear that carly(tm) thinks OpenVMS is strategic. I've got money > riding on it that January 18th doesn't change that. > I've got money riding on you complaining no matter what happens. ------------------------------ Date: 13 Jan 2005 13:23:00 -0800 From: susan_skonetski@hotmail.com Subject: Re: Mass mailing from Gorham Message-ID: <1105651380.408374.78910@c13g2000cwb.googlegroups.com> John Vottero wrote: > "John Smith" wrote in message > news:kJmdnQDkCZhaTnvcRVn-iA@igs.net... > > > > > You don't hear that carly(tm) thinks OpenVMS is strategic. I've got money > > riding on it that January 18th doesn't change that. > > > > I've got money riding on you complaining no matter what happens. I sent mail to my distribution lists asking that this letter not be posted until the 18. sue ------------------------------ Date: Thu, 13 Jan 2005 17:23:31 -0500 From: "John Smith" Subject: Re: Mass mailing from Gorham Message-ID: John Vottero wrote: > "John Smith" wrote in message > news:kJmdnQDkCZhaTnvcRVn-iA@igs.net... > >> >> You don't hear that carly(tm) thinks OpenVMS is strategic. I've got >> money riding on it that January 18th doesn't change that. >> > > I've got money riding on you complaining no matter what happens. Will the world beat a path to your door if you build a better mousetrap and tell nobody about it? I'd like to hear from the coiffed cute one, in her own words, about this. I know that many others in the VMS community are also interested in the words and the tone of the words she speaks. ------------------------------ Date: Thu, 13 Jan 2005 17:20:58 -0500 From: "John Smith" Subject: Re: Mass mailing from Gorham Message-ID: susan_skonetski@hotmail.com wrote: > John Vottero wrote: >> "John Smith" wrote in message >> news:kJmdnQDkCZhaTnvcRVn-iA@igs.net... >> >>> >>> You don't hear that carly(tm) thinks OpenVMS is strategic. I've got >>> money riding on it that January 18th doesn't change that. >>> >> >> I've got money riding on you complaining no matter what happens. > > I sent mail to my distribution lists asking that this letter not be > posted until the 18. Sorry Sue, I'm not on your list. Maybe Mark likes me more than other may think :-) ------------------------------ Date: Thu, 13 Jan 2005 21:20:26 -0500 From: "Brad McCusker" Subject: Re: Mass mailing from Gorham Message-ID: <60GFd.6323$846.4295@fe05.lga> "John Smith" wrote in message news:OJ-dnZvOLIxHaXvcRVn-qQ@igs.net... > and the tone of the words she speaks. > Ahhh.. Hedging your bets I see. So, even if she does have thing to say, you've set yourself up nicely to bitch about the tone. Thank god for kill files. ------------------------------ Date: 13 Jan 2005 20:24:27 -0600 From: Kilgallen@SpamCop.net (Larry Kilgallen) Subject: Re: Mass mailing from Gorham Message-ID: In article <60GFd.6323$846.4295@fe05.lga>, "Brad McCusker" writes: > > "John Smith" wrote in message > news:OJ-dnZvOLIxHaXvcRVn-qQ@igs.net... > >> and the tone of the words she speaks. >> > Ahhh.. Hedging your bets I see. So, even if she does have thing to say, > you've set yourself up nicely to bitch about the tone. > > Thank god for kill files. Kill files are not perfect. For instance they tend to let through posts from people commenting about posts by the killfiled individual :-) ------------------------------ Date: Thu, 13 Jan 2005 15:17:50 -0500 From: "John Smith" Subject: Re: Microvax II questions Message-ID: Dan Foster wrote: > In article <41e6503d$1_2@127.0.0.1>, JF Mezei > wrote: >> >>>> It is a 40 MHz crystal, but divided, divided, and divided until you >>>> get to a cpu cycle time of 200ns. (Whatever cpu cycle means here) > > Wouldn't that yield 5 MHz for the CPU? > > 1 / (200 / 1000000000) > > Not really that shabby considering what it can do and the quality of > the OS running on it. My first VAX that I personally owned was a > MicroVAX II in perhaps a BA23 cabinet? > > The original IBM PC, as a point of comparison, had 120ns RAM which > allowed for a CPU as fast as 8.33 MHz. It bought IBM some time and > CPU speed-ups before they had to look at faster and more expensive > RAM. > > The Apple II had 150ns RAM which allowed for as fast as 6.67 MHz CPU, > but only used 1.023 MHz for the CPU with the later IIc at 4 MHz. > >> Interesting. If this were marketed today, would they market it as a >> 40mhz CPU ? Or in other words, the machines that are marketed at 3 >> gigahertz today, is that just the crystal's rate, or the actual rate >> fed to the CPU ? > > Well, it depends, but the 'master clock oscillator' often drives > different operations. > > For instance, the Apple II's 6502 had a 14.31818 MHz oscillator > crystal. > > It was divided into a "7M" (7.159 MHz -- half of 14.31818 MHz) signal > fed to the peripheral bus. > > The other 7.159 MHz was divided in half to produce a 3.58 MHz clock > that drove the NTSC video signal generation -- both dot clock and > color burst. > > 7.159 MHz was also divided by 7 to yield the effective clock frequency > of 1.023 MHz, for both the phase 0 and phase 1 signals that drove the > peripheral bus access cycles. > > A 1.023 MHz clock signal also drove the 6502's processor operations > and memory accesses. > > Apple marketing called it a 1 MHz machine for simplicity, because that > was essentially what the CPU got to work with. It also rolls off the > tongue a lot better than 1.023 MHz. :-) > > The original clock frequency drove everything important -- the CPU, > memory, video, and peripheral slots' timing... but no single component > got the full 14 MHz clock cycle. > > I doubt that marketing today would normally list the internal clock > oscillator's rating. Even as sneaky as certain companies can be, they > just don't do that. (Though they sometimes use different and > misleading numbers in other areas to sucker the typical consumer.) The original IBM PC/AT had a socketed crystal which could be field 'upgraded'. I did this to a number of machines - IIRC the machine could be overclocked by 33% before it became unstable. (just threw out the last of these overclocking crystals that were hiding in my parts bin a few weeks ago). After about 4-6 months of shipping the PC/AT, IBM soldered the crystals in place at the factory. ------------------------------ Date: 13 Jan 2005 13:40:20 -0800 From: "Ed Wilts" Subject: Re: Multiple-site Fibre Channel configurations Message-ID: <1105652420.417125.153970@c13g2000cwb.googlegroups.com> Makes good sense now. Can I request a doc change to clarify this? It's the sentence "For the Volume Shadowing and cluster software, there is no distinction between local and remote in multiple-site Fibre Channel configurations." that is misleading. Also, Rob, the online help hasn't been fixed in a released remedial kit yet since I'm current and the problem is still there. It's probably still in the queue... Thanks again to you and Rob! .../Ed ------------------------------ Date: Thu, 13 Jan 2005 15:12:30 -0500 From: "John Smith" Subject: New VMS TCO study due to be posted Message-ID: http://www.openvms.org/phorum/read.php?f=6&i=403&t=403 probably in time for the webcast on the 18th ------------------------------ Date: Thu, 13 Jan 2005 13:39:00 -0700 From: "Barry Treahy, Jr." Subject: Re: New VMS TCO study due to be posted Message-ID: <41E6DC64.1050803@MMaz.com> John Smith wrote: >http://www.openvms.org/phorum/read.php?f=6&i=403&t=403 > >probably in time for the webcast on the 18th > > > I would love to see the TCO in relationship to cost vs. performance metrics too and how they compare with Alpha. Prior posts state that I2 hardware, and VMS which I don't get, are cheaper than Alpha... But if you need multiple systems to replace an Alpha, is it really cheaper? Barry -- Barry Treahy, Jr E-mail: Treahy@MMaz.com Midwest Microwave, Inc. Phone: 480/314-1320 Vice President & CIO FAX: 480/661-7028 ... but it's a DRY HEAT! ------------------------------ Date: Thu, 13 Jan 2005 19:02:39 GMT From: "Marco" Subject: OpenVMS 8.2 on Itanium... any emulators? Message-ID: Soon-to-be-released OpenVMS for Intel... Makes me wonder if there's any emulators (VMware or others) that might support a OpenVMS 8.2 install... What do you guys think? Marco ------------------------------ Date: 13 Jan 2005 16:12:45 -0600 From: Kilgallen@SpamCop.net (Larry Kilgallen) Subject: Re: OpenVMS 8.2 on Itanium... any emulators? Message-ID: In article , "Marco" writes: > Soon-to-be-released OpenVMS for Intel... > > Makes me wonder if there's any emulators (VMware or others) that might > support a OpenVMS 8.2 install... > > What do you guys think? I think you are taking a write-before-read approach to this newsgroup. By the time you read this comment, presumably you will have found the existing topics where emulation of non-Vaxen for VMS has been discussed. ------------------------------ Date: Thu, 13 Jan 2005 12:35:36 -0800 From: Ken Fairfield Subject: Re: Q: How to have SEARCH match a "null" line Message-ID: David J Dachtera wrote: > Ken Fairfield wrote: [...] >>Thanks Keith, I know I can do it in two stages. I was trying >>to avoid that, and in particular, I now always try to avoid >>using PIPE in batch jobs (where this will run). There's a know >>issue with PIPE on SMP processors, and for reasons I don't fully >>understand, running in batch seems to expose that issue (possibly >>just from building up statics on multiple runs...) > > > Tell me more, first I've heard of this. In my environment, we run scores of "monitors" periodically, with intervals varying from every 15 minutes to once a day, or once a week. Non-normal conditions detected by these monitors result in notification being sent to the system manager and/or various application owners, etc. Most of these monitors are simple DCL procedures run in batch. Often times we have coded these monitors to use PIPE since it gets tedious (for the programmer) to manage temporary files. There have been a variety of problems reported against PIPE, and I think we've encountered all of them at one time or another. :-( The most typical symptom is that, looking at the monitor's .LOG file, we find a job "hung" in PIPE. The last line of DCL being executed is the PIPE command, but the job is sitting there and never completes. Typically we need to do a DELETE/ENTRY to get rid of it, and occasionally, a STOP/QUEUE/ABORT. The last straw was, having installed the latest patches (to VMS 7.3-1) which claimed to fix the last remaining PIPE bugs, we had a monitor that could reproducibly produce incorrect output some fraction of the time, say 20%-35% of the time. The "reproducer" was sent to VMS engineering, and the person working the case was able to reproduce the problem not only on ES40's at VMS 7.3-1 (which we run), but on a variety of other Alpha hardware and VMS versions. I don't have the case in front of me, but the final answer was that this is a known restriction with SMP systems (and affects unix SMP systems as well) having to do with one of the "downstream" processes in the PIPE completing before the "upstream" processes have completed. OK, found the article that resulted from our problem report. Look in DSNlink, or the ITRC "Natural Language Search", for the article titled, "DCL Pipe Command Inconsistant in Versions 7.3-1 and 7.3-2". -Ken -- I don't speak for Intel, Intel doesn't speak for me... Ken Fairfield D1C Automation VMS System Support who: kenneth dot h dot fairfield where: intel dot com ------------------------------ Date: 13 Jan 2005 16:18:03 -0600 From: Kilgallen@SpamCop.net (Larry Kilgallen) Subject: Re: Q: How to have SEARCH match a "null" line Message-ID: In article , Ken Fairfield writes: > Ken Fairfield wrote: > >> Graham Burley wrote: > [...] >> >> This is helpful, Graham, thanks! >> >> I'm already doing a SORT using a /Specification file, with defined >> fields and conditions, etc., and doing a Difference on the original >> file with the sorted file. So adding the /Omit's and/or /Data's >> in producing the sorted file will be easy and could be the answer >> to my problem. :-) > > One last follow-up on my original problem. > > To be specific, I needed to write a "monitor" that checks a system's > SYSUAF for duplicate UICs, by which I mean, more than one Username I suppose licensing a product http://ljk.com/ljk/LJK_SECURITY_DOCUMENTATION/LJK_SECURITY_D_019.HTML#tests_uaf_uicshare that also checks 211 other things, would be overkill :-) > shares a single UIC. In addition, there are certain known "non-user" > usernames which do share UICs, e.g., SYSTEM and SYS15, which I want > to skip. Ah, those exceptions keep you from simply setting the "UIC" RMS key to Noduplicates. ------------------------------ Date: Fri, 14 Jan 2005 00:33:48 +0000 (UTC) From: Graham Burley Subject: Re: Q: How to have SEARCH match a "null" line Message-ID: <41E7135C.6C738EDA@encompasserve-or-this.org> David J Dachtera wrote: > > Yeah! > > I used /SPECIFICATION for the "first" time recently at work: I needed to > extract stuff from a Classic Radnet system control file into a format > that DCL could use (V7.3, still has the "short strings" limitation). > Great stuff, but a bit obscure. Opportunity for improved documentation. It is a wonderful thing! One feature that probably is obscure is that SORT can be convinced *not* to sort. This is useful when you want to use the field processing without re-ordering the records. Give it a 0 length key field and /STABLE and records are output in their original order. For example, using SORT to copy a file: $ sort/spec=sys$input a.a b.b /field=(name=none, pos:1, siz:0) /key=none /stable ------------------------------ Date: Thu, 13 Jan 2005 22:40:26 +0000 (UTC) From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) Subject: Re: TCPIP SSH failure logging Message-ID: In article <3f119ada0501131051789c9894@mail.gmail.com>, DeanW writes: >TCPIP 5.4: >I'm getting an increasing number of hits on SSH, attempting to crack >my systems. Other than the immediate intrusions records, the IP >addresses these come from don't seem to be logged in the audit log. >Anyone know where I might find them? At least in the OPERATOR.LOG -- 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: Thu, 13 Jan 2005 21:13:06 -0600 From: Hunter Goatley Subject: Re: VTfm new release 2.3-2 Message-ID: <41E738C2.9080203@goatley.com> Vladimir Vershinin wrote: > VTfm new release 2.3-2 has the following new features: > > - Supports EXTENDED FILE SPECIFICATIONS on ODS-5 volumes; > - command line is up to 4K chars; > - supports use of F1-F10 keys on some keyboards instead of default > F11-F20 keys; > - User's Guide in PDF format included in the kit. > > You can download VTfm kit from Web-page: http://www.vtfm.narod.ru > Vladimir Vershinin, > vershinin-vk(AT)tochka.ru > I've added this to my archive, too. http://www.process.com/openvms/ ftp://ftp.process.com/vms-freeware/fileserv/vtfm.zip http://vms.process.com/ftp/vms-freeware/fileserv/vtfm.zip ftp://ftp.tmk.com/vms-freeware/fileserv/vtfm.zip http://www.tmk.com/ftp/vms-freeware/fileserv/vtfm.zip Hunter ------ Hunter Goatley, Process Software, http://www.process.com/ http://www.goatley.com/hunter/ ------------------------------ End of INFO-VAX 2005.027 ************************