INFO-VAX Wed, 05 Mar 2008 Volume 2008 : Issue 130 Contents: Re: Can a MicroVAX 3100 run on 208V? Re: DCPS downloadable? Re: DCPS downloadable? Re: DCPS downloadable? Re: DCPS downloadable? Re: Encryption on VMS Re: Encryption on VMS Re: Encryption on VMS Re: Encryption on VMS Re: Long term archiving of VMS stuff Re: Long term archiving of VMS stuff monitor incompatibility - progress Re: Noob File-Transfer (License to .com) Question Re: Noob File-Transfer (License to .com) Question OT: now CA schools want to indoctrinate your kids with communism OT: Scientists meet in NY to show global warming natural Re: OT: Universal healthcare in England failing - boy dies ! Re: SCSI bus Termination with KZPBA-CB controllers Re: SCSI bus Termination with KZPBA-CB controllers Re: SCSI bus Termination with KZPBA-CB controllers Simple Clustering question Re: Simple Clustering question Re: Simple Clustering question Re: Simple Clustering question Re: Simple Clustering question Re: VMS advertising ! Re: VMS advertising ! Re: Weekly Boot Camp Update Re: Weekly Boot Camp Update ---------------------------------------------------------------------- Date: 5 Mar 2008 08:02:12 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Can a MicroVAX 3100 run on 208V? Message-ID: In article <67uos3ppf89ah94s8mcp2fli5tk7928bft@4ax.com>, gerry77@no.spam.mail.com writes: > > BTW, here (Italy) we have 220 single-phase and 380 three-phase power. In > residences we have only 220 volts single-phase power which is used for > everything ranging from TV and PCs to washing machines and electric ovens. > It's my understanding that in the US many people have 120 volts for lightning > and 208 volts for heavy use such as ovens and dryers, am I right? We have 110 for lighting and small appliances, 220 for ovens, dryers, and such. Had an interesting problem when we received a collection of PCs that had been temporarily located in Europe. None would power up. Eventually someone saw that they were all set to 220V. ------------------------------ Date: 5 Mar 2008 08:59:16 +0100 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: Re: DCPS downloadable? Message-ID: <47ce60e4$1@news.langstoeger.at> In article <47cdc181$0$25417$c3e8da3@news.astraweb.com>, JF Mezei writes: >If Someone has the 2.6 DCPS kit and willing to send me a copy privately, >I would be more than happy to very quietly let others have it below the >radar and under the table. (ok, I'd have to direct FTP requests to the >alpha that is under a table for this to be true :-) You know already know where you can find it. ftp://ftp.itrc.hp.com/product_patches/hpdecprintsupervisor/dcps/openvms/ because ECO3 is current, and the ECO kit is in fact a full kit... -- 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: Wed, 05 Mar 2008 08:10:03 GMT From: =?ISO-8859-1?Q?Jan-Erik_S=F6derholm?= Subject: Re: DCPS downloadable? Message-ID: Peter 'EPLAN' LANGSTOeGER wrote: > In article <47cdc181$0$25417$c3e8da3@news.astraweb.com>, JF Mezei writes: >> If Someone has the 2.6 DCPS kit and willing to send me a copy privately, >> I would be more than happy to very quietly let others have it below the >> radar and under the table. (ok, I'd have to direct FTP requests to the >> alpha that is under a table for this to be true :-) > > You know already know where you can find it. > > ftp://ftp.itrc.hp.com/product_patches/hpdecprintsupervisor/dcps/openvms/ > because ECO3 is current, and the ECO kit is in fact a full kit... > The full ZIP'ed 2.6 DCPS kit I have is aprox 2.3 Mbytes. The ZIP'ed ECO3 I have is aprox 700 Kbytes. I do not think that the ECO3 kit on that link is a complete 2.6 kit. Jan-Erik. ------------------------------ Date: Wed, 05 Mar 2008 04:08:50 -0500 From: JF Mezei Subject: Re: DCPS downloadable? Message-ID: <47ce6392$0$1515$c3e8da3@news.astraweb.com> A DCPS 2.6 kit has inexpicably found its way to one of my hard drives. It appears to have come from the land of people who jump out of saunas naked to roll in snow. If you wish to take a look at it totally under the table, send me a private email. (remember to remove the smapnot from the email address) ------------------------------ Date: Wed, 05 Mar 2008 08:54:25 -0500 From: Paul Anderson Subject: Re: DCPS downloadable? Message-ID: In article <47ce60e4$1@news.langstoeger.at>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) wrote: > because ECO3 is current, and the ECO kit is in fact a full kit... Although DCPS ECOs are cumulative, in that they contain all files from previous ECOs to that version, they are not full kits. Paul -- Paul Anderson OpenVMS Engineering Hewlett-Packard Company ------------------------------ Date: Wed, 5 Mar 2008 08:52:51 -0800 (PST) From: res0o7il@yahoo.com Subject: Re: Encryption on VMS Message-ID: I think I need to clarify some stuff. I used the only copy of PGP for VMS I could find, one dating back to 1996. It ran for about 40 minutes to encrypt the save set, with no errors. However, when I decrypted I only got a tenth of the save set back. There were no errors reported. I used the GPG provided by the VMS Security team. It took eleven hours to encrypt the save set, which is an image backup of a large RAID disk. I can't get to the machine to check the size right now (VPN problems), but it's several gig. My machine is on VMS 6.2, but the customer's machine is on 7.3-2. They have the encryption subsystem, we don't. I do have a couple of machines I could load the later version of VMS onto, but I'd rather not unless I know it's going to be a productive operation. Larry, you mention that there's an encryption product that integrates with VMS backup. I don't seem to have that on VMS 6.2 Alpha. Would that be on the VMS distribution CDs somewhere, or does the Alpha version not turn up until 8.3? Shane ##### #-O-O-# # L # #===# ### ------------------------------ Date: Wed, 5 Mar 2008 08:57:32 -0800 (PST) From: res0o7il@yahoo.com Subject: Re: Encryption on VMS Message-ID: <23bf2d45-a5e7-4829-b277-988a662568cb@u10g2000prn.googlegroups.com> Just managed to get into the target machine. The save set I need to encrypt is 74,525,913 blocks. Shane ##### #-O-O-# # L # #===# ### ------------------------------ Date: Wed, 5 Mar 2008 09:17:41 -0800 (PST) From: AEF Subject: Re: Encryption on VMS Message-ID: <0afa43e4-2491-49e3-bdfd-4cc322a35412@n77g2000hse.googlegroups.com> On Mar 5, 12:57 pm, res0o...@yahoo.com wrote: > Just managed to get into the target machine. The save set I need to > encrypt is 74,525,913 blocks. > > Shane > > ##### > #-O-O-# > # L # > #===# > ### Are you encrypting directly to tape? If you are creating an on-disk file, have you tried SET FILE/RMS_EXTEND=/BLOCK=127? AEF ------------------------------ Date: Wed, 5 Mar 2008 12:04:45 -0600 (CST) From: sms@antinode.org (Steven M. Schweda) Subject: Re: Encryption on VMS Message-ID: <08030512044500_20203A01@antinode.org> From: res0o7il@yahoo.com > I think I need to clarify some stuff. Yup. > I used the only copy of PGP for VMS I could find, one dating back to > 1996. It ran for about 40 minutes to encrypt the save set, with no > errors. However, when I decrypted I only got a tenth of the save set > back. There were no errors reported. Well, yeah. It doubtless pre-dates large files, which I assume you're trying to use. > [...] I can't get to the machine to check the size right now (VPN > problems), but it's several gig. > [...] The save set I need to > encrypt is 74,525,913 blocks. As I said, large. From: AEF > Are you encrypting directly to tape? If you are creating an on-disk > file, have you tried SET FILE/RMS_EXTEND=/BLOCK=127? Probably a good idea, even with my GnuPG kit. I didn't try to add my usual RMS default parameter adjustment code to GnuPG, but I probably should see how hard that would be. As usual, a little investigation might reveal where the bottleneck lies. If the thing is CPU-bound, then even a big I/O improvement might be insignificant. ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: Wed, 05 Mar 2008 00:29:23 -0700 From: Mark Berryman Subject: Re: Long term archiving of VMS stuff Message-ID: <47cddb54$1@mvb.saic.com> VAXman- @SendSpamHere.ORG wrote: > In article <47c9ce15$0$10289$c3e8da3@news.astraweb.com>, JF Mezei writes: >> Main, Kerry wrote: >> >>> A company selling that many copies per month also does not decide to slash >>> the cost of the product world wide in order to stimulate demand. >> There are lots of issues with regards to Vista. Microsoft decided to >> have a very stringent hardware requiremenent for it. Without it, your >> machine is essentially an XP machine because the Vista glitz is disabled. >> >> Because of this, very few people are bothering with an upgrade of their >> existing machine. They will get Vista whenever they upgrade their PC. >> >> As such, Microsoft isn't seeing much demand for upgrades. And its >> revenue base is from the Winodws tax charged to wintel makers for just >> about every wintel box built. This is assured revenu no matter how >> good/bad windows is, it is simply relative to the health of the PC industry. >> >> Note that the recent price cut announcement only deals with *certain* >> versions of the shrinkwrapped Windows product. It does not cover the >> price charged by Microsoft to wintel manufacturer. >> >> While many rejoice at any bad news related to Microsoft, those news >> cover only a small portion of the Windows revenus since nobody has >> mentioned bad news with regards to the window tax on newbuilt Witel boxes. > > This is one of the reasons that I'm reluctant to buy new intel based Macs. > Am I or am I not paying Billzebub tax because a new Mac can run Weendoze? > Nobody in the Apple camp has been able to clarify this. > The tax in question is a tactic used by Microsoft where they essentially tell the PC vendor that they must sell a Windows license for very PC they sell or Microsoft will not grant them the license discount rate they need in order to be able to sell Windows systems competitively. Since Apple is not a Windows software reseller, they are not subject to this particular strong-arm tactic and are able to sell Intel product without paying a tax to Redmond. (No, Office for Mac doesn't count - the terms for that particular deal were spelled out when Apple settled its lawsuit with Microsoft.) Mark Berryman ------------------------------ Date: 5 Mar 2008 08:06:58 -0600 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: Long term archiving of VMS stuff Message-ID: <2m$kG+jbmq1Z@eisner.encompasserve.org> In article , John Santos writes: > > Good theory, but wrong. The fallacy of the excluded middle. I bet > you can boot Linux or BSD or VMS-x86 on it, without paying any MAC > or M$ tax (just the Intel tax.) :-) > Yeah, I'd like tosee you boot VMS-86 on it! 8-) ------------------------------ Date: Wed, 5 Mar 2008 03:32:18 -0800 (PST) From: IanMiller Subject: monitor incompatibility - progress Message-ID: <5f7c43b8-0b70-4bfb-9fb3-689cc22a76b0@2g2000hsn.googlegroups.com> Some people in this place will be interested to know that installing the recently released VMS732_MONTOR-V0100 patch kit means that in a cluster with VMS 7.3-2 and V8.3 then MONITOR CLUSTER now works. ftp://ftp.itrc.hp.com/openvms_patches/alpha/V7.3-2/VMS732_MONTOR-V0100.txt ftp://ftp.itrc.hp.com/openvms_patches/alpha/V7.3-2/VMS732_MONTOR-V0100.ZIPEXE If you have VAX V7.3 systems clustered with VMS 8.3 then MONITOR CLUSTER still does not work for the VAXes. ------------------------------ Date: Tue, 04 Mar 2008 21:00:20 +0100 From: Ralf Folkerts Subject: Re: Noob File-Transfer (License to .com) Question Message-ID: Jan-Erik Söderholm schrieb: Hi Jan-Erik, >>> Just "$ CREATE lic.com " and paste the mail. >> >> thanks for the hint! Well., I tried that too, using minicom on a Host >> next to the AlphaStation via SSH from my Laptop. However, I was too >> greedy and wanted to paste the hole License Document into the Editor. > > Note, CREATE is not an "editor". CREATE just copies whatever is > written to the terminal (from the paste) to a file. thanks for pointing me at that! I indeed did not recognize the Capitals you used and assumed you meant to create the file in an editor! Hope I'll keep that Command in my mind just in case! Cheers, _ralf_ ------------------------------ Date: Tue, 04 Mar 2008 21:00:15 +0100 From: Ralf Folkerts Subject: Re: Noob File-Transfer (License to .com) Question Message-ID: sampsal@gmail.com schrieb: Hi sampsal and Bob Koehler, [...] >> I then pop the CD into the VMS box and mount it with the /OVERRIDE=ID, /MEDIA=CDROM >> and /UNDEF=(STREAM_LF:132) options and run the .COM files. >> >> So briefly: > > Oops, hit Send by accident, so the commands to issue are similar to > the following: > (This assumes there are two files on the CD, BASE.COM and LAYER.COM, > and DKAxxx is your CD device) > > MOUNT DKAxxx: /OVERRIDE=ID /MEDIA=CDROM /UNDEF=(STREAM_LF:132) > @DKAxxx:BASE.COM > @DKAxxx:LAYER.COM a BIG thumbs-up! I just re-installed OpenVMS to my Alpha again - and Installation of the Licences went thru like a hot knife through butter! Really great hint that I hope to keep in mind! Cheers, _ralf_ ------------------------------ Date: Wed, 5 Mar 2008 06:54:49 -0800 (PST) From: ultradwc@gmail.com Subject: OT: now CA schools want to indoctrinate your kids with communism Message-ID: http://www.worldnetdaily.com/index.php?fa=PAGE.view&pageId=58061 ------------------------------ Date: Wed, 5 Mar 2008 06:50:57 -0800 (PST) From: ultradwc@gmail.com Subject: OT: Scientists meet in NY to show global warming natural Message-ID: <3812c263-5543-44eb-aa38-100ac1fa0a93@p73g2000hsd.googlegroups.com> http://www.worldnetdaily.com/index.php?fa=PAGE.view&pageId=58024 ------------------------------ Date: Wed, 5 Mar 2008 06:42:31 -0800 (PST) From: ultradwc@gmail.com Subject: Re: OT: Universal healthcare in England failing - boy dies ! Message-ID: On Feb 28, 9:37=A0pm, David J Dachtera wrote: > ultra...@gmail.com wrote: > > > here is a glance of things to come if Hillary or Obama implement > > their healthcare plan ... the bbc did not report it but a teen died > > while waiting on an ambulance that was setting at the hospital > > with a patient inside because the hospitals were full and did not > > want to violate a government mandate of a 4 hour must treat law. > > > Canada people coming here ... England people that have the money > > flying to India to have surgery ... > > > don't worry, those rich liberal Harvard grads will be able to afford > > it while those they claim to want to help will be dying waiting for > > an ambulance ... > > > that what socialism does, it ruins a country ... > > >http://news.bbc.co.uk/1/hi/uk/7249514.stm > > (Relax, folks, Boob is just trying to up the post tallies for this year > to stem the downward spiral of VMS. Maybe the "MI5" guy will help him > out.)- Hide quoted text - > > - Show quoted text - Hillary won ... her socialist healthcare system and these kind of results like above are getting closer to becoming reality here ... ------------------------------ Date: Wed, 5 Mar 2008 10:30:05 -0500 From: "David Turner, Island Computers" Subject: Re: SCSI bus Termination with KZPBA-CB controllers Message-ID: Doesn't that kinda negate the point of a cluster? DT "Chris Scheers" wrote in message news:636k25F26p6guU1@mid.individual.net... > Rich Jordan wrote: >> We're retrofitting some differential (HVD) SCSI gear into a two node >> SCSI cluster using KZPBA-CB controllers. The old setup used the 'Y' >> cable connectors for the Alpha hookups, one leg going to the storage >> equipment, the other terminated. The Alphas are at each end of the >> bus with storage in between. >> >> The back side of this config will have a difficult time fitting with >> those bulky connectors. Does anyone know if we can just hook the >> storage cables directly to the KZPBA-CB controllers, with the console >> setting for termination on those cards enabled? Does this provide >> proper bus termination? I know we will lose the ability to physically >> remove a node from the bus without disrupting the bus if the 'Y' >> cables are removed. > > If you enable termination on the cards, yes. > > I don't know if the termination is still valid if the power to the box is > off. > > Remember that the point of the "Y" cables is to enable you to disconnect > one system or the other for maintenance without disrupting the SCSI bus. > > If you eliminate the "Y" cables and have to disconnect a machine, you will > need to take the other one down, too. > > -- > ----------------------------------------------------------------------- > Chris Scheers, Applied Synergy, Inc. > > Voice: 817-237-3360 Internet: chris@applied-synergy.com > Fax: 817-237-3074 ------------------------------ Date: Wed, 5 Mar 2008 08:56:54 -0800 (PST) From: Rich Jordan Subject: Re: SCSI bus Termination with KZPBA-CB controllers Message-ID: On Mar 5, 9:30 am, "David Turner, Island Computers" wrote: > Doesn't that kinda negate the point of a cluster? > > DT"Chris Scheers" wrote in message > > news:636k25F26p6guU1@mid.individual.net... > > > Rich Jordan wrote: > >> We're retrofitting some differential (HVD) SCSI gear into a two node > >> SCSI cluster using KZPBA-CB controllers. The old setup used the 'Y' > >> cable connectors for the Alpha hookups, one leg going to the storage > >> equipment, the other terminated. The Alphas are at each end of the > >> bus with storage in between. > > >> The back side of this config will have a difficult time fitting with > >> those bulky connectors. Does anyone know if we can just hook the > >> storage cables directly to the KZPBA-CB controllers, with the console > >> setting for termination on those cards enabled? Does this provide > >> proper bus termination? I know we will lose the ability to physically > >> remove a node from the bus without disrupting the bus if the 'Y' > >> cables are removed. > > > If you enable termination on the cards, yes. > > > I don't know if the termination is still valid if the power to the box is > > off. > > > Remember that the point of the "Y" cables is to enable you to disconnect > > one system or the other for maintenance without disrupting the SCSI bus. > > > If you eliminate the "Y" cables and have to disconnect a machine, you will > > need to take the other one down, too. > > > -- > > ----------------------------------------------------------------------- > > Chris Scheers, Applied Synergy, Inc. > > > Voice: 817-237-3360 Internet: ch...@applied-synergy.com > > Fax: 817-237-3074 Yep. It does negate part of the benefit of a cluster. IN defense I'll state that the previous Alphaservers (AS1000As) never had to be removed from the bus in their 10 years of operation; occasionally one was shut down to replace other hardware but the bus stayed connected. I guess the power down would have posed a problem for the termination at that end if not for the 'Y' cables, though. The space will not be an issue after all; they can move the shelves out from the wall enough to accommodate the Y cables. And these are relatively thick cables that won't take a tight bend, so it was an issue. I checked the cards in the system; the resistor packs are present on the cards. I tried running the external towers direct to the cards (without the 'Y's and all seemed good at low load. When a shadow copy was run across the towers, the bus errors went through the roof. So its not a workable solution. Both Alphas were running, all cards have terminator packs, so the bus should have been terminated. I'm switching back to the 'Y' cables and pulling the terminator packs to do the next shadow copy. We'll see how it goes. Do the PK*0_SOFT_TERM console variable have any effect at all on KZPBA- CB cards? Do they need to be set to any particular value, given that actual termination is handled by the presence or absence of the terminator packs? Rich ------------------------------ Date: Wed, 5 Mar 2008 09:57:53 -0800 (PST) From: BHall Subject: Re: SCSI bus Termination with KZPBA-CB controllers Message-ID: On Mar 5, 10:56 am, Rich Jordan wrote: > On Mar 5, 9:30 am, "David Turner, Island Computers" > > > > wrote: > > Doesn't that kinda negate the point of a cluster? > > > DT"Chris Scheers" wrote in message > > >news:636k25F26p6guU1@mid.individual.net... > > > > Rich Jordan wrote: > > >> We're retrofitting some differential (HVD) SCSI gear into a two node > > >> SCSI cluster using KZPBA-CB controllers. The old setup used the 'Y' > > >> cable connectors for the Alpha hookups, one leg going to the storage > > >> equipment, the other terminated. The Alphas are at each end of the > > >> bus with storage in between. > > > >> The back side of this config will have a difficult time fitting with > > >> those bulky connectors. Does anyone know if we can just hook the > > >> storage cables directly to the KZPBA-CB controllers, with the console > > >> setting for termination on those cards enabled? Does this provide > > >> proper bus termination? I know we will lose the ability to physically > > >> remove a node from the bus without disrupting the bus if the 'Y' > > >> cables are removed. > > > > If you enable termination on the cards, yes. > > > > I don't know if the termination is still valid if the power to the box is > > > off. > > > > Remember that the point of the "Y" cables is to enable you to disconnect > > > one system or the other for maintenance without disrupting the SCSI bus. > > > > If you eliminate the "Y" cables and have to disconnect a machine, you will > > > need to take the other one down, too. > > > > -- > > > ----------------------------------------------------------------------- > > > Chris Scheers, Applied Synergy, Inc. > > > > Voice: 817-237-3360 Internet: ch...@applied-synergy.com > > > Fax: 817-237-3074 > > Yep. It does negate part of the benefit of a cluster. IN defense > I'll state that the previous Alphaservers (AS1000As) never had to be > removed from the bus in their 10 years of operation; occasionally one > was shut down to replace other hardware but the bus stayed connected. > I guess the power down would have posed a problem for the termination > at that end if not for the 'Y' cables, though. > > The space will not be an issue after all; they can move the shelves > out from the wall enough to accommodate the Y cables. And these are > relatively thick cables that won't take a tight bend, so it was an > issue. > > I checked the cards in the system; the resistor packs are present on > the cards. I tried running the external towers direct to the cards > (without the 'Y's and all seemed good at low load. When a shadow copy > was run across the towers, the bus errors went through the roof. So > its not a workable solution. Both Alphas were running, all cards have > terminator packs, so the bus should have been terminated. > > I'm switching back to the 'Y' cables and pulling the terminator packs > to do the next shadow copy. We'll see how it goes. > > Do the PK*0_SOFT_TERM console variable have any effect at all on KZPBA- > CB cards? Do they need to be set to any particular value, given that > actual termination is handled by the presence or absence of the > terminator packs? > > Rich Rich, You might have a bad resistor after knocking around in storage for a while or maybe you missed a pin/socket when reinserting them? If you have a VOM you might check them and verify that they all have consist readings. I don't think the srm can tell which version of the KZPBA is installed and therefore it creates the variables whether they are useful or not. I think "show config" identifies both -CA and -CB as "QLogic ISP10x0". Our PK*0_SOFT_TERM variables are set to ON and I'm sure that is the default value. We are using or did use either the DWZZH-03 or DWZZH-05 hubs to eliminate the y-cables/external terminators and still allow removing hosts or arrays from the shared scsi buses. Bill ------------------------------ Date: Wed, 5 Mar 2008 12:10:53 -0500 From: "David Turner, Island Computers" Subject: Simple Clustering question Message-ID: <1vAzj.34$DT3.15@bignews3.bellsouth.net> Can my customer with 2 ES45 cluster with just 1 gigabit ethernet, cluster license and y cables? Does he actually need memory channel? Never been asked this before.... I figured Y cable with term. ----cable-----system ---y cable ----system ----- HSZ80-----term Or am I missing something? David -- Island Computers US Corp PO Box 86 Tybee Island GA 31328 T: 877 636 4332 x201 M: 877 636 4332 x251 I: 00 1 912 786 8501 F: 912 786 8505 E: dturner@islandco.com W: www.islandco.com ------------------------------ Date: Wed, 05 Mar 2008 17:19:44 GMT From: "PL" Subject: Re: Simple Clustering question Message-ID: Yes, he should be able to use the gigabit ethernet to deliver cluster communication between the nodes, unless networking equipment does not support SCS/SCA packets or something else gets in the way. ^P "David Turner, Island Computers" skrev i meddelandet news:1vAzj.34$DT3.15@bignews3.bellsouth.net... > Can my customer with 2 ES45 cluster with just > 1 gigabit ethernet, cluster license and y cables? > > Does he actually need memory channel? > > Never been asked this before.... > > I figured > > Y cable with term. ----cable-----system ---y cable ----system ----- > HSZ80-----term > > Or am I missing something? > > David > > -- > Island Computers US Corp > PO Box 86 > Tybee Island GA 31328 > > T: 877 636 4332 x201 > M: 877 636 4332 x251 > I: 00 1 912 786 8501 > F: 912 786 8505 > E: dturner@islandco.com > W: www.islandco.com > > ------------------------------ Date: Wed, 5 Mar 2008 09:39:01 -0800 (PST) From: Ken.Fairfield@gmail.com Subject: Re: Simple Clustering question Message-ID: On Mar 5, 9:10 am, "David Turner, Island Computers" wrote: > Can my customer with 2 ES45 cluster with just > 1 gigabit ethernet, cluster license and y cables? > > Does he actually need memory channel? > > Never been asked this before.... > > I figured > > Y cable with term. ----cable-----system ---y cable ----system ----- > HSZ80-----term > > Or am I missing something? As PL said, yes you can. I'd suggest using two gigabit NICs in each ES45 with "cross-over" connections (i.e., direct connections) between the two systems and avoid having a network switch in the mix for the cluster traffic. They can connect to their larger network with a 3rd NIC. At my former employer, we used the scheme quite successfully with both 2-node and 3-node ES40 clusters (the 3-node clusters used more NICs to get full redundancy between each pair of nodes). Oh, I should qualify the above by stating we were using 100 mbit NICs, not 1000 mbit, but that shouldn't be an issue... I can't comment on the SCSI connections to the storage...is a SAN out of the question (HSG, Brocade, KGPSA)? -Ken ------------------------------ Date: Wed, 5 Mar 2008 09:59:49 -0800 (PST) From: Volker Halle Subject: Re: Simple Clustering question Message-ID: <7df83e31-3e1b-437d-9cec-c817c95ccd25@s8g2000prg.googlegroups.com> David, you would want to use more than 1 Gigabit Interface as cluster interconnects. Just for redundancy. SCSI cluster configurations are explained in chapter 6 of the Guidelines for OpenVMS Cluster Configurations manual. Volker. ------------------------------ Date: Wed, 5 Mar 2008 13:10:27 -0500 From: "David Turner, Island Computers" Subject: Re: Simple Clustering question Message-ID: Thanks Volker et al. David "Volker Halle" wrote in message news:7df83e31-3e1b-437d-9cec-c817c95ccd25@s8g2000prg.googlegroups.com... > David, > > you would want to use more than 1 Gigabit Interface as cluster > interconnects. Just for redundancy. > > SCSI cluster configurations are explained in chapter 6 of the > Guidelines for OpenVMS Cluster Configurations manual. > > Volker. ------------------------------ Date: 5 Mar 2008 12:52:16 GMT From: billg999@cs.uofs.edu (Bill Gunshannon) Subject: Re: VMS advertising ! Message-ID: <637jc0F25o693U1@mid.individual.net> In article <47cdfb85$0$31313$c3e8da3@news.astraweb.com>, JF Mezei writes: > VAXman- @SendSpamHere.ORG wrote: > >> Does VMS use VMS to process these violations? > > If there was a VMS management eager to grow sales instead of avoid the > next round of HP layoffs, they could donate VMS machines/licences to > this VMS company. Then, with this company running our VMS, every bit of > advertising they would make would benefit our VMS. And this would > totally bypass the advertising ban that seems to be in place at HP > against our VMS. Reality check!!! You couldn't give a VMS system away (well, except maybe to a hobbyist) today. Remember me? I provided VMS systems here, at no cost to the University or the Department. I provided and maintained the hardware. I installed and maintained all the software. I administered the machines. And before you say, "Well they were VAXen and the performance sucked.", you would be wrong. No one ever complained about the performance and at one time it included Alphas as well. They went away because no one wanted to use VMS. Period. End of story. If you offered to give the "VMS" operation a bunch of machines running OpenVMS I would imagine they would be: a) turned down b) never used or, most likely, c) converted to Linux machines. bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves billg999@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: Wed, 5 Mar 2008 05:15:41 -0800 (PST) From: AEF Subject: Re: VMS advertising ! Message-ID: <5fe4c62d-26a6-4007-bfba-de30747e7c4f@8g2000hse.googlegroups.com> On Mar 5, 8:52 am, billg...@cs.uofs.edu (Bill Gunshannon) wrote: > In article <47cdfb85$0$31313$c3e8...@news.astraweb.com>, > JF Mezei writes: > > > VAXman- @SendSpamHere.ORG wrote: > > >> Does VMS use VMS to process these violations? > > > If there was a VMS management eager to grow sales instead of avoid the > > next round of HP layoffs, they could donate VMS machines/licences to > > this VMS company. Then, with this company running our VMS, every bit of > > advertising they would make would benefit our VMS. And this would > > totally bypass the advertising ban that seems to be in place at HP > > against our VMS. > > Reality check!!! You couldn't give a VMS system away (well, except > maybe to a hobbyist) today. Remember me? I provided VMS systems > here, at no cost to the University or the Department. I provided > and maintained the hardware. I installed and maintained all the > software. I administered the machines. And before you say, "Well > they were VAXen and the performance sucked.", you would be wrong. > No one ever complained about the performance and at one time it > included Alphas as well. They went away because no one wanted to > use VMS. Period. End of story. If you offered to give the "VMS" > operation a bunch of machines running OpenVMS I would imagine they > would be: > a) turned down > b) never used > or, most likely, > c) converted to Linux machines. > > bill > > -- > Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves > billg...@cs.scranton.edu | and a sheep voting on what's for dinner. > University of Scranton | > Scranton, Pennsylvania | #include Warning: Rant mode: Yeah, no one seems to care about performance these days. We have Outlook/Exchange. We're running JIRA here and it's slow and keeps logging me out most of the time I don't touch it for more than a few minutes, but often I could go on vacation and come back and I'm still logged in! The JIRA admin says it's something I'm doing or it's my cookies or settings. Baloney. Something I'm doing? Yeah, I'm using the bloody product -- that's what I'm doing. In general, browser downloads always seem to be slow. Web page design sucks on average. People seem to like to move the mouse to various things on the page. Just put everything on the page where I can see it so I don't waste time and increase RSI risk doing stupid crap like that. OK, one exception is (train) ticket vending machines. People really do want them to go fast. But sitting at their desks at work or home, no one seems to care to be wasting half their day waiting for browser downloads or moving the mouse to every square centimeter of the page looking for some stupid hidden links. Also, typography on Web pages sucks on average. Drop down lists sometimes take seconds to appear. It's a fricking list of characters. Menus fade in. Please, I've got better things to do than wait for menus to fade in. I always turn that crap off when I can. We've got the fastest chips ever made and still everything is slow. Maybe they should make VMS performance suck, too. Then everyone will flock to VMS. I could go on, but... Rant off. AEF ------------------------------ Date: Wed, 05 Mar 2008 10:48:59 +0000 From: "R.A.Omond" Subject: Re: Weekly Boot Camp Update Message-ID: Sue wrote: > > [..snip...] > > Countries (no particular order) > > UK - Ireland 1 Oh dear. Sue, Ireland is *not* part of the UK. ------------------------------ Date: 05 Mar 2008 13:46:28 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Weekly Boot Camp Update Message-ID: <47cea434$0$5616$607ed4bc@cv.net> In article , "R.A.Omond" writes: >Sue wrote: >> >> [..snip...] >> >> Countries (no particular order) >> >> UK - Ireland 1 > >Oh dear. Sue, Ireland is *not* part of the UK. Perhaps Sue meant the Ununited Kingdom! From Pink Floyd's Us and Them... With, without. And who'll deny it's what the fighting's all about. Seems there was quite a bit of nastiness in England because of this. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" http://tmesis.com/drat.html ------------------------------ End of INFO-VAX 2008.130 ************************