INFO-VAX Tue, 31 Jul 2007 Volume 2007 : Issue 416 Contents: DEBUG-I-DSFMISMATCH on librtl Re: DEBUG-I-DSFMISMATCH on librtl Re: Installing VMS without a bootable CD-ROM drive Is VMSONE.COM off the air permanently? Re: Is VMSONE.COM off the air permanently? Re: July the 4th Re: July the 4th Re: July the 4th RE: July the 4th Re: July the 4th Maximum Java Heap Size/OpenVMS Java Experiences? Re: Maximum Java Heap Size/OpenVMS Java Experiences? Re: Old DEC Monitor quandry Re: Old DEC Monitor quandry remote shutdown/startup of MSA1000? TRADE: BA370 rack mount VMS cluster behind a *NIX firewall Re: VMS cluster behind a *NIX firewall Re: VMS cluster behind a *NIX firewall Re: VMS cluster behind a *NIX firewall RE: VMS cluster behind a *NIX firewall Re: VMS cluster behind a *NIX firewall Re: VMS cluster behind a *NIX firewall Re: VMS cluster behind a *NIX firewall Re: VMS cluster behind a *NIX firewall RE: VMS cluster behind a *NIX firewall RE: VMS cluster behind a *NIX firewall Re: VMS cluster behind a *NIX firewall Re: VMS cluster behind a *NIX firewall Re: VMS cluster behind a *NIX firewall Re: VMS cluster behind a *NIX firewall Re: What does GEM mean? Re: [OT] RIP DTL's Father ---------------------------------------------------------------------- Date: Tue, 31 Jul 2007 08:44:21 -0700 From: Don Subject: DEBUG-I-DSFMISMATCH on librtl Message-ID: <1185896661.857528.174060@i13g2000prf.googlegroups.com> OpenVMS 8.3 with SYS-V0200 on rx3600 When we try to debug a particular program written in C, we get a series of errors beginning with DEBUG-I-DSFMISMATCH for LIBRTL.DSF / LIBRTL.EXE. The relevant directory entrys are: GREGG> dir librtl.*/date Directory SYS$COMMON:[SYSLIB] LIBRTL.AIIF;1 30-JUN-2006 13:06:53.87 LIBRTL.DSF;1 29-JUN-2006 23:17:38.03 LIBRTL.EXE;1 20-DEC-2006 15:06:27.10 LIBRTL.STB;1 20-DEC-2006 15:06:27.31 Total of 4 files. Any idea how we got here? How do I get matching dsf and exe files for librtl? Thanks for the help, Don ------------------------------ Date: Tue, 31 Jul 2007 08:51:03 -0700 From: Volker Halle Subject: Re: DEBUG-I-DSFMISMATCH on librtl Message-ID: <1185897063.966597.271240@57g2000hsv.googlegroups.com> Don, your LIBRTL.EXE and .STB (dated 20-DEC-2006) are from VMS83I_SYS- V0200, but HP failed to deliver an updated/matching LIBRTL.DSF file with that ECO kit. You may ignore this DEBUG-I-DSFMISMATCH unless you want to debug code inside LIBRTL. Volker. ------------------------------ Date: Tue, 31 Jul 2007 05:45:45 -0500 (CDT) From: sms@antinode.org (Steven M. Schweda) Subject: Re: Installing VMS without a bootable CD-ROM drive Message-ID: <07073105454533_202003EE@antinode.org> From: "P. Sture" > > For the record, VMS VAX installation save sets (VMS%%%.%) must be in > > the [000000] directory, or else the installation will fail. (VMS%%%.B > > may be an exception, when it's handled manually, outside the normal > > installation procedure, but it's normally found in the same place as the > > others, and they must be in [000000].) > > Is the [000000] restriction still true? Maybe I'm dreaming, but I > thought it had been relaxed at some point. According to the "OpenVMS VAX Version 7.3 Upgrade and Installation Manual", http://h71000.www7.hp.com/doc/73final/6630/6630pro_003.html The procedure prompts you to specify which drive holds the distribution kit. Enter the device name for the source drive. If all you can specify is a device name, where will it look for the files? In "Chapter 5 Before Upgrading the OpenVMS VAX Operating System", http://h71000.www7.hp.com/doc/73final/6630/6630pro_006.html it still says: 5.1.7 Save Set Location If you move the OpenVMS VAX save sets from the distribution media to the system disk or some other disk before beginning the upgrade, the save sets must be in the [000000] directory of the disk, or the upgrade will fail. [...] But if you can demonstrate that it works when they're somewhere else, I'll be happy to gain enlightenment. ------------------------------------------------------------------------ Steven M. Schweda sms@antinode-org 382 South Warwick Street (+1) 651-699-9818 Saint Paul MN 55105-2547 ------------------------------ Date: Tue, 31 Jul 2007 06:58:39 -0700 From: Rich Jordan Subject: Is VMSONE.COM off the air permanently? Message-ID: <1185890319.702659.248370@e16g2000pri.googlegroups.com> Is VMSONE.COM gone for good? For the past couple of weeks I've been unable to get to it (connections time out), and at least one post in June noted it was not reachable. ------------------------------ Date: Tue, 31 Jul 2007 14:17:45 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Is VMSONE.COM off the air permanently? Message-ID: In article <1185890319.702659.248370@e16g2000pri.googlegroups.com>, Rich Jordan writes: > > >Is VMSONE.COM gone for good? For the past couple of weeks I've been >unable to get to it (connections time out), and at least one post in >June noted it was not reachable. Who took over VMSone.com when John passed away? I would thing one of the DFW lug guys -- David Cathey or Pat Jankoviak. They'd be the best to answer your question. I would love to see VMSone.com back on-line too but at least Google has cached all of the _important_ content that was there. -- 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 ------------------------------ Date: Tue, 31 Jul 2007 02:07:30 -0500 From: Ron Johnson Subject: Re: July the 4th Message-ID: On 07/30/07 12:41, david20@alpha2.mdx.ac.uk wrote: [snip] > > Are you really that ignorant of the situation ? > Yasser Arafat and the PLO accepted Israels right to exist over a decade ago !! > This was formally confirmed when Arafat exchanged letters of PLO-Israel > recognition with the Yitzhak Rabin in 1993. > > See http://www.earlham.edu/~pss/yasser_arafat.htm More sayings of Chairman Arafat, *after* he "accepted" Israel's right to exist... Newspaper interview, Al-Quds, May 10, 1998: Question: Do you feel sometimes that you made a mistake in agreeing to Oslo? Arafat: No... no. Allah's messenger Mohammad accepted the al-Khudaibiya peace treaty and Salah a-Din accepted the peace agreement with Richard the Lion-Hearted." [Note: Salah a-Din (Saladin) was the Muslim leader who, after a cease-fire, declared a jihad against the Crusaders and captured Jerusalem] Published in the official PA newspaper Al-Hayat Al-Jadeeda, April 16, 1998: My colleagues in struggle and in arms, my colleagues in struggle and in jihad. Intensify the revolution and the blessed intifada. Reinforce the strong stance and strengthen the faith. We must burn the ground under the feet of the invaders. Speech to the Organization of the Islamic Conference's Jerusalem Committee in Morocco on July 29, 1998 (Al-Hayat Al-Jadeeda, July 30, 1998): with Arab and Islamic assistance, this shall be our faithful jihad (holy war) - to defend holy Jerusalem from the danger of Judaization and the Zionist plot." Leading Palestinian marchers in a chant, broadcast on official Palestinian television on May 14, 1998 With our soul and blood we will redeem you O Palestine! -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Tue, 31 Jul 2007 02:10:25 -0500 From: Ron Johnson Subject: Re: July the 4th Message-ID: On 07/30/07 02:15, Dirk Munk wrote: > JF Mezei wrote: >> AEF wrote: >>> It doesn't matter what the US does. Hamas, Hezbollah, etc., want to >>> destroy the state of Israel. THEY'VE PLAINLY AND LOUDLY SAID SO! >> >> They acceptance of Israel's existance will be the RESULT of succesful >> negotiations. The USA/Israel require them to accept Israel as a >> pre-condition for negotiations. > > The state of Israel exists since the UN more or less established it in > 1948. The existence of the state of Israel or acceptance of this fact is > not negotiable. > > Almost no one in the middle east is interested in solving the conflict > between Israel and the Palestinians. Why not? Simple, corrupt and > autocratic leaders love to use this conflict to distract their own > people from the problems in their own country. It is the oldest game in > politics, find something abroad to distract the attention of the people > from your own failures. And it works perfectly. Poverty in your own > country? Look at the Americans and Europeans! They help the Israeli > against the Palestinians! They hate Muslems! That is why we are poor! > > Who is paying the salaries of the Palestinian civil servants? Not the > rich Arab countries or other "Muslem brother states". No, the US and the > EU, those damned friends of Israel. A man after my own heart. Are you *sure* that you are European? Or were you stolen at birth from a Texas hospital and spirited across the pond? > Now Israel isn't doing very much to improve the situation either. If you > are a Palestinian or a Israeli citizen of Arab origin, and you can't get > water from the tap while Jewish people are splashing in their private > swimming pool just a few hundred yards from your home, you will not get > the idea that you get treated fairly by the Israeli government. And this > is just a small example of the rather foolish way Israel is fueling the > conflict. This is interesting and sad. Do you have a reference, so that I might get more detail? -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Tue, 31 Jul 2007 02:23:58 -0500 From: Ron Johnson Subject: Re: July the 4th Message-ID: On 07/29/07 15:26, David J Dachtera wrote: > Ron Johnson wrote: >> On 07/29/07 02:16, Paul Raulerson wrote: >> [snip] >>> Just what exactly do you find wrong with that picture, beyond the >>> fact you have exaggerated the analogy? Your own history is pretty >>> well fraught with fighting those exact same "Indians." The >>> Netherlands has fought against Muslims for quite a while, >>> including ganging up against them with Portugal in Malaysia. >> Three words: Theo van Gogh. >> >>> Muslims never bothered us much before they starting slamming >>> airplanes into buildings. We never bothered them until they did >>> that either. Also remember, that wasn't their first attempt- just >>> their first really successful one. >> Two more words: Achille Lauro. > > ...and as I implied at another point in all this, the fact that the flight crews > of those two planes LET them be flown into the WTC will stand before the entire > world as testimony to the U.S.'s greatest weakness. Boy do I agree with that!!!!! Since the 1970s, Americans[0] have been, in essence, told to bend over and accept being terrorized, raped, beaten, mugged, etc. The men of United 93 did a much greater service to this country than "just" preventing that plane from crashing into the WH or Capitol. [0] And possibly the Europeans too, but there's a different dynamic there, given that there haven't been pioneers in Europe since the neolithic era. -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Tue, 31 Jul 2007 15:48:11 +0000 (UTC) From: david20@alpha2.mdx.ac.uk Subject: RE: July the 4th Message-ID: In article <000a01c7d21f$b9132fd0$2b398f70$@com>, "Paul Raulerson" writes: > > >> -----Original Message----- >> From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca] >> Sent: Sunday, July 29, 2007 12:23 PM >> To: Info-VAX@Mvb.Saic.Com >> Subject: Re: July the 4th >> >> Ron Johnson wrote: >> > The whole "they'll like us when we stop supporting Israel" idea is a >> > steaming pile of shit. >> >> >> You wouldn't be able to grasp more evolved concepts such as "our blind >> support for Israel prevents establishement of a real peace in middle >> east". It is VERY different than what you stated above. >> > > >Here's something for you - Israel has as much, if not *more* right to exist >than do some, even most, of the surrounding Arab states. If by that you mean that most of the Arab countries in the Middle East were created by the great powers after the first world war - France and Britain - and that Israel was similarly created by the US (and UN) after the second world war. Then yes - the countries were arbitrarily created. However the people who lived in those areas (citizens of the ottaman empire) still lived in those areas after the Arab countries were created. In contrast Israel was created as a homeland for the Jewish people who had been spread throughout the world in the previous two millenia. The only justification for the existence of this new Israel in this particular location is that such a country once existed in ancient times and the idea from the Old Testament that God had promised them this land and ordered them to forcefully appropriate it from the lands previous inhabitants - the Canaanites. Applying similar historic claims to most other parts of the world would be complicated because there have been so many conquests and reconquests to consider. However I suppose you could say that the prime claim should go to the earliest inhabitants which could be found. In which case Australia should be handed over to the aborigines and North America should be handed over to the Indian tribes etc etc ( Of course this might also be taken as meaning that Israel should be handed back to the Lebanese descendants of the ancient Canaanites - see http://en.wikipedia.org/wiki/Canaanites#Genetics ) >And here's another >one for you, Israel is *directly* connected to the main religion of most of >the population of the United States, which like it or not, is Christianity. > Which is why I suppose that Christians have historically persecuted Jews. Islam is as directly connected to Judaism as Christianity and historically has generally been much more tolerant - probably because Muslims didn't consider the Jews to have killed their messiah. David Webb Security team leader CCSS Middlesex University > ------------------------------ Date: Tue, 31 Jul 2007 17:30:38 +0000 (UTC) From: david20@alpha2.mdx.ac.uk Subject: Re: July the 4th Message-ID: In article <1185761687.729407.239390@g4g2000hsf.googlegroups.com>, AEF writes: >On Jul 29, 8:58 pm, JF Mezei wrote: >> AEF wrote: >> > It doesn't matter what the US does. Hamas, Hezbollah, etc., want to >> > destroy the state of Israel. THEY'VE PLAINLY AND LOUDLY SAID SO! >> >> They acceptance of Israel's existance will be the RESULT of succesful >> negotiations. The USA/Israel require them to accept Israel as a >> pre-condition for negotiations. > >Ridiculous. How do you negotiate with someone who wants to destroy >you? I don't think Hamas or Hezbollah even wants to negotiate. Fatah >maybe. > Although Hamas doesn't officially recognise Israel they definitely want the establishment of a Palestinian state and have recognised the existence of Israel as a reality both now and after the formation of such a state see http://www.guardian.co.uk/israel/Story/0,,1802862,00.html and http://www.smh.com.au/news/world/israel-a-reality-hamas-leader-admits/2007/01/11/1168105115675.html The only way out of this conflict is for bitter enemies to negotiate in good faith. Who would have thought that Gerry Adam's Sinn Fein and Ian Paisley's DUP could be working together in running Northern Ireland a decade ago. David Webb Security team leader CCSS Middlesex University >> You can't require peace be achieved as a pre-condition to negotiate a >> peace agreement. > >Why not a cease fire? You're equating short-term peace with long-term >peace. They're not the same. > >Actually, didn't Rabin try that? Negotiate as if there were no >terrorists and fight the terrorists as if there were no negotations? >It didn't work, did it now? > >> >Israel withdrew UNILATERALLY and UNCONDITIONALLY from Gaza, and looked >> >what happened. The missiles are still coming. Civil war >> >> If Israel widthdrew from Gaza, why then does Israel still prevent free >> movement of people within Palestine, why does it widthhold tax revenus >> preventing the Gaza government (duly elected in a democratic fashion) >> from operating ? > >Because the P's are still firing rockets into Israel proper and still >staging attacks. That's why. > >Hitler arose through democracy as did Hamas. Bush was elected by a >democracy but that doesn't stop you from condemning him. Another clear >indication of your bigotry. OH, it's okay for Hamas, but not for >Bush!!! I think I see some bigotry here! Let's see you weasel your way >out of this one! > >> The removal of israeli citizens from a part of Gaza was a very positive >> move by Israel. But Israel still is very much involved in financial and >> military activities in gaza which irritates the people there to no end >> and which explains why they democratically elected a party that is very >> much against Israel. > >Amusing. Yes, Israel is irritating them by trying to prevent them from >firing rockets at its citizens. What nerve! ;-) Did it ever occur to >you that maybe the P's are irritating Israel? No, you never think of >such things because you are a bigot. What happened to all the aid that >went to the P's? And how did Arafat become a billionaire? (Hint: the >two are related.) > >> Everytime Israel reponds in kind to terrorist attacks on Israel, it >> motivates the palestinian terrorists even more and makes peace harder to >> achieve. > >Funny how that works both ways. Again it demonstrates your bigotry. > >> Everytime Israel cancels peace negotiations because some palestinian >> terrorism threw a bomb over the border, it postpones peace by a long time. > >Then the terrorist shouldn't have thrown the bomb. Duh! > >> The two sides should meet in some hotel and not be allowed to leave >> until they reach an agreement, and during such negotiations, they should >> not be made aware of any terrorist attack on Israel and/or Israeli >> military attack on palestinians. And they should be no media covering >> this, as well as no political opposition/influence. > >If you ever did achieve this, they'd die of old age in the same hotel. > >AEF > >P.S. BTW, I just heard on the radio today that Iraqis celebrated their >victory in the soccer Asia cup by firing guns into the air. Seven >dead, 50 injured, IIRC. Lovely people, eh? > ------------------------------ Date: Tue, 31 Jul 2007 07:09:31 -0700 From: urbancamo Subject: Maximum Java Heap Size/OpenVMS Java Experiences? Message-ID: <1185890971.291076.285690@d55g2000hsg.googlegroups.com> Hi, I was wondering if anyone knows what the maximum heap size is that can be specified for OpenVMS Java 5 on Itanium/Alpha? I could only find reference to a '32-bit' option, which I'm presuming isn't the default. Is there any developers out there using Java on VMS? If so, could they please share their experience with me. I am a developer since Java 1.1 and would welcome your opinions on how developing/deploying on OpenVMS differs to other environments. Many thanks for the help, Mark. ------------------------------ Date: Tue, 31 Jul 2007 18:15:57 +0200 From: "Martin Vorlaender" Subject: Re: Maximum Java Heap Size/OpenVMS Java Experiences? Message-ID: urbancamo wrote: > I was wondering if anyone knows what the maximum heap size is that can > be specified for OpenVMS Java 5 on Itanium/Alpha? I could only find > reference to a '32-bit' option, which I'm presuming isn't the default. > > Is there any developers out there using Java on VMS? If so, could they > please share their experience with me. I am a developer since Java 1.1 > and would welcome your opinions on how developing/deploying on OpenVMS > differs to other environments. I didn't (yet) do too much with Java on VMS, but I think the JDK User's Guide[1] has some useful tips, e.g. "Note that you are better off with a smaller heap that results in more garbage collections than allowing your application to page fault too often due to a large heap size.". There's also a paper on Optimizing Java Performance on OpenVMS[2] that - although it's not quite current - has some tips. HTH, Martin [1] http://h18012.www1.hp.com/java/documentation/1.5.0/ovms/docs/user_guide.html [2] http://h71000.www7.hp.com/ebusiness/OptimizingSDKGuide/OptimizingSDKGuide.html -- One OS to rule them all | Martin Vorlaender | OpenVMS rules! One OS to find them | work: mv@pdv-systeme.de One OS to bring them all | http://www.pdv-systeme.de/users/martinv/ And in the Darkness bind them.| home: martin.vorlaender@t-online.de ------------------------------ Date: Tue, 31 Jul 2007 07:40:34 +0200 From: "P. Sture" Subject: Re: Old DEC Monitor quandry Message-ID: In article <9Dvri.28958$Xa3.28172@attbi_s22>, "John E. Malmberg" wrote: > Larry Kilgallen wrote: > > In article , "John E. Malmberg" > > writes: > > > >>My ELSA card seemed to overdrive an LCD monitor that I had. The same > >>monitor worked fine with a Radeon card on the same DS10. > > > > Are there any situations in which the ELSA is preferable to the Radeon ? > > There must have been one. Island traded me the Radeon for my Elsa even > shortly after I got the LCD. > I remember Island doing such an offer, and wish I had taken it up. One reason I can think of is where you have a fleet of systems with ELSAs and are looking to add new systems or simply keep a stock of spares. Life can be a lot easier if you don't need to maintain separate DECwindows startup files for different video cards. -- Paul Sture Sue's OpenVMS bookmarks: http://eisner.encompasserve.org/~sture/ovms-bookmarks.html ------------------------------ Date: Tue, 31 Jul 2007 14:58:03 +0100 From: "Maciej W. Rozycki" Subject: Re: Old DEC Monitor quandry Message-ID: On Sun, 29 Jul 2007, Ben Myers wrote: > Most of the cable adapters I've seen are for co-ax monitors to SVGA 15-pin > graphics cards. You need the opposite. Not sure if they exist. I wonder if > one of the cable adapters would work if reversed? For sure, do not even > consider a flat panel LCD monitor with a digital video interface connector. DEC and later Compaq used to offer a suitable adapter box with three BNC inputs as of a monitor on one side and a HD15 output on the other. Its order number was 29-32549-01, priced at $30 back in 2000. Iiyama used to offer a similar "reverse" adapter cable as with some of their newer CRTs, e.g. ones with two HD15 inputs only, they recognised the lack of a BNC input as a shortcoming. The order number was MB31, IIRC. Similar parts may be available from independent manufacturers of cables/adapters too. Alternatively, if the workstation's graphics output is 3W3, then DEC and later Compaq used to offer 3W3 to HD15 adapter cables. Their order numbers were 17-03851-01 and 17-04816-01, priced at $86 and $66 back in 2002, respectively. For any of the former DEC parts you could certainly try asking HP through their support channel, but chances are they have gone out of stock now. I have got a pair of the 17-04816-01 cables which I use with my DECstation and VAXstation gear, so 1) they do exist and 2) they work. I spotted one on eBay once too. Of course all these solutions will result in some signal quality degradation, but that is unavoidable anyway if you use analogue signalling and HD15 connectors. Maciej ------------------------------ Date: Tue, 31 Jul 2007 09:13:18 +0100 From: Anton Shterenlikht Subject: remote shutdown/startup of MSA1000? Message-ID: <20070731081318.GA75790@mech-aslap33.men.bris.ac.uk> Is there a supported (or unsupported) way of remote MSA1000 shutdown or startup? I had a look in the contoller and switch commands and manuals, but cannot see any. IN fct there does not seem to be any way of orderly shutdown, just a power button. Perhaps it's all right for a storage array.. thanks anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 ------------------------------ Date: Tue, 31 Jul 2007 06:33:49 -0700 From: sapienzaf Subject: TRADE: BA370 rack mount Message-ID: <1185888829.619153.261260@w3g2000hsg.googlegroups.com> I have a new BA370 rackmount cabinet (5 power supplies, 24 disk slots, compatible with HSZ70 and HSG80 controllers, top gun blue) that I would like to trade for an RA7000/RA8000 pedestal cabinet. There are a variety of alternate part numbers for both the rack unit that I have and the pedestal cabinet that I want. The BA370 and RA7000/RA8000 are basically the same units, but I'd like the nice blue plastic on wheels for my nice office. Alternatively, I might be convinced to buy the plastic skins and caster base if anyone has them lying around going unused. Please contact me directly by email. ------------------------------ Date: Tue, 31 Jul 2007 09:32:10 +0100 From: Anton Shterenlikht Subject: VMS cluster behind a *NIX firewall Message-ID: <20070731083210.GA79688@mech-aslap33.men.bris.ac.uk> I need to have a *NIX frontend for a VMS cluster. In addition I'd like to set up cluster alias. I'm not sure how cluster alias will work together with port forwarding. The cluster has 3 nodes, two ds10l and one rx2620, all running VMS 8.3 with TCP/IP configured and started. Each node has 2 IP addresses, all in 192.168.0.0/24. I haven't yet set up a cluster alias. The frontend is FBSD 6.2 with IPF. The port forwarding rules, I can think of, wll only allow me to forward to a particular IP address in a particular node, e.g.: rdr xxx.xxx.xxx.xxx port 22 -> 192.168.0.xxx port 22 So this node then becomes a single point of failure. As far as I can see, I cannot have a host name in the above rule. Ideally, I'd like to have some sort of load balancing, i.e. let the users log into different nodes chosen automatically based on the current load, which is, I think, is the idea behind a cluster alias. Any advice on how to get a more useful port forwarding? thanks a lot anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 ------------------------------ Date: Tue, 31 Jul 2007 01:56:25 -0700 From: Volker Halle Subject: Re: VMS cluster behind a *NIX firewall Message-ID: <1185872185.198001.14210@w3g2000hsg.googlegroups.com> Anton, to remove the single point of failure, you should set up a FailSAFE IP address on those 3 nodes. This address will be offered on ONE interface at a time on ONE of the 3 nodes. The FailSAFE IP service handles the automatic detection of a failed interface or node and brings up that IP address on a working interface on one of the remaining nodes. Volker. ------------------------------ Date: Tue, 31 Jul 2007 11:06:55 +0100 From: "Richard Brodie" Subject: Re: VMS cluster behind a *NIX firewall Message-ID: "Volker Halle" wrote in message news:1185872185.198001.14210@w3g2000hsg.googlegroups.com... > to remove the single point of failure, you should set up a FailSAFE IP > address on those 3 nodes. This address will be offered on ONE > interface at a time on ONE of the 3 nodes. I might go for 3 alias addresses, with the default interface on different nodes. Then when you figure out how to load balance on the firewall, you are all set. ------------------------------ Date: 31 Jul 2007 13:28:19 +0200 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: Re: VMS cluster behind a *NIX firewall Message-ID: <46af38f3$1@news.langstoeger.at> In article <1185872185.198001.14210@w3g2000hsg.googlegroups.com>, Volker Halle writes: >to remove the single point of failure, you should set up a FailSAFE IP >address on those 3 nodes. This address will be offered on ONE >interface at a time on ONE of the 3 nodes. The FailSAFE IP service >handles the automatic detection of a failed interface or node and >brings up that IP address on a working interface on one of the >remaining nodes. And this interface the IP address sits on should be a LAN failover device (LLxy) as well (if you have more than one interface per machine of course)... -- 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: Tue, 31 Jul 2007 12:25:08 +0000 From: "Main, Kerry" Subject: RE: VMS cluster behind a *NIX firewall Message-ID: > -----Original Message----- > From: Anton Shterenlikht [mailto:mexas@bristol.ac.uk] > Sent: July 31, 2007 4:32 AM > To: Info-VAX@Mvb.Saic.Com > Subject: VMS cluster behind a *NIX firewall > > I need to have a *NIX frontend for a VMS cluster. In > addition > I'd like to set up cluster alias. I'm not sure how cluster alias > will > work together with port forwarding. > > The cluster has 3 nodes, two ds10l and one rx2620, all > running > VMS 8.3 with TCP/IP configured and started. Each node has 2 IP > addresses, > all in 192.168.0.0/24. I haven't yet set up a cluster alias. > > The frontend is FBSD 6.2 with IPF. The port forwarding > rules, I > can think of, wll only allow me to forward to a particular IP > address > in a particular node, e.g.: > > rdr xxx.xxx.xxx.xxx port 22 -> 192.168.0.xxx port 22 > > So this node then becomes a single point of failure. As far > as I > can see, I cannot have a host name in the above rule. Ideally, I'd > like > to have some sort of load balancing, i.e. let the users log into > different nodes chosen automatically based on the current load, > which is, > I think, is the idea behind a cluster alias. > > Any advice on how to get a more useful port forwarding? > > thanks a lot > anton > > -- > Anton Shterenlikht > Room 2.6, Queen's Building > Mech Eng Dept > Bristol University > University Walk, Bristol BS8 1TR, UK > Tel: +44 (0)117 928 8233 > Fax: +44 (0)117 929 4423 Anton, In addition to the options already presented, you might also want to consid= er load balancing using the OpenVMS TCPIP Load Broker. You could also combi= ne the FailSAFE strategy with the Load Broker approach. Advantages: - can load balance across cluster using cpu, memory information which also = reflects batch loads. Many load balancers only use simple connection based = strategies e.g. round robin, number of connections etc. - can set flag (disable users) which allows current connections to complete= , but all new connections to be directed to other nodes in cluster. When cu= rrent connections are completed, that node can be shutdown without having t= o inform end users or get permission from the business - it is transparent = to them. - can define logicals dynamically to assign weights to different servers to= better balance workloads specific to local preferences. Reference: http://tinyurl.com/235epl (DNS Load Broker and TCPIP Load Balancing) http://h71000.www7.hp.com/openvms/journal/v2/articles/tcpip.pdf (TCPIP Fail= SAFE paper) Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT) OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Tue, 31 Jul 2007 06:02:41 -0700 From: "Tom Linden" Subject: Re: VMS cluster behind a *NIX firewall Message-ID: On Tue, 31 Jul 2007 01:32:10 -0700, Anton Shterenlikht wrote: > I need to have a *NIX frontend for a VMS cluster. In addition > I'd like to set up cluster alias. I'm not sure how cluster alias will > work together with port forwarding. Why do you need this front-end? > > The cluster has 3 nodes, two ds10l and one rx2620, all running > VMS 8.3 with TCP/IP configured and started. Each node has 2 IP addresses, > all in 192.168.0.0/24. I haven't yet set up a cluster alias. > > The frontend is FBSD 6.2 with IPF. The port forwarding rules, I > can think of, wll only allow me to forward to a particular IP address > in a particular node, e.g.: > > rdr xxx.xxx.xxx.xxx port 22 -> 192.168.0.xxx port 22 > > So this node then becomes a single point of failure. As far as I > can see, I cannot have a host name in the above rule. Ideally, I'd like > to have some sort of load balancing, i.e. let the users log into > different nodes chosen automatically based on the current load, which is, > I think, is the idea behind a cluster alias. > > Any advice on how to get a more useful port forwarding? > > thanks a lot > anton > -- PL/I for OpenVMS www.kednos.com ------------------------------ Date: Tue, 31 Jul 2007 14:19:09 +0100 From: Anton Shterenlikht Subject: Re: VMS cluster behind a *NIX firewall Message-ID: <20070731131909.GA53337@mech-aslap33.men.bris.ac.uk> On Tue, Jul 31, 2007 at 06:02:41AM -0700, Tom Linden wrote: > On Tue, 31 Jul 2007 01:32:10 -0700, Anton Shterenlikht > wrote: > > > I need to have a *NIX frontend for a VMS cluster. In addition > >I'd like to set up cluster alias. I'm not sure how cluster alias will > >work together with port forwarding. > Why do you need this front-end? The Uni comp. services are not happy for me to connect the cluster to the network directly. I'm not sure whether they do not trust the security of the VMS, or my ability to maintain it. In any case there is nobody with any VMS skills at the comp. services at present, so they probably thought it safer to make me do it this way. -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 ------------------------------ Date: Tue, 31 Jul 2007 14:21:36 +0100 From: Anton Shterenlikht Subject: Re: VMS cluster behind a *NIX firewall Message-ID: <20070731132136.GA54049@mech-aslap33.men.bris.ac.uk> On Tue, Jul 31, 2007 at 12:25:08PM +0000, Main, Kerry wrote: > > -----Original Message----- > > From: Anton Shterenlikht [mailto:mexas@bristol.ac.uk] > > > > I need to have a *NIX frontend for a VMS cluster. In > Anton, > > In addition to the options already presented, you might also want > to consider load balancing using the OpenVMS TCPIP Load Broker. > You could also combine the FailSAFE strategy with the Load Broker approach. thank you to all who replied. I'll do my reading now. anton -- Anton Shterenlikht Room 2.6, Queen's Building Mech Eng Dept Bristol University University Walk, Bristol BS8 1TR, UK Tel: +44 (0)117 928 8233 Fax: +44 (0)117 929 4423 ------------------------------ Date: Tue, 31 Jul 2007 06:27:42 -0700 From: etmsreec@yahoo.co.uk Subject: Re: VMS cluster behind a *NIX firewall Message-ID: <1185888462.177442.185130@o61g2000hsh.googlegroups.com> On 31 Jul, 14:19, Anton Shterenlikht wrote: > On Tue, Jul 31, 2007 at 06:02:41AM -0700, Tom Linden wrote: > > On Tue, 31 Jul 2007 01:32:10 -0700, Anton Shterenlikht > > wrote: > > > > I need to have a *NIX frontend for a VMS cluster. In addition > > >I'd like to set up cluster alias. I'm not sure how cluster alias will > > >work together with port forwarding. > > Why do you need this front-end? > > The Uni comp. services are not happy for me to connect the cluster to the > network directly. I'm not sure whether they do not trust the security > of the VMS, or my ability to maintain it. In any case there is nobody > with any VMS skills at the comp. services at present, so they probably > thought it safer to make me do it this way. > > -- > Anton Shterenlikht > Room 2.6, Queen's Building > Mech Eng Dept > Bristol University > University Walk, Bristol BS8 1TR, UK > Tel: +44 (0)117 928 8233 > Fax: +44 (0)117 929 4423 They've probably heard that VMSclusters are "terribly chatty" on the network and so want to make sure that none of their systems have network packets that might be foreign (i.e. anything that's not IP)! There are consultants around Bristol that could help the CSD guys out with VMS if they wished, or they could take someone on full time... If you need any UK based help let me know. Either way, good to see VMS being used in academia. Well done! Steve ------------------------------ Date: Tue, 31 Jul 2007 13:39:29 +0000 From: "Main, Kerry" Subject: RE: VMS cluster behind a *NIX firewall Message-ID: -----Original Message----- > From: Main, Kerry > Sent: July 31, 2007 8:25 AM > To: Info-VAX@Mvb.Saic.Com > Subject: RE: VMS cluster behind a *NIX firewall > > > -----Original Message----- > > From: Anton Shterenlikht [mailto:mexas@bristol.ac.uk] > > Sent: July 31, 2007 4:32 AM > > To: Info-VAX@Mvb.Saic.Com > > Subject: VMS cluster behind a *NIX firewall > > > > I need to have a *NIX frontend for a VMS cluster. In > > addition > > I'd like to set up cluster alias. I'm not sure how cluster alias > > will > > work together with port forwarding. > > > > The cluster has 3 nodes, two ds10l and one rx2620, all > > running > > VMS 8.3 with TCP/IP configured and started. Each node has 2 IP > > addresses, > > all in 192.168.0.0/24. I haven't yet set up a cluster alias. > > > > The frontend is FBSD 6.2 with IPF. The port forwarding > > rules, I > > can think of, wll only allow me to forward to a particular IP > > address > > in a particular node, e.g.: > > > > rdr xxx.xxx.xxx.xxx port 22 -> 192.168.0.xxx port 22 > > > > So this node then becomes a single point of failure. As > far > > as I > > can see, I cannot have a host name in the above rule. Ideally, > I'd > > like > > to have some sort of load balancing, i.e. let the users log into > > different nodes chosen automatically based on the current load, > > which is, > > I think, is the idea behind a cluster alias. > > > > Any advice on how to get a more useful port forwarding? > > > > thanks a lot > > anton > > > > -- > > Anton Shterenlikht > > Room 2.6, Queen's Building > > Mech Eng Dept > > Bristol University > > University Walk, Bristol BS8 1TR, UK > > Tel: +44 (0)117 928 8233 > > Fax: +44 (0)117 929 4423 > > Anton, > > In addition to the options already presented, you might also want > to consider load balancing using the OpenVMS TCPIP Load Broker. You > could also combine the FailSAFE strategy with the Load Broker > approach. > > Advantages: > - can load balance across cluster using cpu, memory information > which also reflects batch loads. Many load balancers only use > simple connection based strategies e.g. round robin, number of > connections etc. > - can set flag (disable users) which allows current connections to > complete, but all new connections to be directed to other nodes in > cluster. When current connections are completed, that node can be > shutdown without having to inform end users or get permission from > the business - it is transparent to them. > - can define logicals dynamically to assign weights to different > servers to better balance workloads specific to local preferences. > > Reference: > http://tinyurl.com/235epl (DNS Load Broker and TCPIP Load > Balancing) > http://h71000.www7.hp.com/openvms/journal/v2/articles/tcpip.pdf > (TCPIP FailSAFE paper) > > Just re-read the original note and I misread the point about only being abl= e to forward IP physical addresses vs. FQDN's. mmm... it sounds like the key is to build some smarts (filters??) into the= UNIX front end such that a particular port request is translated to a FQDN= which in turn can be load balanced etc. Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT) OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Tue, 31 Jul 2007 16:44:02 +0000 From: "Main, Kerry" Subject: RE: VMS cluster behind a *NIX firewall Message-ID: > -----Original Message----- > From: etmsreec@yahoo.co.uk [mailto:etmsreec@yahoo.co.uk] > Sent: July 31, 2007 9:28 AM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: VMS cluster behind a *NIX firewall > > On 31 Jul, 14:19, Anton Shterenlikht wrote: > > On Tue, Jul 31, 2007 at 06:02:41AM -0700, Tom Linden wrote: > > > On Tue, 31 Jul 2007 01:32:10 -0700, Anton Shterenlikht > > > wrote: > > > > > > I need to have a *NIX frontend for a VMS cluster. In > addition > > > >I'd like to set up cluster alias. I'm not sure how cluster > alias will > > > >work together with port forwarding. > > > Why do you need this front-end? > > > > The Uni comp. services are not happy for me to connect the > cluster to the > > network directly. I'm not sure whether they do not trust the > security > > of the VMS, or my ability to maintain it. In any case there is > nobody > > with any VMS skills at the comp. services at present, so they > probably > > thought it safer to make me do it this way. > > > > -- > > Anton Shterenlikht > > Room 2.6, Queen's Building > > Mech Eng Dept > > Bristol University > > University Walk, Bristol BS8 1TR, UK > > Tel: +44 (0)117 928 8233 > > Fax: +44 (0)117 929 4423 > > They've probably heard that VMSclusters are "terribly chatty" on > the > network and so want to make sure that none of their systems have > network packets that might be foreign (i.e. anything that's not > IP)! > > There are consultants around Bristol that could help the CSD guys > out > with VMS if they wished, or they could take someone on full time... > If you need any UK based help let me know. > > Either way, good to see VMS being used in academia. Well done! > > Steve As a suggestion, it is usually a good practice from a security and scalabil= ity perspective to use VLANS for these types of environments anyway. With O= penVMS, you can direct which NIC(s) will carry the cluster traffic and whic= h ones will carry regular TCPIP traffic. While it does raise the complexity a tad (2 ports per VLAN for HA environme= nts), and not officially required, I like to recommend having 3 VLAN's - MG= MT (remote management, remote console devices, OS provisioning), CLUSTER an= d USER (normal TCPIP). Some larger Cust's also like to have a BACKUP VLAN f= or doing network backups. Regards Kerry Main Senior Consultant HP Services Canada Voice: 613-592-4660 Fax: 613-591-4477 kerryDOTmainAThpDOTcom (remove the DOT's and AT) OpenVMS - the secure, multi-site OS that just works. ------------------------------ Date: Tue, 31 Jul 2007 16:49:44 +0000 From: "Paul Raulerson" Subject: Re: VMS cluster behind a *NIX firewall Message-ID: ----=_vm_0011_W994373768_24136_1185900584 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable >I need to have a *NIX frontend for a VMS cluster. In addition >I'd like to set up cluster alias. I'm not sure how cluster alias will >work together with port forwarding. > >The cluster has 3 nodes, two ds10l and one rx2620, all running >VMS 8.3 with TCP/IP configured and started. Each node has 2 IP addresses= , >all in 192.168.0.0/24. I haven't yet set up a cluster alias. > >The frontend is FBSD 6.2 with IPF. The port forwarding rules, I >can think of, wll only allow me to forward to a particular IP address >in a particular node, e.g.: > >rdr xxx.xxx.xxx.xxx port 22 -> 192.168.0.xxx port 22 > >So this node then becomes a single point of failure. As far as I >can see, I cannot have a host name in the above rule. Ideally, I'd like >to have some sort of load balancing, i.e. let the users log into >different nodes chosen automatically based on the current load, which is= , >I think, is the idea behind a cluster alias. > If I understand you correctly, what you really want then is to use a Free= BSD machine as a load balancing proxy server for the VMS cluster? There are several easy ways to do that, but probably the easiest of all i= s to just write a small custom proxy server, and attach it a specific por= t on the Free BSD machine with xinetd. When started, it can be something as simple as a round robin to each of t= he six IP addresses available on the VMS cluster, or you can make it as c= omlpex as you need it to be. Starting it from xinetd avoids the need to fork() a process for each conn= ection (xinetd already did that for you) and you can then simply read/wri= te to both the external connection (with STDIN and STDOUT already connect= ed for you) and to the remote program. There are source several examples = of simple proxy servers out there you could use or modify. Perl,or C woul= d be easily appropriate to that task. There are also prebuilt proxy packa= ges you just drop in place, but they may be overkill indeed for what you = want. :) The end result is your users would always connect to a specific port on t= he FreeBSD machine (inside the firewall, one port drilled through) and fr= om there make balanced connections to your VMS cluster. -Paul >Any advice on how to get a more useful port forwarding? > >thanks a lot >anton ----=_vm_0011_W994373768_24136_1185900584-- ------------------------------ Date: Tue, 31 Jul 2007 17:33:37 GMT From: Tad Winters Subject: Re: VMS cluster behind a *NIX firewall Message-ID: Anton Shterenlikht wrote in news:20070731083210.GA79688@mech-aslap33.men.bris.ac.uk: [snip..] > > The frontend is FBSD 6.2 with IPF. The port forwarding rules, I > can think of, wll only allow me to forward to a particular IP address > in a particular node, e.g.: > > rdr xxx.xxx.xxx.xxx port 22 -> 192.168.0.xxx port 22 > > So this node then becomes a single point of failure. As far as I > can see, I cannot have a host name in the above rule. Ideally, I'd like > to have some sort of load balancing, i.e. let the users log into > different nodes chosen automatically based on the current load, which is, > I think, is the idea behind a cluster alias. > It sounds like the *nix firewall becomes the single point of failure. ------------------------------ Date: Tue, 31 Jul 2007 17:44:34 GMT From: Tad Winters Subject: Re: VMS cluster behind a *NIX firewall Message-ID: "Paul Raulerson" wrote in news:W994373768241361185900584@webmail31: > If I understand you correctly, what you really want then is to use a > Free BSD machine as a load balancing proxy server for the VMS cluster? > [snip..] This makes the FreeBSD machine a point of failure, along with the *nix firewall. ------------------------------ Date: Tue, 31 Jul 2007 17:52:25 +0000 From: "Paul Raulerson" Subject: Re: VMS cluster behind a *NIX firewall Message-ID: ----=_vm_0011_W406317719_32470_1185904345 Content-Type: multipart/alternative; boundary="--=_vm_0016_W406317719_32470_1185904345" ----=_vm_0016_W406317719_32470_1185904345 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable > It sounds like the *nix firewall becomes the single point of failure. The first SPOF is the actual firewall, but they have probably addressed t= hat by putting some kind of fail-over system in. For example, I use multi= ple Cisco ASA's here at work. Should one fail, the second one picks up th= e load, including the dozens of active VPN sessions connecting our remote= offices. In terms of the Linux proxy-like server, that can be addressed several wa= ys too - I would probably use a virutal server and a second backup hardwa= re platform, such as a blade. Means I have to spend 5 minutes with it if = it fails, but it is fast and simple. Otherwise, I would address it with a= ny of a number of HA solutions. -Paul ----=_vm_0016_W406317719_32470_1185904345 Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable

> It sounds like the *nix firew= all becomes the single point of failure.

The first SPOF is the actual firewall, but they have probably addresse= d that by putting some kind of fail-over system in. For example, I use mu= ltiple Cisco ASA's here at work. Should one fail, the second one picks up= the load, including the dozens of active VPN sessions connecting our rem= ote offices.

In terms of the Linux proxy-like server, that can be addressed several= ways too - I would probably use a virutal server and a second backu= p hardware platform, such as a blade.  Means I have to spend 5 minut= es with it if it fails, but it is fast and simple. Otherwise, I would add= ress it with any of a number of HA solutions.

-Paul

 

----=_vm_0016_W406317719_32470_1185904345-- ----=_vm_0011_W406317719_32470_1185904345-- ------------------------------ Date: Tue, 31 Jul 2007 02:27:25 -0500 From: Ron Johnson Subject: Re: What does GEM mean? Message-ID: On 07/30/07 07:07, Bill Gunshannon wrote: > In article <46AB2BA3.7090100@comcast.net>, > "Richard B. Gilbert" writes: >> P. Sture wrote: >>> In article , >>> John Reagan wrote: >>> >>> >>>> John Reagan wrote: >>>> >>>> >>>>> German words like straße (5 chars long) and STRASSE (6 chars long). >>>> and to compound the problem, I can't count. :-) >>> >>> And why would a compiler author need to? :-) >>> >> Indeed! That's what computers are for! If God had meant us to count >> above twenty he would have given us more digits! ;-) >> > > Look how high computers can count and they only have two digits. Is that the zoological or mathematical definition of "digit"? -- Ron Johnson, Jr. Jefferson LA USA Give a man a fish, and he eats for a day. Hit him with a fish, and he goes away for good! ------------------------------ Date: Tue, 31 Jul 2007 07:23:37 -0500 From: Roy Goodman Subject: Re: [OT] RIP DTL's Father Message-ID: Didier_Toulouse wrote in news:1185767058.961026.121760@o61g2000hsh.googlegroups.com: > Good morning Friends, > > My Father died yesterday in Geneva, Switzerland. His heart just > stopped while sleeping. He was 93 and in bad health. We were expecting > this since a few months. > Didier: I know you don't know me, but I remember you well from notes conferences at DEC all those years ago. Please accept my sympathies on your loss. Roy Goodman (remove "nospa" from address to reply) ------------------------------ End of INFO-VAX 2007.416 ************************