INFO-VAX Thu, 13 Nov 2008 Volume 2008 : Issue 615 Contents: Re: AlphaServer 8400 for Trade Re: AlphaServer 8400 for Trade Re: AlphaServer 8400 for Trade Re: AlphaServer 8400 for Trade Re: EDT: inserting special characters Re: EDT: inserting special characters Re: EDT: inserting special characters Re: EDT: inserting special characters Re: EDT: inserting special characters RE: EDT: inserting special characters Re: Emulation Re: Emulation Re: Emulation Re: Emulation Re: Emulation Re: Emulation Re: Good news: HP has retaken the VMS trademark ! Re: Most impressive VAX installations Re: Most impressive VAX installations OpenSSL 1.3 and smime ? Re: Peek&Spy anyone ? Re: Peek&Spy anyone ? Re: Peek&Spy anyone ? Re: Peek&Spy anyone ? Seagate ST31051N on MicroVAX 3100 Re: Seamonkey browser port for Alpha now available Re: SFF (Send From File) Utility strange disk state Re: strange disk state Re: strange disk state Re: strange disk state Re: strange disk state Re: strange disk state Re: Synchronizing SYSUAF between independent machines Re: VAX-11/785 microdiagnostics help requested [VMS V8.3] PCSI Kit Verification ---------------------------------------------------------------------- Date: 13 Nov 2008 08:59:43 GMT From: Dennis Grevenstein Subject: Re: AlphaServer 8400 for Trade Message-ID: In comp.unix.tru64 Wilm Boerhout wrote: > Alpha's don't count VUPs, see (a.o.) > http://www.compaq.com/alphaserver/performance/vups_297.html > > If you still want to compare, from Table 1 on that page, I infer that > the (SPECxxx) performance of an Alphaserver 8400 5/300 is roughly 10 > times a VAX 7000-610, or 350 VUPs. YMMV. Then again, you can't be sure how those benchmark numbers really scale, so you can't infer how an 11/750 might come out on SPEC. Personally, if you want to compare performance, I would advice to get those old dhrystone sources out of the crypt and get *proper* results as a VAX MIPS rating for both. :-) There's nothing better than collecting data yourself. Much more fun. Dennis -- Don't suffer from insanity... Enjoy every minute of it. ------------------------------ Date: Thu, 13 Nov 2008 06:20:21 -0500 From: Bob Willard Subject: Re: AlphaServer 8400 for Trade Message-ID: Phillip Helbig---remove CLOTHES to reply wrote: > In article > , > vaxorcist writes: > > >>I will trade off my AlphaServer 8400 for a VAX-11/750. > > > I guess the power consumption will stay about the same. > > The VAX is 1 VUP, right? The 8400 must be about 1500. Not a judgement, > just an observation. While I don't have such big machines at home, I > can appreciate the joy of running a VAX (of which I have several, all > small and relatively fast (for VAXes). > The basis for the VUP was the VAX-11/780. IIRC, the VAX-11/750 was ~0.7 VUP. -- Cheers, Bob ------------------------------ Date: Thu, 13 Nov 2008 08:18:01 -0800 (PST) From: Galen Subject: Re: AlphaServer 8400 for Trade Message-ID: Dennis wrote: > > Personally, if you want to compare performance, I would advice to > get those old dhrystone sources out of the crypt and get *proper* > results as a VAX MIPS rating for both. :-) If I recall correctly--and it's been a long time, so I may be inaccurate about the details--good modern compilers (since, maybe, the early 1990's?) may optimize away much or all of a simple benchmark like dhrystone or whetstone, rendering them, in effect, a great big no- op. If so you don't really get a measurement that will help predict the performance of real applications. Also, modern processors may fit most or all of these small benchmark applications into high speed cache or use other hardware features that render the old benchmarks less than useless. And how do multiple processors I am not an expert on any of this, and I hope you'll pardon any blunders or oversights I've made. I also don't know to what extent the most recent OpenVMS VAX compilers are capable of this kind of optimization. But I do recall a lot of debate about the usefulness of dhrystone and other benchmarks way back in the late 1980's or early 1990's when we were evaluating early workstations from Apollo, HP, and other companies. Of course, if all you're looking for is some cheap fun then don't let any of this stand in your way. :-) ------------------------------ Date: 13 Nov 2008 17:08:16 GMT From: Dennis Grevenstein Subject: Re: AlphaServer 8400 for Trade Message-ID: In comp.sys.dec Galen wrote: > > If I recall correctly--and it's been a long time, so I may be > inaccurate about the details--good modern compilers (since, maybe, the > early 1990's?) may optimize away much or all of a simple benchmark > like dhrystone or whetstone, rendering them, in effect, a great big no- > op. If so you don't really get a measurement that will help predict > the performance of real applications. Also, modern processors may fit > most or all of these small benchmark applications into high speed > cache or use other hardware features that render the old benchmarks > less than useless. And how do multiple processors That's just my point. You can't predict how an 11/750 might score on SPEC. It wouldn't be a fair comparison, because modern benchmarks are made to show differences between modern computers. If an old benchmark doesn't move enough data to go beyond the CPU cache on modern machines, that's fine, because old machines don't have a cache like that and that's one reason why they are slower. It all depends on what kind of an application you are thinking about. You can't say that an AS8400 is X times faster than an 11/750. For whom? For a single user editing a text file with vi or for a programmer who can use the entire machine for serious number crunching? In the end, it's only a useless number, but you could say that the newer machine can do the same work much faster than the old machine. If you want to make that comparision, stick to the old type of work that the old machine was built to do. Dennis -- Don't suffer from insanity... Enjoy every minute of it. ------------------------------ Date: Thu, 13 Nov 2008 09:38:47 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: EDT: inserting special characters Message-ID: In article , Peter Weaver writes: > On Nov 12, 10:42=A0am, hel...@astro.multiCLOTHESvax.de (Phillip Helbig--- > remove CLOTHES to reply) wrote: > > In article , hel...@astro.multiCLOTHESvax.de > > (Phillip Helbig---remove CLOTHES to reply) writes: > > > > > A few messages even encode and , so I did the obvious > > > extension: > > > > > INSERT;s|=3D0D||w > > > INSERT;s|=3D0A||w > > > > > However, it doesn't work. =A0No error message; no effect. > > > > > (Note: Of course, and are each represented above as 4 other > > > characters; in the macro the actual ASCII 13 and 10 characters are used= > , > > > inserted via "GOLD GOLD 3".) > > > > > Is what I want to do possible? > > ... > > When I tried your changes it did what you told it to do, but not what > I think you wanted it to so I am not sure why you had "No effect." Can you explain where my thinking went wrong? > The > code below will do what I think you wanted to do; Take the =0D and > replace it with a line-break. I also changed it so it only uses 7-bit > ASCII characters since I never like seeing non-printable characters in > code if I can avoid it. Thanks. It works. > ! > ! Create a buffer with two blank lines (for some reason one > ! blank line is not enough???) > ! > find buffer cr_buffer > insert; > insert; > find last This is strange. A common occurrence is when someone inserts a blank line to form a paragraph. I then see =0D=0A=0D=0A. This corresponds to adding 1 blank line. The stuff above adds 2 blank lines. Removing one of the "insert;" statements results in 0 blank lines. Real EDT wizards, please stand up! ------------------------------ Date: Thu, 13 Nov 2008 11:48:39 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: EDT: inserting special characters Message-ID: <00A828E4.52B86831@SendSpamHere.ORG> In article , helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: >In article >, >Peter Weaver writes: > >> On Nov 12, 10:42=A0am, hel...@astro.multiCLOTHESvax.de (Phillip Helbig--- >> remove CLOTHES to reply) wrote: >> > In article , hel...@astro.multiCLOTHESvax.de >> > (Phillip Helbig---remove CLOTHES to reply) writes: >> > >> > > A few messages even encode and , so I did the obvious >> > > extension: >> > >> > > INSERT;s|=3D0D||w >> > > INSERT;s|=3D0A||w >> > >> > > However, it doesn't work. =A0No error message; no effect. >> > >> > > (Note: Of course, and are each represented above as 4 other >> > > characters; in the macro the actual ASCII 13 and 10 characters are used= >> , >> > > inserted via "GOLD GOLD 3".) >> > >> > > Is what I want to do possible? >> > ... >> >> When I tried your changes it did what you told it to do, but not what >> I think you wanted it to so I am not sure why you had "No effect." > >Can you explain where my thinking went wrong? > >> The >> code below will do what I think you wanted to do; Take the =0D and >> replace it with a line-break. I also changed it so it only uses 7-bit >> ASCII characters since I never like seeing non-printable characters in >> code if I can avoid it. > >Thanks. It works. > >> ! >> ! Create a buffer with two blank lines (for some reason one >> ! blank line is not enough???) >> ! >> find buffer cr_buffer >> insert; >> insert; >> find last > >This is strange. A common occurrence is when someone inserts a blank >line to form a paragraph. I then see =0D=0A=0D=0A. This corresponds to >adding 1 blank line. The stuff above adds 2 blank lines. Removing one >of the "insert;" statements results in 0 blank lines. > >Real EDT wizards, please stand up! I perused over this thread too lightly. What is the question/issue? I saw an attempt to replace quoted-pukeable with something more along the lines of textual-readable with an EDT substitution macro. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM ... pejorative statements of opinion are entitled to constitutional protection no matter how extreme, vituperous, or vigorously expressed they may be. (NJSC) Copr. 2008 Brian Schenkenberger. Publication of _this_ usenet article outside of usenet _must_ include its contents in its entirety including this copyright notice, disclaimer and quotations. ------------------------------ Date: Thu, 13 Nov 2008 16:09:24 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: EDT: inserting special characters Message-ID: In article <00A828E4.52B86831@SendSpamHere.ORG>, VAXman- @SendSpamHere.ORG writes: > >Real EDT wizards, please stand up! > > I perused over this thread too lightly. What is the question/issue? > I saw an attempt to replace quoted-pukeable with something more along > the lines of textual-readable with an EDT substitution macro. Yes, that's the idea. Works fine. Sadly, I need it more and more even for posts in comp.os.vms. The problem I had was replacing =0D with and =0A with . The other substitutions work fine. ------------------------------ Date: Thu, 13 Nov 2008 16:19:31 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: EDT: inserting special characters Message-ID: <00A8290A.29DDE719@SendSpamHere.ORG> In article , helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: >In article <00A828E4.52B86831@SendSpamHere.ORG>, VAXman- >@SendSpamHere.ORG writes: > >> >Real EDT wizards, please stand up! >> >> I perused over this thread too lightly. What is the question/issue? >> I saw an attempt to replace quoted-pukeable with something more along >> the lines of textual-readable with an EDT substitution macro. > >Yes, that's the idea. Works fine. Sadly, I need it more and more even >for posts in comp.os.vms. > >The problem I had was replacing =0D with and =0A with . The >other substitutions work fine. They didn't? I'd define your EDT macro in your EDT init file. Use the special insert ([GOLD] [GOLD] [3]) keypad sequence to put the and into that macro. I know this works as I have such a macro to convert and in a file to printable versions of and . A couple of others too such as tabs and some control char- acters. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM ... pejorative statements of opinion are entitled to constitutional protection no matter how extreme, vituperous, or vigorously expressed they may be. (NJSC) Copr. 2008 Brian Schenkenberger. Publication of _this_ usenet article outside of usenet _must_ include its contents in its entirety including this copyright notice, disclaimer and quotations. ------------------------------ Date: Thu, 13 Nov 2008 16:23:49 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: EDT: inserting special characters Message-ID: In article <00A8290A.29DDE719@SendSpamHere.ORG>, VAXman- @SendSpamHere.ORG writes: > >> I perused over this thread too lightly. What is the question/issue? > >> I saw an attempt to replace quoted-pukeable with something more along > >> the lines of textual-readable with an EDT substitution macro. > > > >Yes, that's the idea. Works fine. Sadly, I need it more and more even > >for posts in comp.os.vms. > > > >The problem I had was replacing =0D with and =0A with . The > >other substitutions work fine. > > They didn't? I'd define your EDT macro in your EDT init file. That's what I did. > Use the > special insert ([GOLD] [GOLD] [3]) keypad sequence to put > the and into that macro. That's what I did. > I know this works as I have such a > macro to convert and in a file to printable versions of > and . A couple of others too such as tabs and some control char- > acters. Strange that my macro didn't work. One followup seemed to indicate that it was clear why it didn't work as expected, but I haven't seen the corresponding explanation. ------------------------------ Date: Thu, 13 Nov 2008 12:40:03 -0500 From: "Peter Weaver" Subject: RE: EDT: inserting special characters Message-ID: <23C780ECBCFD4A0E951483AB90533F93@CHARONLAP> >... > This is strange. A common occurrence is when someone inserts a blank > line to form a paragraph. I then see =0D=0A=0D=0A. This corresponds to > adding 1 blank line. The stuff above adds 2 blank lines. Removing one > of the "insert;" statements results in 0 blank lines. > > Real EDT wizards, please stand up! I do not know why the CR_BUFFER needs two lines, for some reason one of the lines gets removed on the PASTE=CR_BUFFER. But you ware getting two blank lines on the =0D=0A=0D=0A because the line change; 9999('=0D=0A' cutsr paste=cr_buffer) ex replaces the first =0D=0A with a new-line then replaces the second =0D=0A with another new-line. Add the lines; ! Double Linefeed/CR Combination INSERT;%B INSERT;change; 9999('=0D=0A=0D=0A' cutsr paste=cr_buffer) ex Above the =0D=0A line to do what you want there. The reason your original "INSERT;s|=0D||w" line did not do what you wanted is that it was just putting a CR (ASCII 13) character into the record, it was not creating a new record. You can see what I mean by running your writing out the file then do a DUMP/RECORD of the file. Then run the modified macro and DUMP/RECORD the file again. Peter Weaver www.weaverconsulting.ca www.openvmsvirtualization.com www.vaxvirtualization.com www.alphavirtualization.com Winner of the 2007 OpenVMS.org Readers' Choice Award for System Management/Performance ------------------------------ Date: Thu, 13 Nov 2008 06:58:24 -0800 From: "Tom Linden" Subject: Re: Emulation Message-ID: On Wed, 12 Nov 2008 10:19:28 -0800, Wilm Boerhout wrote: > Tom Linden vaguely mentioned on 12-11-2008 18:50: > >> I have wondered before about the legality of limiting it to CHARON. It >> seems like restraint of trade, which is illegal. > > Or not. Your original VMS license only allows you to run VMS on a choice > of VAX, Alpha or Itanium all from HP (see the VMS SPD). That appears to > be/have been legal. > > But first of all, it's HP who grants the transfer licenses, so they > should speak up of what their intention is for commercial use (CHARON > only or any emulator). If it is then contended in a particular > jurisdiction, let the judge decide. Make some lawyers happy... There is a distinction. The HW, OS and CHARON all are third party, so if someone comes along with HW, OS-X and BrandX emulator. Then what? IOW, HP is selling a piece of software for a platform it does not provide. I don't think they can deny sale of said software to another platform. They can, of course, refuse to provide support. > > /Wilm -- PL/I for OpenVMS www.kednos.com ------------------------------ Date: Thu, 13 Nov 2008 07:38:01 -0800 (PST) From: emu Subject: Re: Emulation Message-ID: On Nov 12, 11:19 am, Wilm Boerhout wrote: > But first of all, it's HP who grants the transfer licenses, so they > should speak up of what their intention is for commercial use (CHARON > only or any emulator). If it is then contended in a particular > jurisdiction, let the judge decide. Make some lawyers happy... But HP doesn't speak up, so they probably (?) don't care at all, as long as you paid the transfer license fees. ------------------------------ Date: Thu, 13 Nov 2008 07:41:33 -0800 (PST) From: emu Subject: Re: Emulation Message-ID: On Nov 13, 7:58 am, "Tom Linden" wrote: > There is a distinction. The HW, OS and CHARON all are third party, so > if someone comes along with HW, OS-X and BrandX emulator. Then what? IOW, > HP is selling a piece of software for a platform it does not provide. I > don't think they can deny sale of said software to another platform. They can, > of course, refuse to provide support. And HP provides software support, quote from their web page : "HP fixes software problems if they are also seen in the comparable VAX or Alpha AXP environment." I guess, a lot of people can live with that. Hopefully, they still have the original equipment around to prove that the problem also exist on the original machine ;-) Cheers ------------------------------ Date: 13 Nov 2008 16:04:38 GMT From: billg999@cs.uofs.edu (Bill Gunshannon) Subject: Re: Emulation Message-ID: <6o31gmF1l1o5U1@mid.individual.net> In article , "Tom Linden" writes: > On Wed, 12 Nov 2008 10:19:28 -0800, Wilm Boerhout > wrote: > >> Tom Linden vaguely mentioned on 12-11-2008 18:50: >> >>> I have wondered before about the legality of limiting it to CHARON. It >>> seems like restraint of trade, which is illegal. >> >> Or not. Your original VMS license only allows you to run VMS on a choice >> of VAX, Alpha or Itanium all from HP (see the VMS SPD). That appears to >> be/have been legal. >> >> But first of all, it's HP who grants the transfer licenses, so they >> should speak up of what their intention is for commercial use (CHARON >> only or any emulator). If it is then contended in a particular >> jurisdiction, let the judge decide. Make some lawyers happy... > > There is a distinction. The HW, OS and CHARON all are third party, so > if someone comes along with HW, OS-X and BrandX emulator. Then what? IOW, > HP is selling a piece of software for a platform it does not provide. I > don't > think they can deny sale of said software to another platform. They can, > of > course, refuse to provide support. While I agree with you in principle, Tom, I don't think you really have an argument. They don't "sell" the software, they license it for use on platforms of their choice. Choosing not to offer a license for a particular platform does not restrict anyone's ability to "trade" in SIMH. 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: Thu, 13 Nov 2008 17:14:36 +0100 From: Wilm Boerhout Subject: Re: Emulation Message-ID: <491c5287$0$8577$ba620dc5@nova.planet.nl> emu vaguely mentioned on 13-11-2008 16:38: > But HP doesn't speak up, so they probably (?) don't care at all, as > long as you paid the transfer license fees. Indeed they don't speak up and (so) don't care, or haven't done so in the past few years when I asked them this same question every now and then. We surely moved to the far corner of an edge in a niche... /Wilm [who is, also, a former Digital Equipment employee, involved in software licensing practices a.o.. I left no successors there apparently...] ------------------------------ Date: Thu, 13 Nov 2008 08:44:14 -0800 From: "Tom Linden" Subject: Re: Emulation Message-ID: On Thu, 13 Nov 2008 08:04:38 -0800, Bill Gunshannon wrote: > In article , > "Tom Linden" writes: >> On Wed, 12 Nov 2008 10:19:28 -0800, Wilm Boerhout >> >> wrote: >> >>> Tom Linden vaguely mentioned on 12-11-2008 18:50: >>> >>>> I have wondered before about the legality of limiting it to CHARON. >>>> It >>>> seems like restraint of trade, which is illegal. >>> >>> Or not. Your original VMS license only allows you to run VMS on a >>> choice >>> of VAX, Alpha or Itanium all from HP (see the VMS SPD). That appears to >>> be/have been legal. >>> >>> But first of all, it's HP who grants the transfer licenses, so they >>> should speak up of what their intention is for commercial use (CHARON >>> only or any emulator). If it is then contended in a particular >>> jurisdiction, let the judge decide. Make some lawyers happy... >> >> There is a distinction. The HW, OS and CHARON all are third party, so >> if someone comes along with HW, OS-X and BrandX emulator. Then what? >> IOW, >> HP is selling a piece of software for a platform it does not provide. I >> don't >> think they can deny sale of said software to another platform. They >> can, >> of >> course, refuse to provide support. > > While I agree with you in principle, Tom, I don't think you really > have an argument. They don't "sell" the software, they license it > for use on platforms of their choice. Choosing not to offer a license > for a particular platform does not restrict anyone's ability to "trade" > in SIMH. I don't think that matters, whether you license or sell. > bill > -- PL/I for OpenVMS www.kednos.com ------------------------------ Date: 13 Nov 2008 12:43:30 GMT From: billg999@cs.uofs.edu (Bill Gunshannon) Subject: Re: Good news: HP has retaken the VMS trademark ! Message-ID: <6o2lniF1h6gvU1@mid.individual.net> In article , helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: > In article , > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: > >> In article <6d5b0$472b798b$cef8887a$5047@TEKSAVVY.COM>, JF Mezei writes: >> > >> > In this case, it is moot. Just as the case for the use of "ALL-IN-1" for >> > its printers, HP can use "VMS" for whatever it wants since it owns the >> > company that owned that trademark. >> >> Actually it didn't. HP owns "OpenVMS" but the VMS trademark was >> allowed to expire before VMS bought Compaq. There are now other >> software systems trademarked as "VMS" which would prevent HP >> from using "VMS" as a software trademark, unless they first track >> down and purchase that from the current owner. > > I think only if there is possibility of confusion, such as another OS > called VMS. I don't know if application software would be deemed close > enough to an OS to cause confusion. There was Apple Computer and Apple > Records---no problem, until the former started to deal in music. There > is a vacuum cleaner called the VAX (made famous by the magazine ad > "nothing sucks like a VAX"). The German Stock Exchange has some > software called Eurex; that name is also used for a mark of trousers > (and, I believe, for Venetian blinds or something similar). (Maybe > Oliver Sacks has written a book about the man who mistook his software > for trousers or vice versa, but that's not a mistake a healthy person > would make, except perhaps someone who programs by the seat of his > pants. On the other hand, "software" and "underwear" are linguistically > rather close. I won't talk about hardware in this respect, much less > firmware or wetware.) Forgetting the legal implications, there are other problems with their having let the mark expire and then trying to retake it. The US Government (specifically DOD) currently have product known far and wide as VMS. It is computer related. Now, the government didn't register a trademark but that is irrelevant, because someone registering it now could not prevent their continued use and that use would definitely dilute the effectiveness of the mark among someone who you would expect was a major potential customer. Face it, the mark VMS, as we knew it, is dead. 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: Thu, 13 Nov 2008 08:09:03 -0800 (PST) From: Rich Jordan Subject: Re: Most impressive VAX installations Message-ID: <3423f919-20c7-4e7a-9bbf-ddc17d31e4bf@35g2000pry.googlegroups.com> On Nov 12, 8:49=A0pm, legalize+jee...@mail.xmission.com (Richard) wrote: > [Please do not mail me a copy of your followup] > > johnwalla...@yahoo.co.uk spake the secret code > <3c395fbc-074b-419b-9562-482b0b6fe...@q9g2000yqc.googlegroups.com> thusly= : > > >Meanwhile a Windows licence isn't noticeably cheaper today than it was > >when Win3.11 first came out in the mid 90s [1] (around $100 or so?), > > This is as silly as the memory comparison. =A0You're not buying a Win 3.1= 1 > license when you put out that $100 today. =A0You're getting a *lot* more > software and functionality for that same $100. =A0Remember that 3.11 > doesn't even include TCP/IP. > -- > "The Direct3D Graphics Pipeline" -- DirectX 9 draft available for downloa= d > =A0 =A0 =A0 > > =A0 =A0 =A0 =A0 Legalize Adulthood! But you are buying the 'top of the line' desktop product from that vendor in both cases. Their version of the Eddie Bauer edition Ford Explorer. The 1990 model cost X$, the 2008 model costs Y$, you're probably getting 'a lot more' in 2008 but its the same thing selling to the same group of people with the same (adjusted for time) expectations; only the difference in the real value of the dollar really matters. What do you "need" in a desktop system. Then (for some) it was 3.11. Now (for some) its Vista Ultimate. It is the same product adjusted for time, just more features made standard due to market forces. ------------------------------ Date: Thu, 13 Nov 2008 10:35:38 -0700 From: Keith Parris Subject: Re: Most impressive VAX installations Message-ID: Phillip Helbig---remove CLOTHES to reply wrote: > In article , moroney@world.std.spaamtrap.com > (Michael Moroney) writes: >>Not a single system, but I came across comments in the disk shadowing code >>for a bugfix where a byte field was being treated as a negative number >>if it exceeded 127. That byte field was the number of nodes in a cluster, >>and it was found by a customer (I think I know who), not internal testing. >> >>Also the test followed a decrement of that field, meaning a node left the >>cluster, so the bug wouldn't have been seen unless there were 129 or >>more nodes in the cluster at some point. (Supported limit was/is 96) > > As you point out, there are reasons why that's the supported limit. Reasons, but not any good reasons, IMHO. The design limit for OpenVMS Clusters was 256 nodes, hence the presence of all those 8-bit-wide fields in the cluster internals and data structures. The supported cluster node count has varied over time, with values of 13, 26, and 42 at various points in time. The supported limit at present is still 96. That number was based on a bug in $MOUNT/CLUSTER way back in VMS version 5.2 that caused crashes at above 96 nodes, when C.W. Hobbs (of Cluster Wide Process Services, or "CWPS" fame) cobbled together enough systems to test that many. And although that bug has long since been fixed (5.5-2), the supported number has never been raised. Qualification and test expense, coupled with a lack of customer pressure for higher supported numbers, is why larger configurations haven't been supported since then. It's not based on any inability of the product itself -- any known limits at 96 or 128 nodes (and even one at 224 nodes) have been removed. Perhaps with virtualization it may become practical to economically test OpenVMS Clusters with as many as 255 nodes. ------------------------------ Date: Thu, 13 Nov 2008 11:55:58 GMT From: =?ISO-8859-1?Q?Jan-Erik_S=F6derholm?= Subject: OpenSSL 1.3 and smime ? Message-ID: Hi. I've installed the OpenSSL 1.3 kit for VMS from HP. I'm not trying to send smime mails from the VMS box and read/decrypt them in Thunderbird. I have installed "KeyManager" in Firefox to produce key-pairs and export/import the cert into Thunderbird. I also move and install an CSR "Certificate Signing Request" using the SSL$CERT_TOOL in OpenSSL. So far so good. The mails are created and sent over to my mail account read by Thunderbird. But there something goes wrong. Thunderbird says it can't find the right certificate to decrypt the mail: > Thunderbird cannot decrypt this message > > The sender encrypted this message to you using one > of your digital certificates; however, Thunderbird > was not able to find this certificate and > corresponding private key. Now, has anyone else used the smime tool in the OpenSSL 1.3 kit for VMS ? If so, I'd be interested in a short walk-through on how to get it working... I can supply further details about what I've done so far if it would be of any help. Jan-Erik. ------------------------------ Date: Thu, 13 Nov 2008 01:38:25 -0800 (PST) From: Rob Subject: Re: Peek&Spy anyone ? Message-ID: On Nov 12, 11:50=A0am, Jan-Erik S=F6derholm wrote: > Rob wrote: > > On Nov 12, 8:55 am, Jan-Erik S=F6derholm > > wrote: > >> VAXman- @SendSpamHere.ORG wrote: > >>> In article , =3D?ISO-8859-1?Q?Ja= n-Erik_S=3DF6derholm?=3D writes: > >>>> Quick update... > >>>> I've now runed a few tests with Peek&Spy. > >>>> The major problem right now is that is doesn't log > >>>> data going *in* to VMS if it isn't echo'ed back out. > >>>> Some security concern (not logging non-echoed passwords) > >>>> that I can understand, but that breaks my use of it... > >>>> What I need is something that would give the same logging > >>>> as if I had attached a RS232/serial line "sniffer" to > >>>> the actual terminal server port. That is, log everything > >>>> going in and out no matter the VMS terminal/device settings. > >>>> That's about Peek&Spy... > >>>> About CONTRL... > >>>> I'm currently trying to find a manual for CONTRL, the > >>>>www.raxco.comsiteis *very* succesfull in hiding everything > >>>> having anything with VMS to do. They only wants to sell > >>>> Disk defraggers for Windows, as far as I understand from > >>>> the actual web site. > >>>> If someone could show me how to find anything VMS related > >>>> by clicking fromhttp://www.raxco.com/thatwould be nice. > >>>> Late update... !! > >>>> While writing this I got a couple of manuals mailed > >>>> directly from Raxco... Fine then, now into the manuals. > >>>> I will first see if the manual tells me something about > >>>> similar restrictions on the logging as Peek&Spy has. If not > >>>> I'll try in "live" and we'll see... > >>>> Thanks for listening... > >>>> Jan-Erik. > >>> If you have any specific questions re CONTRL, feel free to ask me. > >> Fine. The manual didn't say anything about blocking non-echoed > >> data (in the say Peek&Spy does). I'll probably test it later > >> today... > > >> Jan-Erik.- Hide quoted text - > > >> - Show quoted text - > > > Jan, we use CONTRL for watching the traffic out of our barcode readers > > and conveyor control boxes, and everything is viewable. > > > Like Vaxman, let me know if you need any help with CONTRL. > > > Rob. > > OK, fine, here comes the first one... :-) > > I have installed CONTRL on a test box (Alpha XP900). > > I opened two telnet/ssh terminals and entered > "CONTRL WATCH /NONOTIFY " > and everything runs just fine. A MONITOR SYS session > is perfectly copied to the watching terminal... > > Now, I tried to enter SET TERM/NOECHO on the watched > terminal, and all input disapears from the watching > terminal. This worries me a little, since as far as > I know, all input from the production equipment is > not echoed back. Is there something to make WATCH > see/show non-echoed input ? > > Jan-Erik.- Hide quoted text - > > - Show quoted text - CONTRL hooks into the session at the terminal device driver level, so whatever you do to the terminal for the user is replicated for CONTRL. If you set NOECHO and the user doesn't see anything, then neither will CONTRL. If you want to capture that type of data, e.g. passwords, then you'll have to capture the packets instead. Rob. ------------------------------ Date: Thu, 13 Nov 2008 11:31:49 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Peek&Spy anyone ? Message-ID: <00A828E1.F8D40AB4@SendSpamHere.ORG> In article , "Tom Simpson" writes: >I used the Raxco product for years and worked fine for most everything. >However, one thing it did have a problem with was FMS screens, which was >what our main application used. When monitoring, you could see everything >that the user saw except what was on the screen before you started your >monitoring session (which is logical). A control-w would not force the FMS I thought CTRL-R was refresh under FMS. >graphics to refresh. If the user moved off the screen to a different >screen, you would see everything. It was an aggravation but not a >show-stopper. So, you'd like to hear a symphony even though you weren't present when it was performed and nobody was there to recording it? Hmm. So, your enhancement request is that CONTRL would prognosticate? With the WINDOW portion of the RaxcoSupport product, this too can be realized. There are myriad other benefits and features using the WINDOW utility. >The Peek&Spy product handled the monitoring task much differently could >display what was on the user's FMS screen before you started your monitor >session (claimed the vendor). This additional functionality significantly >increased the resources required to run the application (from my discussions >with the vendor). Unfortunately, I changed jobs before I actually had a >chance to install and test the Peek&Spy product, but it sounded like would >solve the one issue we had with Raxco's product. No difference. They're buffering all terminal I/O in anticipation of someone wanting to see some terminal screen's contents prior to someone actually showing interest in seeing it. This feature comes at a steeper price both monetarily and as a performance hit. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM ... pejorative statements of opinion are entitled to constitutional protection no matter how extreme, vituperous, or vigorously expressed they may be. (NJSC) Copr. 2008 Brian Schenkenberger. Publication of _this_ usenet article outside of usenet _must_ include its contents in its entirety including this copyright notice, disclaimer and quotations. ------------------------------ Date: Thu, 13 Nov 2008 11:36:58 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: Peek&Spy anyone ? Message-ID: <00A828E2.B1374121@SendSpamHere.ORG> In article , Rob writes: >On Nov 12, 11:50=A0am, Jan-Erik S=F6derholm >wrote: >> Rob wrote: >> > On Nov 12, 8:55 am, Jan-Erik S=F6derholm >> > wrote: >> >> VAXman- @SendSpamHere.ORG wrote: >> >>> In article , =3D?ISO-8859-1?Q?Ja= >n-Erik_S=3DF6derholm?=3D writes: >> >>>> Quick update... >> >>>> I've now runed a few tests with Peek&Spy. >> >>>> The major problem right now is that is doesn't log >> >>>> data going *in* to VMS if it isn't echo'ed back out. >> >>>> Some security concern (not logging non-echoed passwords) >> >>>> that I can understand, but that breaks my use of it... >> >>>> What I need is something that would give the same logging >> >>>> as if I had attached a RS232/serial line "sniffer" to >> >>>> the actual terminal server port. That is, log everything >> >>>> going in and out no matter the VMS terminal/device settings. >> >>>> That's about Peek&Spy... >> >>>> About CONTRL... >> >>>> I'm currently trying to find a manual for CONTRL, the >> >>>>www.raxco.comsiteis *very* succesfull in hiding everything >> >>>> having anything with VMS to do. They only wants to sell >> >>>> Disk defraggers for Windows, as far as I understand from >> >>>> the actual web site. >> >>>> If someone could show me how to find anything VMS related >> >>>> by clicking fromhttp://www.raxco.com/thatwould be nice. >> >>>> Late update... !! >> >>>> While writing this I got a couple of manuals mailed >> >>>> directly from Raxco... Fine then, now into the manuals. >> >>>> I will first see if the manual tells me something about >> >>>> similar restrictions on the logging as Peek&Spy has. If not >> >>>> I'll try in "live" and we'll see... >> >>>> Thanks for listening... >> >>>> Jan-Erik. >> >>> If you have any specific questions re CONTRL, feel free to ask me. >> >> Fine. The manual didn't say anything about blocking non-echoed >> >> data (in the say Peek&Spy does). I'll probably test it later >> >> today... >> >> >> Jan-Erik.- Hide quoted text - >> >> >> - Show quoted text - >> >> > Jan, we use CONTRL for watching the traffic out of our barcode readers >> > and conveyor control boxes, and everything is viewable. >> >> > Like Vaxman, let me know if you need any help with CONTRL. >> >> > Rob. >> >> OK, fine, here comes the first one... :-) >> >> I have installed CONTRL on a test box (Alpha XP900). >> >> I opened two telnet/ssh terminals and entered >> "CONTRL WATCH /NONOTIFY " >> and everything runs just fine. A MONITOR SYS session >> is perfectly copied to the watching terminal... >> >> Now, I tried to enter SET TERM/NOECHO on the watched >> terminal, and all input disapears from the watching >> terminal. This worries me a little, since as far as >> I know, all input from the production equipment is >> not echoed back. Is there something to make WATCH >> see/show non-echoed input ? >> >> Jan-Erik.- Hide quoted text - >> >> - Show quoted text - > >CONTRL hooks into the session at the terminal device driver level, so >whatever you do to the terminal for the user is replicated for CONTRL. > >If you set NOECHO and the user doesn't see anything, then neither will >CONTRL. I can make things like that change quite easily. ;) >If you want to capture that type of data, e.g. passwords, then you'll >have to capture the packets instead. ...but then decoding all that gets ugly. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM ... pejorative statements of opinion are entitled to constitutional protection no matter how extreme, vituperous, or vigorously expressed they may be. (NJSC) Copr. 2008 Brian Schenkenberger. Publication of _this_ usenet article outside of usenet _must_ include its contents in its entirety including this copyright notice, disclaimer and quotations. ------------------------------ Date: Thu, 13 Nov 2008 11:48:09 GMT From: =?ISO-8859-1?Q?Jan-Erik_S=F6derholm?= Subject: Re: Peek&Spy anyone ? Message-ID: VAXman- @SendSpamHere.ORG wrote: > In article , Rob writes: >> On Nov 12, 11:50=A0am, Jan-Erik S=F6derholm >> wrote: >>> Rob wrote: >>>> On Nov 12, 8:55 am, Jan-Erik S=F6derholm >>>> wrote: >>>>> VAXman- @SendSpamHere.ORG wrote: >>>>>> In article , =3D?ISO-8859-1?Q?Ja= >> n-Erik_S=3DF6derholm?=3D writes: >>>>>>> Quick update... >>>>>>> I've now runed a few tests with Peek&Spy. >>>>>>> The major problem right now is that is doesn't log >>>>>>> data going *in* to VMS if it isn't echo'ed back out. >>>>>>> Some security concern (not logging non-echoed passwords) >>>>>>> that I can understand, but that breaks my use of it... >>>>>>> What I need is something that would give the same logging >>>>>>> as if I had attached a RS232/serial line "sniffer" to >>>>>>> the actual terminal server port. That is, log everything >>>>>>> going in and out no matter the VMS terminal/device settings. >>>>>>> That's about Peek&Spy... >>>>>>> About CONTRL... >>>>>>> I'm currently trying to find a manual for CONTRL, the >>>>>>> www.raxco.comsiteis *very* succesfull in hiding everything >>>>>>> having anything with VMS to do. They only wants to sell >>>>>>> Disk defraggers for Windows, as far as I understand from >>>>>>> the actual web site. >>>>>>> If someone could show me how to find anything VMS related >>>>>>> by clicking fromhttp://www.raxco.com/thatwould be nice. >>>>>>> Late update... !! >>>>>>> While writing this I got a couple of manuals mailed >>>>>>> directly from Raxco... Fine then, now into the manuals. >>>>>>> I will first see if the manual tells me something about >>>>>>> similar restrictions on the logging as Peek&Spy has. If not >>>>>>> I'll try in "live" and we'll see... >>>>>>> Thanks for listening... >>>>>>> Jan-Erik. >>>>>> If you have any specific questions re CONTRL, feel free to ask me. >>>>> Fine. The manual didn't say anything about blocking non-echoed >>>>> data (in the say Peek&Spy does). I'll probably test it later >>>>> today... >>>>> Jan-Erik.- Hide quoted text - >>>>> - Show quoted text - >>>> Jan, we use CONTRL for watching the traffic out of our barcode readers >>>> and conveyor control boxes, and everything is viewable. >>>> Like Vaxman, let me know if you need any help with CONTRL. >>>> Rob. >>> OK, fine, here comes the first one... :-) >>> >>> I have installed CONTRL on a test box (Alpha XP900). >>> >>> I opened two telnet/ssh terminals and entered >>> "CONTRL WATCH /NONOTIFY " >>> and everything runs just fine. A MONITOR SYS session >>> is perfectly copied to the watching terminal... >>> >>> Now, I tried to enter SET TERM/NOECHO on the watched >>> terminal, and all input disapears from the watching >>> terminal. This worries me a little, since as far as >>> I know, all input from the production equipment is >>> not echoed back. Is there something to make WATCH >>> see/show non-echoed input ? >>> >>> Jan-Erik.- Hide quoted text - >>> >>> - Show quoted text - >> CONTRL hooks into the session at the terminal device driver level, so >> whatever you do to the terminal for the user is replicated for CONTRL. >> >> If you set NOECHO and the user doesn't see anything, then neither will >> CONTRL. > > I can make things like that change quite easily. ;) > > >> If you want to capture that type of data, e.g. passwords, then you'll >> have to capture the packets instead. > > ...but then decoding all that gets ugly. Just so that there are no misunderstandings... :-) It's *not* passwords I want to "see", it's barcode data sent from barcode readers connected to ports with /noecho. Anyway, I just checked a little in the actual systems, and it seems as we do not use /noecho on the actual ports, I thought we did... I have to run some test on the live prod systems, I guess... :-) Jan-Erik. ------------------------------ Date: Thu, 13 Nov 2008 10:36:44 -0800 (PST) From: shadoooo Subject: Seagate ST31051N on MicroVAX 3100 Message-ID: <88f5e379-94dd-4b64-8c23-1905c4bcb880@z6g2000pre.googlegroups.com> I acquired two Seagate ST31051N to use it on my VAX boxes. I used TEST 75 on the MicroVAX 3100, disk erased correctly. Then I tried to install a VMS 5.5 from another bootable disk I prepared with STABACKIT, but when it tries to INIT/MOUNT it, I received a drive fatal error. I read the ng and found the problem about ARRE ARWE bits, so downloaded the SCU application, connected the drive to the PC and disabled these bits. Then I retried with install operation, but it is the same as before, even with ARRE ARWE disabled. So I tried to disable also EER (early error recovery) but it's the same. I tried to mount the disk also on my 4000-100A, but also there no way. I don't understand why it doesn't work. Any idea? Maybe I need to redo TEST75 after the bit disable operation? ------------------------------ Date: 13 Nov 2008 18:44:25 +0100 From: eplan@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: Re: Seamonkey browser port for Alpha now available Message-ID: <491c7589$1@news.langstoeger.at> In article , Rich Jordan writes: >From OpenVMS.ORG > >http://www.openvms.org/stories.php?story=08/10/24/9322255 > >I may try to get this downloaded this weekend to try. Has anyone used >it or the previous itanium port yet? Nope. I just tried it (on Alpha V8.3) and after a few minutes I can tell: 1) I just don't like to call it CSWB a) it is not from CPQ (why not call it SWB, from HPQ?) b) it is now not possible to have it parallel to Mozilla c) it has a version number which is lower of what was installed before (V1.7-13) 2) It looks like (CSWB/) MOZILLA (with improvements like roaming User) a) even the initial bookmarks (like latest builds) are old (2005/2006) 3) It is still Beta (JAVA doesn't work - as noted) -- Peter "EPLAN" LANGSTÖGER 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 Nov 2008 03:21:34 -0500 From: JF Mezei Subject: Re: SFF (Send From File) Utility Message-ID: <000dbbab$0$9192$c3e8da3@news.astraweb.com> Michael D. Ober wrote: > To: mdo@wakeassoc.com,obermd@alum.mit.edu,obermd@sprintmail.com Your test is not very good since all 3 recipients are on different SMTP servers. But you are right. If you have defined apollo.wakeassoc as a gateway SMTP server (and it seems to be working since the message to all 3 recipients was sent to apollo), one would expect the message to go to apollo once and let apollo distribute the copy 3 times to each different domain. You should try: To: user1@domain1.com, user2@domain1.com, user3@hp.com as well as: To: user1@domain1.com, user2@domain1.com, user3@domain1.com And see if it sends the message twice or 3 times in the first case, and in second case, if it does send only one copy of message out. if you send a message to an invalid domain name, do you get a non-delivery notification from jaguar or from apollo ? ------------------------------ Date: Thu, 13 Nov 2008 10:49:52 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: strange disk state Message-ID: Do to an unfortunate unplanned accidental reboot of two nodes in my cluster, something strange happened. There are three nodes in the cluster. The problem is with a member of the system-disk shadow set on one of the nodes. Like all system-disk shadow sets, both members have a direct connection only to their own node. From two nodes, the shadow set looks like this: Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt DSA122: Mounted 0 OVMSVAXSYS_2 2137348 347 3 $22$DKA300: (DANEEL) ShadowSetMember 0 (member of DSA122:) $22$DKA700: (DANEEL) ShadowSetMember 0 (member of DSA122:) $ sh dev $22$DKA300: Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt DSA122: Mounted 0 OVMSVAXSYS_2 2137348 347 3 $22$DKA300: (DANEEL) ShadowSetMember 0 (member of DSA122:) $22$DKA700: (DANEEL) ShadowSetMember 0 (member of DSA122:) $ sh dev $22$DKA700: Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt DSA122: Mounted 0 OVMSVAXSYS_2 2137348 347 3 $22$DKA300: (DANEEL) ShadowSetMember 0 (member of DSA122:) $22$DKA700: (DANEEL) ShadowSetMember 0 (member of DSA122:) So far, so good. From one of the nodes, however (NOT the node DANEEL, whose system disk is the one showing the problem), I see this: $ sh dev dsa122 Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt DSA122: Mounted 0 OVMSVAXSYS_2 2137348 1 3 $22$DKA700: (DANEEL) ShadowSetMember 0 (member of DSA122:) $ sh dev $22$DKA300: Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt $22$DKA300: (DANEEL) 0 (member of ) $ sh dev $22$DKA700: Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt DSA122: Mounted 0 OVMSVAXSYS_2 2137348 1 3 $22$DKA700: (DANEEL) ShadowSetMember 0 (member of DSA122:) In other words, only one member is showing up. What's up with $22$DKA300:? From the other two nodes, $ SH DEV/FUL gives the standard output: $ sh dev $22$DKA300/fu Disk DSA122:, device type RZ28, is online, mounted, file-oriented device, shareable, available to cluster, error logging is enabled. Error count 0 Operations completed 376024 Owner process "" Owner UIC [SYSTEM] Owner process ID 00000000 Dev Prot S:RWPL,O:RWPL,G:R,W Reference count 276 Default buffer size 512 Total blocks 4110480 Sectors per track 86 Total cylinders 2988 Tracks per cylinder 16 Volume label "OVMSVAXSYS_2" Relative volume number 0 Cluster size 4 Transaction count 347 Free blocks 2137348 Maximum files allowed 411048 Extend quantity 5 Mount count 3 Mount status System Cache name "_DSA122:XQPCACHE" Extent cache size 64 Maximum blocks in extent cache 213734 File ID cache size 64 Blocks currently in extent cache 0 Quota cache size 0 Maximum buffers in FCP cache 1250 Volume owner UIC [1,1] Vol Prot S:RWCD,O:RWCD,G:RWCD,W:RWCD Volume Status: ODS-2, subject to mount verification, protected subsystems enabled, file high-water marking, write-through caching enabled. Volume is also mounted on GLADIA, ELIJAH. Disk $22$DKA300:, device type RZ28, is online, member of shadow set DSA122:, served to cluster via MSCP Server, error logging is enabled. Error count 0 Shadow member operation count 139079 Allocation class 22 Disk $22$DKA700:, device type RZ28, is online, member of shadow set DSA122:, served to cluster via MSCP Server, error logging is enabled. Error count 0 Shadow member operation count 138237 Allocation class 22 From the problematic node, I get: $ sh dev $22$DKA300/fu Disk $22$DKA300:, device type RZ28, is online, member of shadow set, error logging is enabled. Error count 0 Shadow member operation count 11228 Host name "DANEEL" Host type, avail VAX 4000-105A, yes Allocation class 22 From the other two nodes, I get the standard output: $ sh dev dk Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt $22$DKA200: (DANEEL) ShadowSetMember 18 (member of DSA520:) $22$DKA300: (DANEEL) ShadowSetMember 0 (member of DSA122:) $22$DKA400: (DANEEL) Mounted wrtlck 0 OVMSDOC071 61179 1 3 $22$DKA500: (DANEEL) ShadowSetMember 0 (member of DSA510:) $22$DKA700: (DANEEL) ShadowSetMember 0 (member of DSA122:) $33$DKA200: (GLADIA) Online 0 $33$DKA300: (GLADIA) ShadowSetMember 0 (member of DSA520:) $33$DKA400: (GLADIA) Online 0 $33$DKB100: (GLADIA) ShadowSetMember 0 (member of DSA133:) $33$DKB600: (GLADIA) ShadowSetMember 0 (member of DSA133:) $44$DKA100: (ELIJAH) ShadowSetMember 0 (member of DSA144:) $44$DKA200: (ELIJAH) Mounted 0 SCRATCH_4 165972 1 3 $44$DKA300: (ELIJAH) Mounted 0 DATA 1907185 1 3 $44$DKA400: (ELIJAH) ShadowSetMember 0 (member of DSA510:) $44$DKA600: (ELIJAH) ShadowSetMember 0 (member of DSA144:) $131$DKA300: (NADILA) HostUnavailable 0 $131$DKA500: (NADILA) HostUnavailable 0 However, from the problematic node, for the problematic disk, I get: $22$DKA300: (DANEEL) 0 (member of ..*..._$22$DKA400:... ......DANEEL.........VAX VMS ..VAX 4000-105A.................................... .................) Mote that ..*..._$22$DKA400: is the CD-ROM drive! Any ideas how this state of affairs could have arisen? Is there a clean way to correct it, other than a cluster reboot? Everything seems to work fine. I mount the system-disk shadow sets on all nodes mainly so that they are visible from all nodes, so that I could run $ DIFF or whatever. If the one member of DSA122: (system-disk shadow set on DANEEL) becomes unreachable from the problematic node (ELIJAH), then I can't access that shadow set from that node. This shouldn't be any problem, except for interactive work, as mentioned above. However, I would still like to a) understand the problem and b) correct it with a minimum of trouble. ------------------------------ Date: Thu, 13 Nov 2008 03:07:47 -0800 (PST) From: "winston19842005@yahoo.com" Subject: Re: strange disk state Message-ID: On Nov 13, 5:49=A0am, hel...@astro.multiCLOTHESvax.de (Phillip Helbig--- remove CLOTHES to reply) wrote: > Do to an unfortunate unplanned accidental reboot of two nodes in my > cluster, something strange happened. =A0There are three nodes in the > cluster. =A0The problem is with a member of the system-disk shadow set on > one of the nodes. =A0Like all system-disk shadow sets, both members have = a > direct connection only to their own node. =A0From two nodes, the shadow > set looks like this: > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > DSA122: =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Mounted =A0 =A0 =A0 =A0 =A0 =A0 = =A00 =A0OVMSVAXSYS_2 =A0 2137348 =A0 347 =A0 3 > $22$DKA300: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $ sh dev $22$DKA300: > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > DSA122: =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Mounted =A0 =A0 =A0 =A0 =A0 =A0 = =A00 =A0OVMSVAXSYS_2 =A0 2137348 =A0 347 =A0 3 > $22$DKA300: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $ sh dev $22$DKA700: > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > DSA122: =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Mounted =A0 =A0 =A0 =A0 =A0 =A0 = =A00 =A0OVMSVAXSYS_2 =A0 2137348 =A0 347 =A0 3 > $22$DKA300: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > > So far, so good. =A0From one of the nodes, however (NOT the node DANEEL, > whose system disk is the one showing the problem), I see this: > > $ sh dev dsa122 > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > DSA122: =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Mounted =A0 =A0 =A0 =A0 =A0 =A0 = =A00 =A0OVMSVAXSYS_2 =A0 2137348 =A0 =A0 1 =A0 3 > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $ sh dev $22$DKA300: > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > $22$DKA300: =A0 (DANEEL) =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 0 = =A0(member of ) > $ sh dev $22$DKA700: > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > DSA122: =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Mounted =A0 =A0 =A0 =A0 =A0 =A0 = =A00 =A0OVMSVAXSYS_2 =A0 2137348 =A0 =A0 1 =A0 3 > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > > In other words, only one member is showing up. =A0What's up with > $22$DKA300:? =A0From the other two nodes, $ SH DEV/FUL gives the standard > output: > > $ sh dev $22$DKA300/fu > > Disk DSA122:, device type RZ28, is online, mounted, file-oriented device, > =A0 =A0 shareable, available to cluster, error logging is enabled. > > =A0 =A0 Error count =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0 =A0Opera= tions completed =A0 =A0 =A0 =A0 =A0 =A0 376024 > =A0 =A0 Owner process =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 "" =A0 =A0Owner UIC= =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0[SYSTEM] > =A0 =A0 Owner process ID =A0 =A0 =A0 =A000000000 =A0 =A0Dev Prot =A0 =A0 = =A0 =A0 =A0 =A0S:RWPL,O:RWPL,G:R,W > =A0 =A0 Reference count =A0 =A0 =A0 =A0 =A0 =A0 =A0276 =A0 =A0Default buf= fer size =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 512 > =A0 =A0 Total blocks =A0 =A0 =A0 =A0 =A0 =A0 4110480 =A0 =A0Sectors per t= rack =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A086 > =A0 =A0 Total cylinders =A0 =A0 =A0 =A0 =A0 =A0 2988 =A0 =A0Tracks per cy= linder =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A016 > > =A0 =A0 Volume label =A0 =A0 =A0"OVMSVAXSYS_2" =A0 =A0Relative volume num= ber =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A00 > =A0 =A0 Cluster size =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 4 =A0 =A0Transac= tion count =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 347 > =A0 =A0 Free blocks =A0 =A0 =A0 =A0 =A0 =A0 =A02137348 =A0 =A0Maximum fil= es allowed =A0 =A0 =A0 =A0 =A0 =A0411048 > =A0 =A0 Extend quantity =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A05 =A0 =A0Mount cou= nt =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 3 > =A0 =A0 Mount status =A0 =A0 =A0 =A0 =A0 =A0 =A0System =A0 =A0Cache name = =A0 =A0 =A0 =A0 =A0 "_DSA122:XQPCACHE" > =A0 =A0 Extent cache size =A0 =A0 =A0 =A0 =A0 =A0 64 =A0 =A0Maximum block= s in extent cache =A0 213734 > =A0 =A0 File ID cache size =A0 =A0 =A0 =A0 =A0 =A064 =A0 =A0Blocks curren= tly in extent cache =A0 =A0 =A00 > =A0 =A0 Quota cache size =A0 =A0 =A0 =A0 =A0 =A0 =A0 0 =A0 =A0Maximum buf= fers in FCP cache =A0 =A0 =A0 1250 > =A0 =A0 Volume owner UIC =A0 =A0 =A0 =A0 =A0 [1,1] =A0 =A0Vol Prot =A0 = =A0S:RWCD,O:RWCD,G:RWCD,W:RWCD > > =A0 Volume Status: =A0ODS-2, subject to mount verification, protected sub= systems > =A0 =A0 =A0 enabled, file high-water marking, write-through caching enabl= ed. > =A0 Volume is also mounted on GLADIA, ELIJAH. > > Disk $22$DKA300:, device type RZ28, is online, member of shadow set DSA12= 2:, > =A0 =A0 served to cluster via MSCP Server, error logging is enabled. > > =A0 =A0 Error count =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0 =A0Shado= w member operation count =A0 =A0139079 > =A0 =A0 Allocation class =A0 =A0 =A0 =A0 =A0 =A0 =A022 > > Disk $22$DKA700:, device type RZ28, is online, member of shadow set DSA12= 2:, > =A0 =A0 served to cluster via MSCP Server, error logging is enabled. > > =A0 =A0 Error count =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0 =A0Shado= w member operation count =A0 =A0138237 > =A0 =A0 Allocation class =A0 =A0 =A0 =A0 =A0 =A0 =A022 > > From the problematic node, I get: > > $ sh dev $22$DKA300/fu > > Disk $22$DKA300:, device type RZ28, is online, member of shadow set, erro= r > =A0 =A0 logging is enabled. > > =A0 =A0 Error count =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0 =A0Shado= w member operation count =A0 =A0 11228 > =A0 =A0 Host name =A0 =A0 =A0 =A0 =A0 =A0 =A0 "DANEEL" =A0 =A0Host type, = avail =A0 =A0 VAX 4000-105A, yes > =A0 =A0 Allocation class =A0 =A0 =A0 =A0 =A0 =A0 =A022 > > From the other two nodes, I get the standard output: > > $ sh dev dk > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > $22$DKA200: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 18 =A0(member of DSA5= 20:) > $22$DKA300: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $22$DKA400: =A0 (DANEEL) =A0Mounted wrtlck =A0 =A0 =A0 0 =A0OVMSDOC071 = =A0 =A0 =A0 61179 =A0 =A0 1 =A0 3 > $22$DKA500: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A510:) > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $33$DKA200: =A0 (GLADIA) =A0Online =A0 =A0 =A0 =A0 =A0 =A0 =A0 0 > $33$DKA300: =A0 (GLADIA) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A520:) > $33$DKA400: =A0 (GLADIA) =A0Online =A0 =A0 =A0 =A0 =A0 =A0 =A0 0 > $33$DKB100: =A0 (GLADIA) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A133:) > $33$DKB600: =A0 (GLADIA) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A133:) > $44$DKA100: =A0 (ELIJAH) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A144:) > $44$DKA200: =A0 (ELIJAH) =A0Mounted =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0SCRAT= CH_4 =A0 =A0 =A0 165972 =A0 =A0 1 =A0 3 > $44$DKA300: =A0 (ELIJAH) =A0Mounted =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0DATA = =A0 =A0 =A0 =A0 =A0 1907185 =A0 =A0 1 =A0 3 > $44$DKA400: =A0 (ELIJAH) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A510:) > $44$DKA600: =A0 (ELIJAH) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A144:) > $131$DKA300: =A0(NADILA) =A0HostUnavailable =A0 =A0 =A00 > $131$DKA500: =A0(NADILA) =A0HostUnavailable =A0 =A0 =A00 > > However, from the problematic node, for the problematic disk, I get: > > $22$DKA300: =A0 (DANEEL) =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 0 = =A0(member of ..*..._$22$DKA400:... > ......DANEEL.........VAX VMS ..VAX 4000-105A.............................= ....... > .................) > > Mote that ..*..._$22$DKA400: is the CD-ROM drive! > > Any ideas how this state of affairs could have arisen? > > Is there a clean way to correct it, other than a cluster reboot? > > Everything seems to work fine. =A0I mount the system-disk shadow sets on > all nodes mainly so that they are visible from all nodes, so that I > could run $ DIFF or whatever. =A0If the one member of DSA122: (system-dis= k > shadow set on DANEEL) becomes unreachable from the problematic node > (ELIJAH), then I can't access that shadow set from that node. =A0This > shouldn't be any problem, except for interactive work, as mentioned > above. =A0However, I would still like to a) understand the problem and b) > correct it with a minimum of trouble. I'd say you need to reboot the "problematic" node... The problem is something on this node is corrupted. Simple. Why? Dunno. BTW: Inquiring minds want to know: Which is the problematic node? Gladia or Elijah? ------------------------------ Date: Thu, 13 Nov 2008 06:44:06 -0600 From: VMS is Virus Free Subject: Re: strange disk state Message-ID: <528oh4957da57fios17i1936ldd9cps0kn@4ax.com> On Thu, 13 Nov 2008 10:49:52 +0000 (UTC), helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) wrote: >Is there a clean way to correct it, other than a cluster reboot? A cluster reboot should clear the condition, of course. But what you can try (perhaps on a Saturday, just in case), is to dismount the "missing" shadow member from all nodes, verify that you have a 1-member shadowset, then mount that missing member back into your shadow set. That may work. It's really hard to say because the underlying tables that describe this condition are inconsistent among your cluster member (that is NOT a good thing). If you try this and it doesn't work, I would reboot the entire cluster just to clean things up. That's why the 'do it on Saturday' suggestion. ------------------------------ Date: Thu, 13 Nov 2008 07:09:59 -0800 (PST) From: H Vlems Subject: Re: strange disk state Message-ID: <6d8a8c9c-cff6-494f-a22e-0f365eea36d5@p35g2000prm.googlegroups.com> On 13 nov, 11:49, hel...@astro.multiCLOTHESvax.de (Phillip Helbig--- remove CLOTHES to reply) wrote: > Do to an unfortunate unplanned accidental reboot of two nodes in my > cluster, something strange happened. =A0There are three nodes in the > cluster. =A0The problem is with a member of the system-disk shadow set on > one of the nodes. =A0Like all system-disk shadow sets, both members have = a > direct connection only to their own node. =A0From two nodes, the shadow > set looks like this: > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > DSA122: =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Mounted =A0 =A0 =A0 =A0 =A0 =A0 = =A00 =A0OVMSVAXSYS_2 =A0 2137348 =A0 347 =A0 3 > $22$DKA300: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $ sh dev $22$DKA300: > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > DSA122: =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Mounted =A0 =A0 =A0 =A0 =A0 =A0 = =A00 =A0OVMSVAXSYS_2 =A0 2137348 =A0 347 =A0 3 > $22$DKA300: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $ sh dev $22$DKA700: > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > DSA122: =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Mounted =A0 =A0 =A0 =A0 =A0 =A0 = =A00 =A0OVMSVAXSYS_2 =A0 2137348 =A0 347 =A0 3 > $22$DKA300: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > > So far, so good. =A0From one of the nodes, however (NOT the node DANEEL, > whose system disk is the one showing the problem), I see this: > > $ sh dev dsa122 > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > DSA122: =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Mounted =A0 =A0 =A0 =A0 =A0 =A0 = =A00 =A0OVMSVAXSYS_2 =A0 2137348 =A0 =A0 1 =A0 3 > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $ sh dev $22$DKA300: > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > $22$DKA300: =A0 (DANEEL) =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 0 = =A0(member of ) > $ sh dev $22$DKA700: > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > DSA122: =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Mounted =A0 =A0 =A0 =A0 =A0 =A0 = =A00 =A0OVMSVAXSYS_2 =A0 2137348 =A0 =A0 1 =A0 3 > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > > In other words, only one member is showing up. =A0What's up with > $22$DKA300:? =A0From the other two nodes, $ SH DEV/FUL gives the standard > output: > > $ sh dev $22$DKA300/fu > > Disk DSA122:, device type RZ28, is online, mounted, file-oriented device, > =A0 =A0 shareable, available to cluster, error logging is enabled. > > =A0 =A0 Error count =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0 =A0Opera= tions completed =A0 =A0 =A0 =A0 =A0 =A0 376024 > =A0 =A0 Owner process =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 "" =A0 =A0Owner UIC= =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0[SYSTEM] > =A0 =A0 Owner process ID =A0 =A0 =A0 =A000000000 =A0 =A0Dev Prot =A0 =A0 = =A0 =A0 =A0 =A0S:RWPL,O:RWPL,G:R,W > =A0 =A0 Reference count =A0 =A0 =A0 =A0 =A0 =A0 =A0276 =A0 =A0Default buf= fer size =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 512 > =A0 =A0 Total blocks =A0 =A0 =A0 =A0 =A0 =A0 4110480 =A0 =A0Sectors per t= rack =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A086 > =A0 =A0 Total cylinders =A0 =A0 =A0 =A0 =A0 =A0 2988 =A0 =A0Tracks per cy= linder =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A016 > > =A0 =A0 Volume label =A0 =A0 =A0"OVMSVAXSYS_2" =A0 =A0Relative volume num= ber =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A00 > =A0 =A0 Cluster size =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 4 =A0 =A0Transac= tion count =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 347 > =A0 =A0 Free blocks =A0 =A0 =A0 =A0 =A0 =A0 =A02137348 =A0 =A0Maximum fil= es allowed =A0 =A0 =A0 =A0 =A0 =A0411048 > =A0 =A0 Extend quantity =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A05 =A0 =A0Mount cou= nt =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 3 > =A0 =A0 Mount status =A0 =A0 =A0 =A0 =A0 =A0 =A0System =A0 =A0Cache name = =A0 =A0 =A0 =A0 =A0 "_DSA122:XQPCACHE" > =A0 =A0 Extent cache size =A0 =A0 =A0 =A0 =A0 =A0 64 =A0 =A0Maximum block= s in extent cache =A0 213734 > =A0 =A0 File ID cache size =A0 =A0 =A0 =A0 =A0 =A064 =A0 =A0Blocks curren= tly in extent cache =A0 =A0 =A00 > =A0 =A0 Quota cache size =A0 =A0 =A0 =A0 =A0 =A0 =A0 0 =A0 =A0Maximum buf= fers in FCP cache =A0 =A0 =A0 1250 > =A0 =A0 Volume owner UIC =A0 =A0 =A0 =A0 =A0 [1,1] =A0 =A0Vol Prot =A0 = =A0S:RWCD,O:RWCD,G:RWCD,W:RWCD > > =A0 Volume Status: =A0ODS-2, subject to mount verification, protected sub= systems > =A0 =A0 =A0 enabled, file high-water marking, write-through caching enabl= ed. > =A0 Volume is also mounted on GLADIA, ELIJAH. > > Disk $22$DKA300:, device type RZ28, is online, member of shadow set DSA12= 2:, > =A0 =A0 served to cluster via MSCP Server, error logging is enabled. > > =A0 =A0 Error count =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0 =A0Shado= w member operation count =A0 =A0139079 > =A0 =A0 Allocation class =A0 =A0 =A0 =A0 =A0 =A0 =A022 > > Disk $22$DKA700:, device type RZ28, is online, member of shadow set DSA12= 2:, > =A0 =A0 served to cluster via MSCP Server, error logging is enabled. > > =A0 =A0 Error count =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0 =A0Shado= w member operation count =A0 =A0138237 > =A0 =A0 Allocation class =A0 =A0 =A0 =A0 =A0 =A0 =A022 > > From the problematic node, I get: > > $ sh dev $22$DKA300/fu > > Disk $22$DKA300:, device type RZ28, is online, member of shadow set, erro= r > =A0 =A0 logging is enabled. > > =A0 =A0 Error count =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0 =A0Shado= w member operation count =A0 =A0 11228 > =A0 =A0 Host name =A0 =A0 =A0 =A0 =A0 =A0 =A0 "DANEEL" =A0 =A0Host type, = avail =A0 =A0 VAX 4000-105A, yes > =A0 =A0 Allocation class =A0 =A0 =A0 =A0 =A0 =A0 =A022 > > From the other two nodes, I get the standard output: > > $ sh dev dk > > Device =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Device =A0 =A0 =A0 =A0 =A0 Erro= r =A0 =A0Volume =A0 =A0 =A0 =A0 Free =A0Trans Mnt > =A0Name =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Status =A0 =A0 =A0 =A0 =A0 Co= unt =A0 =A0 Label =A0 =A0 =A0 =A0Blocks Count Cnt > $22$DKA200: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 18 =A0(member of DSA5= 20:) > $22$DKA300: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $22$DKA400: =A0 (DANEEL) =A0Mounted wrtlck =A0 =A0 =A0 0 =A0OVMSDOC071 = =A0 =A0 =A0 61179 =A0 =A0 1 =A0 3 > $22$DKA500: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A510:) > $22$DKA700: =A0 (DANEEL) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A122:) > $33$DKA200: =A0 (GLADIA) =A0Online =A0 =A0 =A0 =A0 =A0 =A0 =A0 0 > $33$DKA300: =A0 (GLADIA) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A520:) > $33$DKA400: =A0 (GLADIA) =A0Online =A0 =A0 =A0 =A0 =A0 =A0 =A0 0 > $33$DKB100: =A0 (GLADIA) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A133:) > $33$DKB600: =A0 (GLADIA) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A133:) > $44$DKA100: =A0 (ELIJAH) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A144:) > $44$DKA200: =A0 (ELIJAH) =A0Mounted =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0SCRAT= CH_4 =A0 =A0 =A0 165972 =A0 =A0 1 =A0 3 > $44$DKA300: =A0 (ELIJAH) =A0Mounted =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0DATA = =A0 =A0 =A0 =A0 =A0 1907185 =A0 =A0 1 =A0 3 > $44$DKA400: =A0 (ELIJAH) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A510:) > $44$DKA600: =A0 (ELIJAH) =A0ShadowSetMember =A0 =A0 =A00 =A0(member of DS= A144:) > $131$DKA300: =A0(NADILA) =A0HostUnavailable =A0 =A0 =A00 > $131$DKA500: =A0(NADILA) =A0HostUnavailable =A0 =A0 =A00 > > However, from the problematic node, for the problematic disk, I get: > > $22$DKA300: =A0 (DANEEL) =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 0 = =A0(member of ..*..._$22$DKA400:... > ......DANEEL.........VAX VMS ..VAX 4000-105A.............................= ....... > .................) > > Mote that ..*..._$22$DKA400: is the CD-ROM drive! > > Any ideas how this state of affairs could have arisen? > > Is there a clean way to correct it, other than a cluster reboot? > > Everything seems to work fine. =A0I mount the system-disk shadow sets on > all nodes mainly so that they are visible from all nodes, so that I > could run $ DIFF or whatever. =A0If the one member of DSA122: (system-dis= k > shadow set on DANEEL) becomes unreachable from the problematic node > (ELIJAH), then I can't access that shadow set from that node. =A0This > shouldn't be any problem, except for interactive work, as mentioned > above. =A0However, I would still like to a) understand the problem and b) > correct it with a minimum of trouble. I agree that a reboot ought to solve the problem. What worries me is how the problem was created. Both the cluster software and the shadow disks are prerequisites for high availability solutions and even though corruption of a datastructure may be a possible cause for the problem, the point is that it shouldn't have occurred at all! Phillip, are you sure you didn't make a typo while mounting a shadowset? Or gave the CDrom a different scsi id? Hans ------------------------------ Date: Thu, 13 Nov 2008 16:01:22 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: strange disk state Message-ID: In article , "winston19842005@yahoo.com" writes: Note: Without the KQP macro discussed in another thread, your post would have been completely unreadable, mainly because all spaces following another space were replaced with =A0. > I'd say you need to reboot the "problematic" node... > The problem is something on this node is corrupted. Simple. Why? > Dunno. I'll give it a try---that's not too much trouble. > BTW: Inquiring minds want to know: Which is the problematic node? > Gladia or Elijah? ELIJAH. The strange disk is on DANEEL. DANEEL and GLADIA had the unplanned reboot. I don't know why they rebooted and ELIJAH didn't. GLADIA is an ALPHA; ELIJAH and DANEEL are VAXes. That's good: The DISK$USER shadow set has 2 members, one with a direct connection to ELIJAH, one to DANEEL. By dismounting and remounting the member on ELIJAH from GLADIA (an ALPHA), I will get only a minicopy instead of a copy. ------------------------------ Date: Thu, 13 Nov 2008 16:07:07 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: strange disk state Message-ID: In article <6d8a8c9c-cff6-494f-a22e-0f365eea36d5@p35g2000prm.googlegroups.com>, H Vlems writes: > I agree that a reboot ought to solve the problem. What worries me is > how the problem was created. > Both the cluster software and the shadow disks are prerequisites for > high availability solutions and even though > corruption of a datastructure may be a possible cause for the problem, > the point is that it shouldn't have occurred at all! > Phillip, are you sure you didn't make a typo while mounting a > shadowset? Or gave the CDrom a different scsi id? Absolutely sure. What happened was, while I was in the hospital, Mozilla, which my wife was using (on NADILA, a satellite node, the trusty 5305/1200) got hung. Rather than trying to kill the process, or shut down NADILA from the console, she just switched NADILA off. This apparently caused DANEEL and GLADIA (though not ELIJAH) to reboot. At boot time, the disks are mounted by procedures in the startup sequence which I haven't changed in months. I am NOT current on patches. Now that the worst of my illness is over but I am not back at work yet, I plan to catch up. The ALPHAs are at 7.3-2 and the VAXes at 7.3. Needing hours for a full shadow copy, it's not practical to test all unpretty scenarios (rebooting a node without shutting applications down and/or dismounting disks, just switching the power on and off etc---both for one or many or all nodes). Planned reboots, either of one node or of the entire cluster, work perfectly. ------------------------------ Date: Thu, 13 Nov 2008 09:14:27 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: Synchronizing SYSUAF between independent machines Message-ID: In article , Ramon Jimenez writes: > Hello > > We have two HP Integrity boxes (OS Version V8.3-1H1) > > One of them must be keept as a failover, so we need to keep > information synchronized between both machines. First, why not a proper cluster rather than failover? > The application and data has been confined into a Logical Disk, so we > just dismount the shadow and copy the files into the other machine. If > we need to switch over we just need to mount the volumes on the > failover server. > > We also need to keep synchronized the sysuaf.dat (and related files > rights.dat...). > > Which could be the best method to do it? Since you are copying lots of files anyway, why not just copy SYSUAF as well? Personally, I long ago moved SYSUAF and friends off the system disk. The reason is simple: this file contains the "personality" of the cluster, and as such is something I want to keep even though I might swap out some or all machines in the cluster. Also, being on a disk accessible to all members, I have no problem with synchronisation. ------------------------------ Date: Thu, 13 Nov 2008 10:42:46 +0100 From: Johnny Billquist Subject: Re: VAX-11/785 microdiagnostics help requested Message-ID: iking@killthewabbit.org skrev: > Progress: > > I've determined that the microdiagnostic monitor loads into LSI-11 > memory just fine. However, its first block is all zeros - I've > verified in the listing that this is correct. So it loads a block of > zeros at 70000 (octal), with a jump instruction into the code at > 71000. After loading this file, the console jumps to 70000, where of > course it halts. I note that the value in R1 is 70000. Does anyone > have any thoughts on this? > > I've also tried loading the diagnostic supervisor. This wasn't > working. Then, while digging into the engineering drawings, I > realized that the TOY clock gets its power from the battery backup > unit, whether or not its batteries work (which, after all these years, > ours don't). Contrast this with any given MicroVAX, where the battery > backup is just that, and if the machine is powered up the TOY is > powered off the internal mains. Not on a VAX-11/78x.... After > reconnecting the BBU, I can successfully load and run the DS, and the > VAX autosizer reports all my devices - hooray! But I'd still like to > run some diagnostics.... Sounds somewhat similar to the 86x0 machines. If the BBU isn't ok, the TOY won't work, and the machine will not boot at all. You just get dumped into uODT of the T11. There is a way of cheating the 86x0 machines when the BBU don't work anymore (which is probably common if you have one, since the batteries are probably dead). The workaround is to strap a signal on the backplane, so that it looks like you have a power ok-signal from the BBU. > Does anyone out there have a copy of RX8 from the VAX-11 diag set > (also known as AS-E161N-DF), and RX25, AS-F748H-DE? I would be > incredibly grateful if someone can share a copy with me. I can > restore a disk image, or would be happy to pay for shipping both ways > of a couple of floppies for copying. I really want to run some > serious diagnostics on these olde beastes before launching into a VMS > install.... Thanks -- Ian You know what files they contain? Johnny -- Johnny Billquist || "I'm on a bus || on a psychedelic trip email: bqt@softjar.se || Reading murder books pdp is alive! || tryin' to stay hip" - B. Idol ------------------------------ Date: 13 Nov 2008 15:44:20 +0100 From: eplan@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: [VMS V8.3] PCSI Kit Verification Message-ID: <491c4b54$1@news.langstoeger.at> I just downloaded the SeaMonkey for VMS and began to install it: $ prod inst cswb %PCSI-I-VERIFYPDB, performing product database verification ... %PCSI-I-VERIFYSUC, successfully completed product database verification Performing product kit validation of signed kits ... %PCSI-I-VALPASSED, validation of DSA1:[INSTALL_KITS]CPQ-AXPVMS-CSWB-T0101-10-1.P CSI$COMPRESSED;1 succeeded %PCSI-I-VALPASSED, validation of DSA1:[INSTALL_KITS]CPQ-AXPVMS-GTK-V0102-10-1.PC SI$COMPRESSED;1 succeeded %PCSI-I-VALPASSED, validation of DSA1:[INSTALL_KITS]CPQ-AXPVMS-OPL-V0100-0A9-1.P CSI$COMPRESSED;1 succeeded The following product has been selected: CPQ AXPVMS CSWB T1.1-10 Layered Product Do you want to continue? [YES] I see, that the PRODUCT INSTALL of the CSWB checks for the validity of the required GTK and OPL kitfiles (and succeeds). BUT: These 2 kits are already installed (for the MOZILLA V1.7-13 also named CSWB - why got SeaMonkey not a new kit name like SWB or similar? - I installed some years ago), and these kitfiles are there per coincidence (left over from the last installation) and so nothing needs to be done with these kit files (and iff the files there are really corrupt *now*, this would be meaningless and must not be a possible reason to eventually abort the installation) as all the requirements for CSWB installation are already fulfilled (and nothing needs to be installed now in addition). So, I see it as PCSI checks in the wrong order (first prerequisites, then kits, then if prerequisites are already fulfilled)... Do you agree (and can someone from VMS engineering fix this)? Is this something in PCSI itself, or in the PCSI$DESCRIPTION of SeaMonkey? TIA -- Peter "EPLAN" LANGSTÖGER Network and OpenVMS system specialist E-mail Peter@LANGSTOeGER.at A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist ------------------------------ End of INFO-VAX 2008.615 ************************