INFO-VAX Wed, 31 Oct 2007 Volume 2007 : Issue 596 Contents: RE: 9-Track tapes on integrity Re: 9-Track tapes on integrity Re: EST Fubar Re: EST Fubar Re: Happy Anniversary VMS - 30 years young Re: Happy Anniversary VMS - 30 years young Re: Happy Anniversary VMS - 30 years young Re: Happy Anniversary VMS - 30 years young Re: Happy Anniversary VMS - 30 years young Re: Happy Anniversary VMS - 30 years young Re: HP *is* celebrating 30 years of OpenVMS Re: Newbie directory protection question Re: Pathworks vs CIFS performance Re: Pathworks vs CIFS performance Re: Pathworks vs CIFS performance Re: Revisited: Wireless (WPA) authentication and OpenVMS Re: Still more NFS problems - and happy birthday VMS ---------------------------------------------------------------------- Date: Wed, 31 Oct 2007 07:36:29 -0400 From: "Ebinger . Eric" Subject: RE: 9-Track tapes on integrity Message-ID: No, the goal is to get the whole system on Itanium. If that is not = possible than the rational for moving any of it to Itanium is extremely = questionable. =20 Eric ________________________________ From: Gorazd Kikelj [mailto:gorazd.kikelj@ne.spamaj.hp.com] Sent: Tue 10/30/2007 3:01 PM To: Info-VAX@Mvb.Saic.Com Subject: Re: 9-Track tapes on integrity "Ebinger . Eric" wrote in message news:B3C98F093EDF2146ADB62CDBD5B7FB060FEC7B@andexch01.drc.com... Is it possible (I know it's not supported) to use a SCSI 9-track tape = drive (TSZ07) on a VMS Integrity server (RX26660) by way of an Ultra320 scsi adapter? We would also need to use a 4mm DDS-3 drive but I can't = imagine that would be a problem if the 9-track would work. It is possible to put a little alfa behind 9-track? And then MSCP serve = tape drive to itanium. This should work. Best, Gorazd ------------------------------ Date: Wed, 31 Oct 2007 06:18:31 -0700 From: "Tom Linden" Subject: Re: 9-Track tapes on integrity Message-ID: On Wed, 31 Oct 2007 04:36:29 -0700, Ebinger . Eric wrote: > No, the goal is to get the whole system on Itanium. If that is not > possible than the rational for moving any of it to Itanium is extremely > questionable. Well you could think of a small alpha as a tape controller > Eric > > ________________________________ > > From: Gorazd Kikelj [mailto:gorazd.kikelj@ne.spamaj.hp.com] > Sent: Tue 10/30/2007 3:01 PM > To: Info-VAX@Mvb.Saic.Com > Subject: Re: 9-Track tapes on integrity > > > > > "Ebinger . Eric" wrote in message > news:B3C98F093EDF2146ADB62CDBD5B7FB060FEC7B@andexch01.drc.com... > > Is it possible (I know it's not supported) to use a SCSI 9-track tape > drive > (TSZ07) on a VMS Integrity server (RX26660) by way of an Ultra320 scsi > adapter? We would also need to use a 4mm DDS-3 drive but I can't imagine > that would be a problem if the 9-track would work. > > > It is possible to put a little alfa behind 9-track? And then MSCP serve > tape > drive to itanium. This should work. > > Best, Gorazd > > > > -- PL/I for OpenVMS www.kednos.com ------------------------------ Date: Wed, 31 Oct 2007 13:25:36 +0100 From: "Dr. Dweeb" Subject: Re: EST Fubar Message-ID: <47287441$0$7612$157c6196@dreader2.cybercity.dk> AEF wrote: > On Oct 29, 9:03 pm, "Dr. Dweeb" wrote: >> JF Mezei wrote: >>> Neil Rieck wrote: >>>> Timezone FYI, >> >>>> This morning (Sunday Oct 28) I got a call about a Solaris >>>> application that desided to transition from EDT to EST a week >>>> early. >> >>> Couple weeks ago, I installed a new thermostat to save energy. The >>> PDF I have foudn on the net provided listing of the "register" >>> values that could be set and had 0 for no automatic DST, and 1 for >>> automatic DST. So I set it to "1" since I had just bought this >>> thermostat and figured it would be according to current rules of >>> EST/DST. >> >>> It changed time this morning anyways. This time, I checked the >>> printed manual to get the register number so I could disable it, >>> and low and behold, the printed manual showed a possible value of 2 >>> for DST changes for 2007 and beyond ! >> >>> Interesting that the printed manual had been updated, but not the >>> on-line PDFs ! >> >> Today I looked at my SQLServer Job logs, there was a job scheduled >> for 03:00 this morning, which of course started on time, but >> finished before it started. This resulted in a VERY bizarre duration >> value in the log. >> >> Dr. Dweeb > > So tell us this bizarre value! Unless it was just a negative time. > > AEF -3976.21:20:27 ------------------------------ Date: 31 Oct 2007 12:52:32 GMT From: billg999@cs.uofs.edu (Bill Gunshannon) Subject: Re: EST Fubar Message-ID: <5orc4gFnst6sU1@mid.individual.net> In article <47287441$0$7612$157c6196@dreader2.cybercity.dk>, "Dr. Dweeb" writes: > AEF wrote: >> On Oct 29, 9:03 pm, "Dr. Dweeb" wrote: >>> JF Mezei wrote: >>>> Neil Rieck wrote: >>>>> Timezone FYI, >>> >>>>> This morning (Sunday Oct 28) I got a call about a Solaris >>>>> application that desided to transition from EDT to EST a week >>>>> early. >>> >>>> Couple weeks ago, I installed a new thermostat to save energy. The >>>> PDF I have foudn on the net provided listing of the "register" >>>> values that could be set and had 0 for no automatic DST, and 1 for >>>> automatic DST. So I set it to "1" since I had just bought this >>>> thermostat and figured it would be according to current rules of >>>> EST/DST. >>> >>>> It changed time this morning anyways. This time, I checked the >>>> printed manual to get the register number so I could disable it, >>>> and low and behold, the printed manual showed a possible value of 2 >>>> for DST changes for 2007 and beyond ! >>> >>>> Interesting that the printed manual had been updated, but not the >>>> on-line PDFs ! >>> >>> Today I looked at my SQLServer Job logs, there was a job scheduled >>> for 03:00 this morning, which of course started on time, but >>> finished before it started. This resulted in a VERY bizarre duration >>> value in the log. >>> >>> Dr. Dweeb >> >> So tell us this bizarre value! Unless it was just a negative time. >> >> AEF > > -3976.21:20:27 Stardate? Klingon clock? Quantum Thread Time? :-) bill -- Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves bill@cs.scranton.edu | and a sheep voting on what's for dinner. University of Scranton | Scranton, Pennsylvania | #include ------------------------------ Date: Wed, 31 Oct 2007 07:59:30 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: Happy Anniversary VMS - 30 years young Message-ID: In article <4727437E.7060900@comcast.net>, "Richard B. Gilbert" writes: > Do consider, however, the fact that RISC architectures are generally > faster than CISC. Isn't RISC inherently SLOWER than CISC, but in practice this is more than made up for by the fact that the time between design and production for RISC is so much shorter than for CISC that Moore's law gives RISC an advantage? (In other words, we are comparing a RISC processor based on one-year-old hardware with a CISC processor based on three-year-old hardware, or whatever.) ------------------------------ Date: Wed, 31 Oct 2007 08:00:34 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: Happy Anniversary VMS - 30 years young Message-ID: In article , "Tom Linden" writes: > On Tue, 30 Oct 2007 07:45:18 -0700, Richard B. Gilbert > wrote: > > > Do consider, however, the fact that RISC architectures are generally > > faster than CISC. > > > Not completely true. For simple things yes. Alpha had a 3:1 bloat factor > over VAX Actually, this just comes from being RISC. With memory so cheap these days, it's not really an issue. ------------------------------ Date: Wed, 31 Oct 2007 04:51:07 -0500 From: Ron Johnson Subject: Re: Happy Anniversary VMS - 30 years young Message-ID: On 10/30/07 18:26, John Smith wrote: [snip] > > Given that Hurd has just invested a boatload of money to create NeoView on > NSK, it's doubtful that HP will be moving off Itanic any time soon. But HP > may well become Itanic's only customer, which puts it in the same boat as > Alpha was in. Itanium has excellent floating point performance, so it's go a big backing in the HPC crowd. It's a visible, but small, market. And Unisys (I think, still) likes it for it's large SMP systems. But even they are moving to "Xeon64". -- 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: Wed, 31 Oct 2007 05:13:59 -0500 From: Ron Johnson Subject: Re: Happy Anniversary VMS - 30 years young Message-ID: On 10/31/07 02:59, Phillip Helbig---remove CLOTHES to reply wrote: > In article <4727437E.7060900@comcast.net>, "Richard B. Gilbert" > writes: > >> Do consider, however, the fact that RISC architectures are generally >> faster than CISC. > > Isn't RISC inherently SLOWER than CISC, but in practice this is more > than made up for by the fact that the time between design and production > for RISC is so much shorter than for CISC that Moore's law gives RISC an > advantage? (In other words, we are comparing a RISC processor based on > one-year-old hardware with a CISC processor based on three-year-old > hardware, or whatever.) No. The genesis of RISC was the observations that most compiler writers didn't use every exotic opcode, but just strung together sets of the simple, fast ones. They used that knowledge to build spare designs: - only memory instructions are load/store - no indirect addressing modes - longword addressing - *lots* of registers - no microcode, all opcodes hard-wired This simplicity just worked better than CISC because microcode is slow, registers are fast, and the simplicity let them crank up the MHz. Of course, RISC has been getting more complex ever since those first ROMP, MIPS and SPARC chips. But that's "allowable" because chips now have so many transistors that you can still maintain the essence of RISC. -- 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: Wed, 31 Oct 2007 06:08:24 -0700 From: "Tom Linden" Subject: Re: Happy Anniversary VMS - 30 years young Message-ID: On Wed, 31 Oct 2007 01:00:34 -0700, Phillip Helbig---remove CLOTHES to reply wrote: > In article , "Tom Linden" > writes: > >> On Tue, 30 Oct 2007 07:45:18 -0700, Richard B. Gilbert >> wrote: >> >> > Do consider, however, the fact that RISC architectures are generally >> > faster than CISC. >> > >> Not completely true. For simple things yes. Alpha had a 3:1 bloat >> factor >> over VAX > > Actually, this just comes from being RISC. With memory so cheap these > days, it's not really an issue. > It is not just memory, it has other design requirements such as bigger memory bandwidth and larger caches -- PL/I for OpenVMS www.kednos.com ------------------------------ Date: Wed, 31 Oct 2007 06:16:22 -0700 From: "Tom Linden" Subject: Re: Happy Anniversary VMS - 30 years young Message-ID: On Wed, 31 Oct 2007 03:13:59 -0700, Ron Johnson wrote: > On 10/31/07 02:59, Phillip Helbig---remove CLOTHES to reply wrote: >> In article <4727437E.7060900@comcast.net>, "Richard B. Gilbert" >> writes: >> >>> Do consider, however, the fact that RISC architectures are generally >>> faster than CISC. >> >> Isn't RISC inherently SLOWER than CISC, but in practice this is more >> than made up for by the fact that the time between design and production >> for RISC is so much shorter than for CISC that Moore's law gives RISC an >> advantage? (In other words, we are comparing a RISC processor based on >> one-year-old hardware with a CISC processor based on three-year-old >> hardware, or whatever.) > > No. > > The genesis of RISC was the observations that most compiler writers > didn't use every exotic opcode, but just strung together sets of the > simple, fast ones. > Well those observations were not particularly keen and largely based, AIR, on analysis of C, Pascal and F77 and ignore Cobol and PL/I. Note that the only Risc design to have done so was Power from IBM, which even allowed unaligned access for a minimal (one tick) cost > They used that knowledge to build spare designs: > - only memory instructions are load/store > - no indirect addressing modes > - longword addressing > - *lots* of registers > - no microcode, all opcodes hard-wired > > This simplicity just worked better than CISC because microcode is > slow, registers are fast, and the simplicity let them crank up the MHz. > > Of course, RISC has been getting more complex ever since those first > ROMP, MIPS and SPARC chips. But that's "allowable" because chips > now have so many transistors that you can still maintain the essence > of RISC. Actually if Dec had continued to a 64 bit VAX it would have looked that way as well, compare z-series with 370 > -- PL/I for OpenVMS www.kednos.com ------------------------------ Date: Wed, 31 Oct 2007 09:14:20 +0100 From: "Fred Zwarts" Subject: Re: HP *is* celebrating 30 years of OpenVMS Message-ID: "David J Dachtera" wrote in message = news:4727E63C.9B8C6D07@spam.comcast.net... > AEF wrote: >>=20 >> On Oct 30, 9:22 am, "Syltrem" wrote: >> > I want to thank HP for putting the "Celebrating 30 years of = OpenVMS" on >> > their main page. >> > >> > That's great ! >> > >> > And of course I thank everyone that worked on making this 30-years = site. >> > Now that it's more properly advertised, I hope the number of visits = will >> > raise. >> > >> > I know, some here will not be so happy and find ways to complain, = but I >> > think this is the best thing that happened to VMS for a very long = time, >> > advertisement-wise. Not only preaching to the choir this time. >> > >> > Happy anniversary VMS ! >> > >> > -- >> > Syltremhttp://pages.infinit.net/syltrem(OpenVMS information and = help, en fran=E7ais) >>=20 >> Happy Anniversary VMS!!! >>=20 >> Yep, I just checked: It's still there on the front f------ page! >>=20 >> Hooray again! >=20 > Which "front" page? Didn't see it on http://www.hp.com/ , just some = stuff about > Halloween specials. It is now at: http://h71000.www7.hp.com/openvms/30th/index.html ------------------------------ Date: Wed, 31 Oct 2007 08:19:40 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: Newbie directory protection question Message-ID: In article , JF Mezei writes: > VMS is Virus Free wrote: > > Normal protection on the MFD would be (RWE,RWE,RE,E). > > Interesting. All my MFDs are at RWED RWED RE E, with owner 1,1. Same here, except a) one I have explicitly changed (by mistake; I changed the protection on all top-level directories on a public disk and the MFD got caught in the wildcard) and b) the documentation CD, which has (RE,RE,RE,RE) and I am SURE that I did not change that! :-) > While I know that NFS has screwed with the MFD on one disk, I haven't > mapped any other drives and I have to assume that VMS created those MFDs > at time of disk initialisation (or perhaps BACKUP ?). I have never used NFS. ------------------------------ Date: Wed, 31 Oct 2007 15:44:44 GMT From: VMS is Virus Free Subject: Re: Pathworks vs CIFS performance Message-ID: On Mon, 29 Oct 2007 21:42:55 -0000, keithparris_NOSPAM@yahoo.com wrote: >Could you divide the directories up, spread the files around among >smaller directories, and use a search list logical name to locate >them, putting the most-used files in the first directory and seldom- >used files in the last directory in the search list? Maybe I'm missing the obvious here; please help me out if it's just too simple to see. Assume disk:[a] is overloaded. Create some sub-dirs to hold the files: $ create/dir disk:[a.a1] $ create/dir disk:[a.a2] $ create/dir disk:[a.a3] Make the searchlist: $ define/system a_searchlist disk:[a.a1],disk:[a.a2],disk:[a.a3] Move some files around: $ rename disk:[a]a*.*,b*.*,c*.* disk:[a.a1] $ rename disk:[a]d*.*,e*.*,f*.* disk:[a.a2] $ rename disk:[a]g*.*,h*.*,i*.* disk:[a.a3] Create the share in Pathworks: $ admin add share/dir test a_searchlist /desc="Searchlist version" %PWRK-E-ERRADDSHARE, error adding share TEST_A -LM-E-ERROR_INVALID_N, invalid name If i replace the a_searchlist logical with something like disk:[a] then the command works as expected. I am probably just missing something but I couldn't figure out a way to make Pathworks happy with a searchlist. Any hints? Well, to heck with the hints, any outright answers? ------------------------------ Date: Wed, 31 Oct 2007 16:06:57 GMT From: =?ISO-8859-1?Q?Jan-Erik_S=F6derholm?= Subject: Re: Pathworks vs CIFS performance Message-ID: OK, neither a hint nor an answer, more like a shoot in the dark... :-) VMS is Virus Free wrote: Try somthing like : > $ create/dir disk:[a.a1] > $ create/dir disk:[a.a2] > $ create/dir disk:[a.a3] > > $ def/sys/tran=(con,term) f_root - > $ disk:[a.a1.], - > $ disk:[a.a2.], - > $ disk:[a.a3.] > > Move some files around... > > $ admin add share/dir test f_root:[000000] - > /desc="Searchlist version" > It works from plain DCL anyway. Not tested with PW... Jan-Erik. > On Mon, 29 Oct 2007 21:42:55 -0000, keithparris_NOSPAM@yahoo.com > wrote: > >> Could you divide the directories up, spread the files around among >> smaller directories, and use a search list logical name to locate >> them, putting the most-used files in the first directory and seldom- >> used files in the last directory in the search list? > > Maybe I'm missing the obvious here; please help me out if it's just > too simple to see. > > Assume disk:[a] is overloaded. Create some sub-dirs to hold the files: > > $ create/dir disk:[a.a1] > $ create/dir disk:[a.a2] > $ create/dir disk:[a.a3] > > Make the searchlist: > > $ define/system a_searchlist disk:[a.a1],disk:[a.a2],disk:[a.a3] > > Move some files around: > > $ rename disk:[a]a*.*,b*.*,c*.* disk:[a.a1] > $ rename disk:[a]d*.*,e*.*,f*.* disk:[a.a2] > $ rename disk:[a]g*.*,h*.*,i*.* disk:[a.a3] > > Create the share in Pathworks: > > $ admin add share/dir test a_searchlist /desc="Searchlist version" > %PWRK-E-ERRADDSHARE, error adding share TEST_A > -LM-E-ERROR_INVALID_N, invalid name > > If i replace the a_searchlist logical with something like disk:[a] > then the command works as expected. > > I am probably just missing something but I couldn't figure out a way > to make Pathworks happy with a searchlist. > > Any hints? Well, to heck with the hints, any outright answers? ------------------------------ Date: Wed, 31 Oct 2007 17:48:20 GMT From: VMS is Virus Free Subject: Re: Pathworks vs CIFS performance Message-ID: On Wed, 31 Oct 2007 16:06:57 GMT, Jan-Erik Söderholm wrote: >OK, neither a hint nor an answer, more like >a shoot in the dark... :-) > >VMS is Virus Free wrote: > >Try somthing like : > > > $ create/dir disk:[a.a1] > > $ create/dir disk:[a.a2] > > $ create/dir disk:[a.a3] > > > > $ def/sys/tran=(con,term) f_root - > > $ disk:[a.a1.], - > > $ disk:[a.a2.], - > > $ disk:[a.a3.] > > > > Move some files around... > > > > $ admin add share/dir test f_root:[000000] - > > /desc="Searchlist version" > > > >It works from plain DCL anyway. Not tested with PW... > >Jan-Erik. Good guess; seems entirely reasonable, Pathworks even lets you mount the "disk", and you can map the drive from Windows Explorer, you can even create new files on the share. However, when you get to the drive on Windows, all you see are the files in the first searchlist translation. The technique works okay on VMS: you see all files as if they were in one directory. My guess is that, when Pathworks looks for the files, the developer never considered the possibility of using a searchlist such as this and thus never walked the translations. My guess is that the code uses the old $TRAN (or whatever it was called) rather than the $TRNLNM system service or it uses the newer one but doesn't do the looping needed to catch all files. ------------------------------ Date: Wed, 31 Oct 2007 15:50:20 -0000 From: IanMiller Subject: Re: Revisited: Wireless (WPA) authentication and OpenVMS Message-ID: <1193845820.877204.203040@22g2000hsm.googlegroups.com> Do you mean this RADIUS server ? http://starlet.deltatel.ru/radiusvms/ ------------------------------ Date: Wed, 31 Oct 2007 06:34:27 -0700 From: etmsreec@yahoo.co.uk Subject: Re: Still more NFS problems - and happy birthday VMS Message-ID: <1193837667.656005.256820@50g2000hsm.googlegroups.com> On 28 Oct, 07:59, Gremlin wrote: > Hi JF > > Well, in the proxy database on vms I mapped system and my username, > system to UID 0 and GID 0, my username to the same value specified in > the password file in Windows. > > Then, using the MS tool on the server, I mapped the "unix" username > SYSTEM to the Windows username Domain Administrator and gave it UID 0 > and GID 0. I then mapped my "unix" username to my Windows username and > gave it the same UID and GID as in the passwd file on Windows. > Anonymous access UID -2, GID -2 is allowed on the Windows NFS server as > a "catch all" in case mapping fails.... > > On Windows, the nfssvc service is running happily and logging > mount/unmounts, so I can only assume that it is really there - VMS also > is happy to see it. There don't appear to be any other bits on the > Windows side, expected or running. > > The client is running on the VMS side (hence the mount/unmount). > > On VMS, I have tried mounting using every combination of UID/GUI/path > etc - and apart from obvious errors (all logged), the mount/unmount > process works fine. The reads fail no matter what combination I have > tried on anything..... > > So, just jumped onto a SuSE and Redhat client, used the linux NFS client > to map and browse - works like a charm. Therefore, Windows2003 is > serving it fine, something on VMS is broken.....stopped and restarted > the client on VMS - still won't do it. > > > > JF Mezei wrote: > > Gremlin wrote: > >> On Windows, the UNIX (ie VMS) username SYSTEM is mapped to a domain > >> administrator account, anonymous access is allowed and a NFS share is > >> created. The Windows log shows no errors > > > "SYSTEM" doesn't really exist in NFS parlance. UID and GID values are > > exchanged. > > > On VMS, when you issue a TCPIP MOUNT command, the GID/UID values are > > either specified in the MOUNT command, or VMS looks at the proxy > > database for a record with your username in it and uses UID/GID > > associated with this, or of there is nothing available, it uses a > > default value of -2 -2 . > > > On unix servers, you have options of mapping the incoming remote UID/GID > > values to the local user database. You can't really do that on windows > > or VMS since the UID/GID values have no meaning on their systems. > > > On all systems, you also have options to map incoming requests to > > specific local usernames. > > >> From VMS, I can mount the NFS disk and unmount it, but I cannot do > >> any file access at all. > > > It appears that just mounting doesn't do that much. (mountd deamon), but > > when you then try to access files, it is a different process (nfsd) > > which handles the requests. > > >> wasn't really happening - just timing out as the VMS message says. > > > Perhaps the mountd process is running, but the nfsd deamon isn't running ? > > > The fact that VMS is waiting for a response and timing out is an > > indication that, instead of getting a bad status code from the server, > > VMS isn't getting any responses. Remember that NFS is usually UDP > > traffic, so there is no concept of a dropped connection. You can reboot > > the server and the client can happily continue to access his files.- Hide quoted text - > > - Show quoted text - Is this a home environment or a production one? If it's a home environment, why not try the field test of Samba (a.k.a. CIFS) and see if that's any easier to work with? Steve ------------------------------ End of INFO-VAX 2007.596 ************************