1 INFO-VAX	Fri, 21 Apr 2006	Volume 2006 : Issue 221       Contents: Even lower DS10 pricing * Re: HP Storageworks Arrays are Bulletproof* Re: HP Storageworks Arrays are Bulletproof* Re: manually resetting file revision dates' Re: OpenVMS is the worlds best desktop! ' Re: OpenVMS is the worlds best desktop! . Re: Opinion: I was just trying to sell OpenVMS. Re: Opinion: I was just trying to sell OpenVMS. Re: Opinion: I was just trying to sell OpenVMS. Re: Opinion: I was just trying to sell OpenVMS Re: OT: Sparc not dead yet Re: OT: Sparc not dead yet Re: OT: Sparc not dead yet Re: OT: Sparc not dead yet Re: OT: Sparc not dead yet6 Overland Data neo 2000/MSL series drive + OpenVMS 8.2>< Re: Partitioned cluster question (reboot during lost quorum) Replacing drives in bricks Re: Sending mail via TLS Strange disk device state  Re: Strange disk device state  Re: Strange disk device state  Re: Strange disk device state  Re: The Undocumented OpenVMS Re: The Undocumented OpenVMS Re: The Undocumented OpenVMS Re: The Undocumented OpenVMS Re: The Undocumented OpenVMS Re: The Undocumented OpenVMS Re: VAX hardware error code 3 ?  Re: VAX hardware error code 3 ?  Re: VAX Q-bus and SCSI Disks?   F ----------------------------------------------------------------------  % Date: Fri, 21 Apr 2006 13:32:06 -0400 C From: "David Turner, Island Computers US Corp" <dbturner@icusc.com>   Subject: Even lower DS10 pricing: Message-ID: <jF82g.25883$Kh5.17201@bignews8.bellsouth.net>    Go to www.islandco.com/ds10.html   We have a large qty   @ DS10 466 with 512MB memory and 30GB IDE disk for only $1895 !!!!         --     David B Turner Island Computers US Corp 2700 Gregory St, Suite 180 Savannah GA 31404  Tel: 912 447 6622 X201 Cell: 912 447 6622 X251  Fax: 912 201 0402  Email: dbturner@icusc.com  Web: http://www.islandco.com% ===================================== < All orders are subject to the following terms and conditions. of sale. These should be read before ordering.% http://www.islandco.com/warranty.html    ------------------------------  # Date: Fri, 21 Apr 2006 14:01:27 GMT " From:   VAXman-  @SendSpamHere.ORG3 Subject: Re: HP Storageworks Arrays are Bulletproof 0 Message-ID: <00A548B3.04E58B1D@SendSpamHere.ORG>  i In article <am32g.137108$zI1.91227@fe2.news.blueyonder.co.uk>, Alan Greig <greigaln@netscape.net> writes:  >  > * >http://www.theinquirer.net/?article=31165 >Which begins: > I >HP HAS A TENDENCY to use gun metaphors when talking about their storage  0 >solutions, and the newest one is 'bulletproof'. > F >Instead of the wussy technical term for it, they went out and shot a F >hole clean through a half-million dollar Storageworks XP12000 server  >while it was running. > J >The bullet went right through it, but only half way through the fishtank F >on the other side. That is what you get with .308 rounds, not enough 7 >stopping power vs high-end arrays backed by fishtanks.  > F >In either case, they literally proved that their Storageworks arrays A >will take a bullet while streaming video, never missing a frame.  >......  >  >  >--  >Alan Greig     & FFFFF U   U  CCC   K   K    H   H PPPP' F     U   U C   C  K  K     H   H P   P ' F     U   U C      K K      H   H P   P & FFFF  U   U C      KK       HHHHH PPPP# F     U   U C      K K      H   H P # F     U   U C   C  K  K     H   H P # F      UUU   CCC   K   K    H   H P   K Another video demonstrating technology in a technology limiting *W*e-don't- I even-want-to-know-you-if *M*icro$uck-is-not-in-your *V*isions WMV format.    --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------    Date: 21 Apr 2006 10:14:05 -0700 From: sean@obanion.us 3 Subject: Re: HP Storageworks Arrays are Bulletproof C Message-ID: <1145639645.322076.237470@u72g2000cwu.googlegroups.com>   * Just don't let them do a firmware upgrade.  E The first one they did, and assured us we could leave all of our HPUX G systems running, had a bug that disabled all the FC ports on the array. +  Not surprisingly, all the systems crashed.      Sean   Alan Greig wrote: + > http://www.theinquirer.net/?article=31165  > Which begins:  > I > HP HAS A TENDENCY to use gun metaphors when talking about their storage 1 > solutions, and the newest one is 'bulletproof'.  > F > Instead of the wussy technical term for it, they went out and shot aF > hole clean through a half-million dollar Storageworks XP12000 server > while it was running.  > J > The bullet went right through it, but only half way through the fishtankF > on the other side. That is what you get with .308 rounds, not enough8 > stopping power vs high-end arrays backed by fishtanks. > F > In either case, they literally proved that their Storageworks arraysB > will take a bullet while streaming video, never missing a frame. > .....  >  >  > --   > Alan Greig   ------------------------------  % Date: Fri, 21 Apr 2006 08:22:23 -0700  From: foo@bar.com 3 Subject: Re: manually resetting file revision dates 8 Message-ID: <oith42lmoquogf4re0fdh580l62tq7dc8p@4ax.com>  7 On 12 Apr 2006 17:14:58 -0700, mabbuttg@yahoo.ca wrote:   B >I'm testing a script of mine which is checking some file revisionH >dates.  For proper testing, I need to create files with a few different >dates.  > G >I did some searching through past newsgroup posts, and found some info C >on changing the rev. date with some of the freeware apps (DFU) and F >someone posted a C program.  However, we're running a stock 7.3-1/AXPD >install - no freeware, just COBOL installed as an extra (other than >TCP/IP, of course). > D >Is there any way to do this on my setup without changing the system >time repeatedly? G >I'm hesistant to start installing additional things if I don't have to  >(agressive SOX auditors, etc).     F Speaking as a SOX IT auditor (for a Big 4) and as Hoff has pointed outC (	assuming you are subject to SOX section 404), if I found that you E were changing dates and doing non-production testing on a system that F supported your financial reporting systems, that would probably resultF in a finding of at least "significant deficiency" in Internal ControlsA Over Financial Reporting (ICOFR), which requires reporting to the D Audit Committee, if not a material weakness which requires reporting in the annual report.   @ If you are not a publicly traded company but still have external@ auditors, the Big 4 are now using a modified SOX audit which now@ delves much more deeply into the IT infrastructure, specifically) change control, sdlc and access controls.   B If you are referring to an Internal Audit, discovery of this wouldD probably result in a risk rating of "high" or "very high", resultingB in this specific item being reported to the Audit Committee. In my0 clients this kind of stuff causes heads to roll.  	 Soo...... ; 1) Do this stuff on a test system, not a production system. D 2) Ensure that any work that MUST be done on a production is solidlyC backed up by Change Control records and audit trail records showing , before state, actions taken and after state.* ------------------------------------------C NO, use John Mee3 at comcast dot net (remove the spaces and do the  # obvious with the "dot" and the "at"    ------------------------------  # Date: Fri, 21 Apr 2006 16:46:45 GMT 1 From: Keith Parris <keithparris_NOSPAM@yahoo.com> 0 Subject: Re: OpenVMS is the worlds best desktop!2 Message-ID: <V%72g.6661$v74.2498@news.cpqcorp.net>   Bill Gunshannon wrote:C > How does that apply tp HP?  They killed Alpha.  They no longer do G > hardware.  They sell other people's hardware.  At razor thin margins.   D HP has chosen to move to using microprocessors from the leading two I companies in the industry (Intel and AMD) rather than continuing to make  H its own, incompatible processor chips. That's a long way from no longer  making its own hardware.  ? Only selling other people's hardware? What about EVA, MSA, and  E SmartArray controllers? What about Superdome and all the specialized  F hardware involved in that, including the "sx" series of chipsets that E give HP hardware a competitive performance edge? What about ProLiant  ) servers? What about all of HP's printers?   B > VMS is about the only thing they have left with any real value.   E Having a majority marketshare position in printers certainly doesn't  ; hurt. :-) Actually, HP has a lot of things with real value.    ------------------------------    Date: 21 Apr 2006 09:56:27 -0700 From: bob@instantwhip.com 0 Subject: Re: OpenVMS is the worlds best desktop!C Message-ID: <1145638587.735608.290160@e56g2000cwe.googlegroups.com>   1 don't you give away things that have NO value? :)    ------------------------------  # Date: Fri, 21 Apr 2006 15:01:45 GMT 1 From: Keith Parris <keithparris_NOSPAM@yahoo.com> 7 Subject: Re: Opinion: I was just trying to sell OpenVMS 2 Message-ID: <tt62g.6650$%24.5022@news.cpqcorp.net>   JF Mezei wrote: J > If VMS is going to be competing in the same niche markets as Tandem/NSK,G > is there a point for the owner of both VMS and NSK to maintain both ?   H We see both VMS and NSK in mission-critical areas like stock exchanges, H but they rarely compete directly. Instead, the customer tends to select @ a software package, and the package choice in turn dictates the  underlying platform.  G NSK can do some things (like process pairs) that VMS can't, and VMS no  E longer supports fault-tolerant hardware. Conversely, VMS can do some  E things that NSK can't (as one example, VMS can run Oracle, while NSK  H cannot). So neither one can readily replace the other. And so HP is not E likely to eliminate one in favor of the other, or try to force users   from one platform to the other.    ------------------------------  # Date: Fri, 21 Apr 2006 15:07:45 GMT 1 From: Keith Parris <keithparris_NOSPAM@yahoo.com> 7 Subject: Re: Opinion: I was just trying to sell OpenVMS 2 Message-ID: <5z62g.6651$o44.4584@news.cpqcorp.net>   JF Mezei wrote: J > NASDAQ was probably paid a very large sum of money by Microsoft to frontI > its web site on Wintel crap, money to compensate for the higher risk of   > its web site being compromised   Not likely.    ------------------------------  # Date: Fri, 21 Apr 2006 15:15:57 GMT 1 From: Keith Parris <keithparris_NOSPAM@yahoo.com> 7 Subject: Re: Opinion: I was just trying to sell OpenVMS 2 Message-ID: <NG62g.6652$Z44.4666@news.cpqcorp.net>   Bill Gunshannon wrote: > It's not like for E > every hospital that migrates away another one migrates towards VMS.   I Not true. There is still a steady stream of hospitals installing VMS for  B the first time, driven by applications. If they choose a software I package from Cerner, IDX, or SMS, it's very likely they'll be running it  B on VMS, even if that means setting up a brand-new VMS environment.  A Same is true for stock (and other) exchanges. If they choose OMX  8 software, it's almost certain they'll be running on VMS.   ------------------------------    Date: 21 Apr 2006 12:15:23 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 7 Subject: Re: Opinion: I was just trying to sell OpenVMS 3 Message-ID: <$$hnWF+Cd49X@eisner.encompasserve.org>   f In article <NG62g.6652$Z44.4666@news.cpqcorp.net>, Keith Parris <keithparris_NOSPAM@yahoo.com> writes:  C > Same is true for stock (and other) exchanges. If they choose OMX  : > software, it's almost certain they'll be running on VMS.  E   Which stock exchange is HP traded on, and how much does that stock     exchange depend on VMS?   E   How many VMS systems does HP use to run it's own business (probably 9   few since VMS is one of HP's more recent acquisitions)?    ------------------------------  % Date: Fri, 21 Apr 2006 08:37:20 +0200 + From: Karsten Nyblad <nospam@nospam.nospam> # Subject: Re: OT: Sparc not dead yet = Message-ID: <44487d95$0$60787$157c6196@dreader1.cybercity.dk>   
 Andrew wrote: I > There is no evidence that Sun ever seriously considered replacing SPARC I > with x86 only speculation. What of course has changed is Sun's adoption F > of x86 a strategy driven by the realisation that they can make moneyB > out of x86 platforms hosting Linux, Windows and Solaris plus theF > adoption of a standard server packaging to support low end SPARC and > x86 servers.  G I have to second Andrew on that.  It seems like people have missed the  F fact that the x86 sales make up only a fraction of Sun's sales.  Last F quarter their revenue on x86 was about $100mn and their total revenue H $3.3bn.  It seems like Suns customers prefer Sparc to x86 and that that  will not change any time soon.   ------------------------------  % Date: Fri, 21 Apr 2006 04:13:25 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> # Subject: Re: OT: Sparc not dead yet , Message-ID: <44489424.DCF75F13@teksavvy.com>   Karsten Nyblad wrote: H > I have to second Andrew on that.  It seems like people have missed theG > fact that the x86 sales make up only a fraction of Sun's sales.  Last G > quarter their revenue on x86 was about $100mn and their total revenue I > $3.3bn.  It seems like Suns customers prefer Sparc to x86 and that that   > will not change any time soon.    D Strategies/visions often take a long time to materialise. Sun signedH stragegic agreements with AMD for the use of Opteron. Sun didn't do that to become a wintel box maker.   H There is no question that at present, the installed base wants continued development of Sparc.   A But with Solaris on 64 bit 8086 already, the application software G portfolio is  also supporting this platform more and more.  This allows D a "natural" migration decided entirely by customers based on what is@ available in terms of software and the true costs and not on any' artificial deadlines set by the vendor.   H Compare this with VMS with an artificial end of sales of Alpha at a timeH when even a crucial piece of software like Oracle isn't yet available on that IA64 thing for VMS.    G And in a way, this is great because it allows Sun to focus Sparc on the G truly high end, and still provide Solaris for the low and medium end on F the 8086. Compare this to VMS and HP-UX which are being relegated ontoH to High End with statements from the vendor (HP) that the low and mediumF end business usiers should focus on 8086 based servers/workstations on* which neither VMS nor HP-UX (nor NSK) run.        E In 2001, had the owners of VMS announced that they would continue the D Alpha Roadmap at least until EV8 and would also port VMS to the 8086E (industry standard) to regain some market share in the low and medium D end markets, this would have given lots of times and product/versionF cycles for developpers to release on both VMS and 8086 and by the timeF Alpha would have been formally killed, most of VMS and its application9 portfolio would already be available on the 8086 version.    ------------------------------    Date: 21 Apr 2006 05:15:44 -0700- From: "Andrew" <andrew_harrison@symantec.com> # Subject: Re: OT: Sparc not dead yet C Message-ID: <1145621744.022473.315760@j33g2000cwa.googlegroups.com>    JF Mezei wrote:  > Andrew wrote: K > > There is no evidence that Sun ever seriously considered replacing SPARC  > > with x86 only speculation. >   > Correct. There is no EVIDENCE. > D > But if the 8086 starts to scale to large systems, the 8086/SolarisH > market will start to eat into the Sparc/Solaris market and eventually,> > it may not be worth the effort to continue to develop Sparc. >   C Large x86 systems have been available for some time, Unisys and IBM C both produce >16 CPU x86 based servers. Before that Sequent who IBM % bought did a 64 way x86 based system.   G Sun themselves are due to release a 16 core (8 module) AMD based server G in the middle of this year to flesh out the top end of their x86 range.   F It has always been possible to build large x86 systems give or take 64C bit support etc. The problem has been that x86 lacks the HW/SW/Apps C ecosystem to support very large servers except as organ donors to a  VMWARE etc type environment.  A Ironically Sun with Solaris x86 are probably the only vendor with E enough of the pieces to build really big x86 syystems if they wanted.   E But the reality is that they don't want to because they don't see the = point. Larger than 4-8 way systems require entirely different D architectures to the Hyperfabric/Frontside type busses/switches thatF work well for commodity platforms. Once you have architected your veryG high bandwidth low latency proprietary bus, hot plug infrastructure etc @ etc the cost of the commodity or otherwise CPU's that sit in the& platform becomes an incidental factor.  G >From a ecosystem perspective you have a different group of ISV's or at E least different products that drive the revenues for low/mid and high E end systems so there isn't much to be gained by being commodity there G either and most fo the small to mid range OS's (Linux, Windows, FreeBSD B etc) simply will not scale to support very large high end systems.  G If you view it in this light then Sun's decision to support 16 CPU's in G their AMD based platforms is eminently sensible because its the largest 2 kind of system that the x86 ecosystem can support.  G Ironically if you did this analysis 5 years ago you would probably have D reached a similar conclusion. The gap between what is required for aB high end server ecosystem to perform compared with a low/mid range- ecosystem hasn't changed much over this time.    Regards  Andrew Harrison    ------------------------------    Date: 21 Apr 2006 09:51:40 -0700 From: etmsreec@yahoo.co.uk# Subject: Re: OT: Sparc not dead yet C Message-ID: <1145638300.042542.126610@u72g2000cwu.googlegroups.com>   @ Why should they prefer anything else?  And, for that matter, why? shouldn't Sun continue believing in Sparc (these are rhetorical  questions).   F If Sun wanted to be a pure hardware vendor then I could see them beingA interested in selling x86 systems and their customer base who are F probably as loyal as VMS people would buy them.  Why would Sun want to7 compete in BillyBox land though?  For a company with an F enterprise-systems background, it doesn't make sense does it?  There'sB margin in high end reliable systems.  There's bugger all margin inA low-end x86 servers since HP and Dell have them sliced and diced.    ------------------------------    Date: 21 Apr 2006 09:40:07 -0700- From: "Doug Phillips" <dphill46@netscape.net> # Subject: Re: OT: Sparc not dead yet C Message-ID: <1145637607.610448.199630@v46g2000cwv.googlegroups.com>    Dave Froble wrote: > JF Mezei wrote:  <...> J > > Neother do I.  But if the market and the 8086 develops to a point thatJ > > the 8086 can take on the vast majority of configurations now done onlyJ > > in Sparc, and the 8086 is cheaper, then Sun will also be responsive to. > > cutsomer demands for lower priced systems. > > C > > Sparc competes against power and 8086. Both Sparc and Power are @ > > essentially proprietary chips without competition within theG > > architectures. (yes, I know sparc has multiple sources, but do they  > > really *compete* ? ) > J > What you're ignoring is that AMD at least doesn't give a rat's ass aboutJ > the top end.  The desktop (say volumn) is just getting to dual core, andD > the majority of servers is in the 4-8 core size.  That's where the. > volume is.  Volume is where the money is at. >   B Are you kidding? AMD and Intel *covet* the top end. AMD is pushing, Intel & x86 upward like noone else ever has.  E To anyone paying attention it's obvious that x86 (AMD64 / EM64T) will D keep advancing upward. Anyone willing to "do the math" will see thatE the R&D $ pushing the x86 advancement is greater than other competing G chips. Okay, forget the money. Consider inertia. There is more software F running on x86 than any other instruction set. There are tools alreadyC developed for x86 that must be ported or re-written for development C under another set. Plus, the Alpha folks are working in both camps. ) Some competitive feelings there, perhaps?   C Plot it out. SUN will reach a point where they can't afford to keep E dumping money into SPARC when x86 does the job faster and cheaper and F doesn't put such a strain on their budget. Intel will reach that pointG with Itanium because the x86 is one business Intel doesn't want to lose G and AMD will keep on eating into market share if Intel doesn't keep up. E And IBM? Well, I don't see them pouring tons of money into Power just @ for pride. At some point it's wiser to stop swimming against the" current and climb aboard the raft.   ------------------------------  % Date: Fri, 21 Apr 2006 13:07:07 -0400 C From: "David Turner, Island Computers US Corp" <dbturner@icusc.com> ? Subject: Overland Data neo 2000/MSL series drive + OpenVMS 8.2> : Message-ID: <Vh82g.25866$Kh5.23655@bignews8.bellsouth.net>  K We have a customer to whom we shipped an Overland Data Neo 2000 series tape . library (same as MSL from HP) with two DLT8000< DLT Tape drives installed. Cabling and termination included.  J before shipping we tested it on an Alphaserver DS10 617Mhz running OpenVMS 7.3-2 K We backed up to each drive of the library and then did a backup back to the # hard drive to ensure data integrity      Everything worked beautifully   > Now it is at a customer site and they cannot even get past the $mount/foreign commandK It hangs their system (which is an Itanium II RX2620 with U3 SCSI interface  runninng 8.2+ patches    Anyone got any ideas?   I One HP tech said that support in VMS for "older" tape drives is no longer . there and customer should have something newer   David    --     David B Turner Island Computers US Corp 2700 Gregory St, Suite 180 Savannah GA 31404  Tel: 912 447 6622 X201 Cell: 912 447 6622 X251  Fax: 912 201 0402  Email: dbturner@icusc.com  Web: http://www.islandco.com% ===================================== < All orders are subject to the following terms and conditions. of sale. These should be read before ordering.% http://www.islandco.com/warranty.html    ------------------------------  + Date: Fri, 21 Apr 2006 07:37:14 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)E Subject: Re: Partitioned cluster question (reboot during lost quorum) $ Message-ID: <e2a23a$l9v$1@online.de>  ? In article <4TM1g.6582$Bt3.4752@news.cpqcorp.net>, Hoff Hoffman " <hoff-remove-this@hp.com> writes:    > JF Mezei wrote:  > > Hoff Hoffman wrote: I > >>    If you want to boot outside the cluster, then I tend to prefer to @ > >> avoid enabling NISCS_LOAD_PEA0 and I don't load VAXCLUSTER. > > I > > No, the idea is to be able to boot one node first before you bring in L > > the other ones, so that first one has to form the cluster and the others > > join in later. > H >    Welll, my own idea is to be able to boot distinctly and completely C > and fully outside the cluster, and to thus reduce the chances of  $ > creatively configured corruptions.  K >    My goal is to win with the least effort and the least risk, and -- as  J > a rule, I prefer to configure and operate the configuration such that I K > don't have to get "creative".  I'd prefer to win by avoiding the battle,  J > because battles are expensive and wasteful.  (Yes, I've been re-reading K > a translation of a very old book, but I digress.)  If I can configure to  J > avoid the problem (through hardware or software upgrades, or documented 4 > management procedures and sequences), then I will.  A Something else JF might want to try, which I think Hoff has also  F mentioned here before, is to set up various system roots with various F parameters, perhaps with and without VAXcluster etc, then boot from a  non-standard root if required.   ------------------------------  # Date: Fri, 21 Apr 2006 17:02:58 GMT % From: Rob Brown <mylastname@gmcl.com> # Subject: Replacing drives in bricks D Message-ID: <Pine.LNX.4.61.0604211057130.6747@localhost.localdomain>  # On Thu, 20 Apr 2006, syslost wrote:   F > But wait there's more.. I also have 48 ba350-sb shelves.  Some have E > 150 watt power supplys, and all have two fans.  Same deal, you pay   > only the shipping.  D This sounds attractive.  Now here is a hypothetical.  I get a BA350 G and some rz28s in storageworks bricks.  n years down the road an rz28s  D dies.  What is the wisdom of opening up the bricks in replacing the = dead drive with a modern drive (with appropriate cabling and  F terminators etc.) so that I can continue to use the slot in the BA350?  8 Would you do it as a hobbyist?  Would you do it at work?   Thanks.    - Rob      --    B Rob Brown                        b r o w n a t g m c l d o t c o mA G. Michaels Consulting Ltd.      (866)438-2101 (voice) toll free! 6 Edmonton                         (780)438-9343 (voice)6                                    (780)437-3367 (FAX)3                                    http://gmcl.com/    ------------------------------   Date: 21 Apr 2006 12:26:54 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)! Subject: Re: Sending mail via TLS + Message-ID: <4as1ceFu2079U3@individual.net>   6 In article <4448444A.504C2A2B@neoasrptahmlnionwk.net>,C 	"David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes:  > "Jeffrey H. Coffield" wrote: >>  K >> One of my customers needs to send e-mail from Alpha VMS 7.3-2 using TLS. F >> I see that PMDF from Process Software does this but are there other
 >> solutions?  > : > Even though I can't help, may I ask what "TLS" might be?    Transport Layer Security   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Fri, 21 Apr 2006 16:56:06 +0200 , From: Albrecht Schlosser <ajs567@tiscali.de>" Subject: Strange disk device state, Message-ID: <dqra2e.5b6.ln@news.hus-soft.de>   Problem summary:  D There is a disk that can't be accessed because the system gives the  error message:  < %SYSTEM-W-DEVALLOC, device already allocated to another user  G although it has been dismounted successfully. Device state is "online"  1 (not allocated). However, show device/full shows:     Reference count       4294967295   which is hex FFFFFFFF, see:    $ x = 4294967295 $ sho sym x /    X = -1   Hex = FFFFFFFF  Octal = 37777777777   F History: The device has had lots of errors ( > 35000 ), and the error G count has been reset. There was a process running (BADBLOCKSCAN), that  I has been terminated with STOP/ID. The local system manager said, that he  " did an ANA/DISK, but no ANA/MEDIA.  G There is a replacement disk that should be inserted and mounted (in an  F Itanium RX2620 box, hot swappable), but obviously, this can't be done.  H Does anybody have an idea, what to do without a reboot? What else can I C look for? (BTW: SDA> sho dev /chan doesn't work, though I think it  D should, as can be seen in the included help output. It doesn't work = under OpenVMS Alpha 7.3-2 either. So, what am I doing wrong?)    Thanks for all help    Albrecht    F The following output is what I found out so far (sorry, if some lines  wrap and for the long text)   
 $ sho time    21-APR-2006 16:32:34    $ tcpip sho vers  K    HP TCP/IP Services for OpenVMS Industry Standard 64 Version V5.5 - ECO 1 :    on an HP rx2620  (1.30GHz/3.0MB) running OpenVMS V8.2-1   $ sho dev dka100  F Device                  Device           Error    Volume         Free 	 Trans Mnt H   Name                   Status           Count     Label        Blocks 	 Count Cnt . $1$DKA100:     (KKP1H)  Online               0   $ sho dev dka100 /full  I Disk $1$DKA100: (KKP1H), device type HP 146 G MAW3147NC, is online, file- H      oriented device, shareable, available to cluster, error logging is  enabled.  >      Error count                    0    Operations completed 	    389095 3      Owner process                 ""    Owner UIC   [SYSTEM]2      Owner process ID        00000000    Dev Prot  S:RWPL,O:RWPL,G:R,W =      Reference count       4294967295    Default buffer size  	       512 2      Current preferred CPU Id       0    Fastpath 	         1 ;      Total blocks           286749488    Sectors per track  	        96 =      Total cylinders            31115    Tracks per cylinder  	        96 %      Allocation class               1   @    Device error count was last reset on: 21-APR-2006 16:21:32.86     $ allo dka100:< %SYSTEM-W-DEVALLOC, device already allocated to another user   $ mount/for/noassi dka100:; %MOUNT-F-DEVALLOC, device already allocated to another user   	 $ ana/sys    OpenVMS system analyzer    SDA> sho dev /chan# %CLI-W-SYNTAX, error parsing 'CHAN'  SDA> help sho dev    SHOW  	    DEVICE   I         Displays a list of all devices in the system and their associated H         data structures, or displays the data structures associated with"         a given device or devices.           Format  E           SHOW DEVICE  [device-name[:] |/ADDRESS=ucb-address|/BITMAP|   2                        /CDT=cdt_address|/CHANNELS|  7                        /HOMEPAGE|/PDT|/UCB=ucb-address]         &      Additional information available:        Parameter  QualifiersK      /ADDRESS   /CDT       /BITMAP    /CHANNELS  /HOMEPAGE  /PDT       /UCB    SDA> sho dev dka100:  H $1$DKA100 [KKP1H$DKA100]                       HP 146 G MAW3147NC  UCB:  8955BB40  6 Device status:   18000010 online,exfunc_supp,fast_path9 Characteristics: 1C454008 dir,fod,shr,avl,elg,idv,odv,rnd /                   21010201 clu,nnm,nlt,scsi,dtn ( SUD Status       00000001 path_available  E Owner UIC [000001,000004]   Operation count     389095   ORB address   8955BAC0F        PID        00000000   Error count              0   DDB address 
   894B32C0E Alloc. lock ID   00000000   Reference count   ********   DDT address   883737A0E Alloc. class            1   Online count             0   SUD address   894EBE40E Class/Type          01/36   Retry cnt/max        16/16   CRB address   89488980H Def. buf. size        512   BOFF              00000000   I/O wait queue  8955BC707 DEVDEPEND        798B6060   Byte count        00000000    7 DEVDEPND2        00000000   SVAPTE            00000000    7 DEVDEPND3        01000001   DEVSTS            00000004     FLCK index             3A    DLCK address     894B3340    Preferred CPUDB  8941A000    Preferred CPUID        00     0                   --  Device Path Information --     UCB: 8955BB40 Path: PKA0.1      Press RETURN for more.  SDA>   I/O data structures  -------------------   H                  --- Primary Class Driver Data Block (CDDB) 894AB400 ---   Status:              00000000  Controller Flags:    0000   D Allocation class       1    CDRP Queue      00000000    DDB address  894B32C0D System ID       00000000    Restart Queue   00000000    CRB address  89488980C                  00000000    DAP Count              0    CDDB link  
   00000000D Contrl. ID      00000000    Contr. timeout         0    PDT address  00000000F                  00000000    Reinit Count           0    Original UCB 
   00000000B Response ID     00000000    Wait UCB Count         0    UCB chain  00000000 MSCP Cmd status 00000000      (          *** PORT I/O queue is empty ***  *          *** DEVICE I/O queue is empty ***    +          *** I/O request queue is empty ***  SDA>   ------------------------------  # Date: Fri, 21 Apr 2006 15:22:31 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>& Subject: Re: Strange disk device state2 Message-ID: <XM62g.6653$L44.1202@news.cpqcorp.net>   Albrecht Schlosser wrote:  > Problem summary: > F > There is a disk that can't be accessed because the system gives the  > error message: > > > %SYSTEM-W-DEVALLOC, device already allocated to another user > I > although it has been dismounted successfully. Device state is "online"  3 > (not allocated). However, show device/full shows:  > " > Reference count       4294967295 >  > which is hex FFFFFFFF, see:   G    This appears to be an erroneously decremented reference count, most  & likely.  A kernel bug, in other words.  @    Since this is V8.2-1 (based on some later information in the I posting), I'd check the ECO kits and (if the system is current) I'd then  D get the support organization involved.  This definitely looks to be  worthy of a formal report.  B    I'd also look at any third-party kernel software that might be G around, too.  (That such would be a trigger here is unlikely, but I'll   mention it anyway.)   B    Short of patching the kernel, I don't know a way to clear this  without a reboot.    ------------------------------  % Date: Fri, 21 Apr 2006 18:39:19 +0200 , From: Albrecht Schlosser <ajs567@tiscali.de>& Subject: Re: Strange disk device state, Message-ID: <vr1b2e.ei7.ln@news.hus-soft.de>   Hoff Hoffman wrote:  > Albrecht Schlosser wrote:  >> Problem summary:  >>G >> There is a disk that can't be accessed because the system gives the   >> error message:  >>? >> %SYSTEM-W-DEVALLOC, device already allocated to another user  >>J >> although it has been dismounted successfully. Device state is "online" 4 >> (not allocated). However, show device/full shows: >># >> Reference count       4294967295  >> >> which is hex FFFFFFFF, see: > H >   This appears to be an erroneously decremented reference count, most ( > likely.  A kernel bug, in other words.  @ Thanks for this fast response, Hoff. That's what I thought, too.  K >   Since this is V8.2-1 (based on some later information in the posting),  I > I'd check the ECO kits and (if the system is current) I'd then get the  J > support organization involved.  This definitely looks to be worthy of a  > formal report.  L The system has been installed in Dec. 2005, but no current patches have been
 installed.  G ----------------------------------- ----------- ----------- ----------- G PRODUCT                             KIT TYPE    OPERATION   DATE AND... G ----------------------------------- ----------- ----------- ----------- G HP I64VMS TCPIP V5.5-11ECO1         Full LP     Install     22-DEC-2005 G HP I64VMS TCPIP V5.5-11             Full LP     Remove      22-DEC-2005 G HP I64VMS VMS821I_IPC V1.0          Patch       Install     22-DEC-2005 G HP I64VMS VMS821I_MUP V1.0          Patch       Install     22-DEC-2005 G HP I64VMS AVAIL_MAN_BASE V8.2-1     Full LP     Install     14-DEC-2005 G HP I64VMS CDSA V2.1-355             Full LP     Install     14-DEC-2005 G HP I64VMS DWMOTIF V1.5              Full LP     Install     14-DEC-2005 G HP I64VMS KERBEROS V2.1-75          Full LP     Install     14-DEC-2005 G HP I64VMS OPENVMS V8.2-1            Platform    Install     14-DEC-2005 G HP I64VMS TCPIP V5.5-11             Full LP     Install     14-DEC-2005 G HP I64VMS TDC_RT V2.1-78            Full LP     Install     14-DEC-2005 G HP I64VMS VMS V8.2-1                Oper System Install     14-DEC-2005 G ----------------------------------- ----------- ----------- -----------   K >   I'd also look at any third-party kernel software that might be around,  I > too.  (That such would be a trigger here is unlikely, but I'll mention  
 > it anyway.)   O No "kernel software", but, of course, application software. The concerned disk, ! however, is only used for backup.   3 The two other disks, however, use volume shadowing.   K >   Short of patching the kernel, I don't know a way to clear this without   > a reboot.   P We'll leave the system in this state at least until monday, just in case someone would have another idea    Thanks for your help   Albrecht   ------------------------------    Date: 21 Apr 2006 13:19:15 -0500. From: brooks@cuebid.zko.hp.nospam (Rob Brooks)& Subject: Re: Strange disk device state, Message-ID: <q7wVJiU8u+Pw@cuebid.zko.hp.com>  . Albrecht Schlosser <ajs567@tiscali.de> writes: > Hoff Hoffman wrote:  >>  I >>   This appears to be an erroneously decremented reference count, most  ) >> likely.  A kernel bug, in other words.   N Yup; I just checked with the engineer who deals with MOUNT, and he's not awareN of any outstanding problem in this area.  That, of course, doesn't absolve us,7 but there is no known patch kit to address this issue.    L >>   Short of patching the kernel, I don't know a way to clear this without  >> a reboot. > K > We'll leave the system in this state at least until monday, just in case  ! > someone would have another idea    There is no other idea.   @ You'll need to reboot, unless you want to patch the longword at O ucb->ucb$l_refc to be 0, instead of FFFFFFFF.  This is, of course, unsupported, $ and may lead to further instability.  O Having said that, I've done it with seemingly good results, but I'm not running E any production systems, and picking through crash dumps are a way of   life for me :-)    --    H Rob Brooks    VMS Engineering -- Exec Group     brooks!cuebid.zko.hp.com   ------------------------------  + Date: Fri, 21 Apr 2006 08:00:26 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)% Subject: Re: The Undocumented OpenVMS $ Message-ID: <e2a3eq$oa4$1@online.de>  @ In article <4448449B.EA101E7A@NeOaSrPtAhMlNiOnWk.net>, "David J.9 Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes:    : > Think: multicast powerpoint, with audio and a chat line.   ^^^^^            ^^^^^^^^^^  Isn't that a contradiction?    ------------------------------  % Date: Fri, 21 Apr 2006 04:31:47 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> % Subject: Re: The Undocumented OpenVMS , Message-ID: <44489871.8B65A698@teksavvy.com>  / Phillip Helbig---remove CLOTHES to reply wrote: < > > Think: multicast powerpoint, with audio and a chat line. >   ^^^^^            ^^^^^^^^^^  > Isn't that a contradiction?     C No. It simply means that I won't even bother checking that web site = because it is very unlikely I can view that prorietary thing.    ------------------------------  % Date: Fri, 21 Apr 2006 05:54:59 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> % Subject: Re: The Undocumented OpenVMS 9 Message-ID: <h_12g.2709$DR6.202477@news20.bellglobal.com>   ; "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message  & news:4447CCAA.F45A73AC@teksavvy.com... > Laura wrote:@ >> This one-hour Webinar will cover the undocumented features ofH >> OpenVMS--some humorous, some useful, and some you might already know. > J > Sounds interesting. But what exactly is a "one hour webinar". A bunch fo, > web pages availabel only during one hour ? > I > A Flash movie ? and Windows media stream ? Some JAVA chat application ?    Webinar = Web + Seminar.  K Back in the day it was possible to leave your desk for weeks (or more) per  M year to acquire training but it is almost impossible to do this today. As an  G alternative, you sign up for a webinar, sometimes download a free Java  M client, then join the session at the appropriate time. Many webinars support  5 sound but a few require that you call a 1-800 number.   M While this does not "usually" do a good job of replacing classroom technical  J training, it can be an adjunct to the technical overviews associated with  supporting sales.   M p.s. If you sign-up for a dud, it is easy to exit and continue with the rest  M of your day. OpenVMS webinars (like those associated with Parsec) are almost  K always worth while. Also, I can only assume that things like webinar "sign  E up stats" are probably sent to marketing folks to indicate whether a  , particular market is worth pursuing (or not)  
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html: http://www3.sympatico.ca/n.rieck/links/openvms_demos.html    ------------------------------  # Date: Fri, 21 Apr 2006 13:16:50 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>% Subject: Re: The Undocumented OpenVMS 1 Message-ID: <6X42g.6645$FI3.634@news.cpqcorp.net>    Neil Rieck wrote: = > "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message  ( > news:4447CCAA.F45A73AC@teksavvy.com... >> Laura wrote: A >>> This one-hour Webinar will cover the undocumented features of I >>> OpenVMS--some humorous, some useful, and some you might already know. K >> Sounds interesting. But what exactly is a "one hour webinar". A bunch fo - >> web pages availabel only during one hour ?  >>J >> A Flash movie ? and Windows media stream ? Some JAVA chat application ? >  > Webinar = Web + Seminar. ...   H    The Halo conferencing system is certainly immersive -- think of some C of the classic surround-style theme park entertainment -- see some  & details at: <http://www.hp.com/halo/>.  E    As for the upcoming OpenVMS undocumented features presentation, I  G have the usual piles of new slides (which means a similarly large pile  I of old slides now have to go, else Sue has to further increase the hours  F that can fit into day) for the upcoming Boot Camp presentations.  New ? hints, new kinks, updates for V8.3, and for recent engineering  G activities.  And another couple of days to figure out how to stuff all  H of the new material into the time-slot alloted.  (The bootcamp sessions 7 will not be available on nor offered on Halo, however.)    ------------------------------    Date: 21 Apr 2006 08:49:29 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) % Subject: Re: The Undocumented OpenVMS 3 Message-ID: <UeMbpB3NsLEp@eisner.encompasserve.org>   y In article <4448449B.EA101E7A@NeOaSrPtAhMlNiOnWk.net>, "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes:  > JF Mezei wrote:  >>   >> Laura wrote: B >> > This one-hour Webinar will cover the undocumented features ofJ >> > OpenVMS--some humorous, some useful, and some you might already know. >>  K >> Sounds interesting. But what exactly is a "one hour webinar". A bunch fo - >> web pages availabel only during one hour ?  >>  J >> A Flash movie ? and Windows media stream ? Some JAVA chat application ? > : > Think: multicast powerpoint, with audio and a chat line.  C And perhaps Javascript, given their demonstrated security attitude.    ------------------------------  # Date: Fri, 21 Apr 2006 13:45:59 GMT " From:   VAXman-  @SendSpamHere.ORG% Subject: Re: The Undocumented OpenVMS 0 Message-ID: <00A548B0.DBC1E71A@SendSpamHere.ORG>  y In article <4448449B.EA101E7A@NeOaSrPtAhMlNiOnWk.net>, "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes:  >  >  >JF Mezei wrote: >>   >> Laura wrote: B >> > This one-hour Webinar will cover the undocumented features ofJ >> > OpenVMS--some humorous, some useful, and some you might already know. >>  K >> Sounds interesting. But what exactly is a "one hour webinar". A bunch fo - >> web pages availabel only during one hour ?  >>  J >> A Flash movie ? and Windows media stream ? Some JAVA chat application ? > 9 >Think: multicast powerpoint, with audio and a chat line.    Is it really powerpuke?   N I know HPs webinars that they keep excluding me from attending require the useO of Billzebub-warez Infernonet Exploiter but I was able to pass the requirements M for Parsec's webinar using my Safari borwser on my Powerbook running OS X.  I L didn't get to attend the last seminar tho.  I'm looking forward to this next9 one  only to see what's actually involved in a "webinar".    --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------    Date: 20 Apr 2006 23:40:59 -0700 From: tomarsin2015@comcast.net( Subject: Re: VAX hardware error code 3 ?C Message-ID: <1145601659.483548.235360@v46g2000cwv.googlegroups.com>   E According to the manual the number following the question mark is the  most important information/ 18..17..16..15...14..13..12..11..10..09..08..07 F This is the line you need to look for ------->?5F 2 15 FF 0000 0000 02 ; SUBTEST_5F_15, DE_SGEC.LIS P1=00000000 P2ETC  06..05..04..03.. Normal operation not possible.   ------------------------------  # Date: Fri, 21 Apr 2006 12:52:04 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>( Subject: Re: VAX hardware error code 3 ?1 Message-ID: <Uz42g.6644$X04.760@news.cpqcorp.net>    tomarsin2015@comcast.net wrote: G > According to the manual the number following the question mark is the  > most important information1 > 18..17..16..15...14..13..12..11..10..09..08..07 H > This is the line you need to look for ------->?5F 2 15 FF 0000 0000 02 > ; SUBTEST_5F_15, DE_SGEC.LIS > P1=00000000 P2ETC  > 06..05..04..03..  > Normal operation not possible. >   D    Based solely on the diagnostic name, there's some sort of a hard B failure within the on-board Ethernet hardware.  Assuming that the G address PROM is present, there probably isn't much you can do for this  ) problem without also involving SMT tools.   I   In traditional DECspeak, "SGEC" is the Second Generation Ethernet Chip.    ------------------------------  % Date: Fri, 21 Apr 2006 07:52:39 -0700  From: foo@bar.com & Subject: Re: VAX Q-bus and SCSI Disks?8 Message-ID: <0bsh42p5u5o3nl008fpbologt48c37nl4c@4ax.com>  0 On Thu, 20 Apr 2006 23:15:27 -0400, "Neil Rieck" <n.rieck@sympatico.ca> wrote:    > < >"JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message ' >news:444819B3.8A02BD0A@teksavvy.com...  >> Neil Rieck wrote:H >>> Thank god some machines (like the uVAX-4300) had a CONFIGURE utilityI >>> available in console mode to help "compute" all those CSR addresses +  >>> vectors. >>I >> Thanks to the VMS engineers who included a CONFIGURE utility in SYSGEN M >> available at the $ sign to help compute all those CSR addresses + vectors.  >> >>H >> Whether "VMS Engineers" = "god" is a question I am not sure I want toK >> get into :-) But I know for a fact that my all mighty Microvax II viewed K >> VMS engineers as GODs since they were the creators of the stuff that ran I >> on it. But contrary to V'GER, it didn't seek to meet with its creator, % >> destroying everything in its path.  > J >Yeah but before someone decides to kill me with a suicide bomber, please / >note that I did not capitalize the letter "g".     E Just make sure you don't create any funny cartoons with your Microvax F II or VMS Engineers or we will be forced to riot and burn down the New Hampshire embassy!* ------------------------------------------C NO, use John Mee3 at comcast dot net (remove the spaces and do the  # obvious with the "dot" and the "at"    ------------------------------   End of INFO-VAX 2006.221 ************************                                                                                                                                                                                                                                                /decus/freewarev40/xv310a/jpeg/jversion.h (667 bytes) started.9 >>> 226 Transfer completed.  371 (8) bytes transferred.> <<< PORT 80,12,133,79,11,2013 >>> 200 Port 11.20 at Host 80.12.133.79 accepted.S <<< RETR jutils.c$c >>> 150 ASCII retrieve of /disk$misc/decus/freewarev40/xv310a/jpeg/jutils.c (4528 bytes) started. : >>> 226 Transfer completed.  3795 (8) bytes transferred. <<< PORT 80,12,133,79,11,21t3 >>> 200 Port 11.21 at Host 80.12.133.79 accepted.> <<< RETR jquant2.ce >>> 150 ASCII retrieve of /disk$misc/decus/freewarev40/xv310a/jpeg/jquant2.c (46985 bytes) started.d; >>> 226 Transfer completed.  47452 (8) bytes transferred.1 <<< PORT 80,12,133,79,11,2283 >>> 200 Port 11.22 at Host 80.12.133.79 accepted.5 <<< RETR jquant1.ce >>> 150 ASCII retrieve of /disk$misc/decus/freewarev40/xv310a/jpeg/jquant1.c (29098 bytes) started.3; >>> 226 Transfer completed.  28996 (8) bytes transferred. <<< PORT 80,12,133,79,11,2323 >>> 200 Port 11.23 at Host 80.12.133.79 accepted.S <<< RETR jpeglib.he >>> 150 ASCII retrieve of /disk$misc/decus/freewarev40/xv310a/jpeg/jpeglib.h (38656 bytes) started.4; >>> 226 Transfer completed.  39079 (8) bytes transferred. <<< PORT 80,12,133,79,11,2423 >>> 200 Port 11.24 at Host 80.12.133.79 accepted.S <<< RETR jpegint.he >>> 150 ASCII retrieve of /disk$misc/decus/freewarev40/xv310a/jpeg/jpegint.h (14630 bytes) started.9; >>> 226 Transfer completed.  14403 (8) bytes transferred. <<< PORT 80,12,133,79,11,2523 >>> 200 Port 11.25 at Host 80.12.133.79 accepted.S <<< RETR jmorecfg.hif >>> 150 ASCII retrieve of /disk$misc/decus/freewarev40/xv310a/jpeg/jmorecfg.h (12106 bytes) started.; >>> 226 Transfer completed.  11785 (8) bytes transferred.> <<< PORT 80,12,133,79,11,2613 >>> 200 Port 11.26 at Host 80.12.133.79 accepted.I <<< RETR jmemsys.hd >>> 150 ASCII retrieve of /disk$misc/decus/freewarev40/xv310a/jpeg/jmemsys.h (8564 bytes) started.: >>> 226 Transfer completed.  8002 (8) bytes transferred. <<< PORT 8