INFO-VAX Wed, 20 Jun 2007 Volume 2007 : Issue 334 Contents: Re: 8086 vs patches Re: Amazing, two new articles on Computerworld.com actual mention OpenVMS! O Re: Amazing, two new articles on Computerworld.com actual mention OpenVMS! Ope Re: Amazing, two new articles on Computerworld.com actual mention OpenVMS! Re: Amazing, two new articles on Computerworld.com actual mention OpenVMS! Re: Amazing, two new articles on Computerworld.com actual mention OpenVMS! OpenV Re: Anyone using HDS XP Series Arrays? Re: Anyone using HDS XP Series Arrays? Re: Anyone using HDS XP Series Arrays? Re: Console monitoring programs help backing up nfs drives (dnfs) Re: help backing up nfs drives (dnfs) Re: help set host/scsi Re: Installing C compiler from Hobbyist CD Re: Is NOONE using XP10000 & OpenVMS??? Re: Itanium hardware Q Re: NFS - what am I doing wrong? Re: NFS - what am I doing wrong? Re: NFS - what am I doing wrong? Re: NFS - what am I doing wrong? OpenVMS - When downtime is not an option Re: OpenVMS - When downtime is not an option Re: OpenVMS - When downtime is not an option Re: OpenVMS - When downtime is not an option Re: OpenVMS - When downtime is not an option Re: OpenVMS - When downtime is not an option Re: OpenVMS - When downtime is not an option Re: OpenVMS - When downtime is not an option Re: OpenVMS - When downtime is not an option Re: Question for the Group Re: Question for the Group Re: Same system disk for multiple architectures ? Re: Same system disk for multiple architectures ? Re: Top ten dead or dying computer skills (guess what's NOT on the list!) Re: Top ten dead or dying computer skills (guess what's NOT on the list!) list!) Re: UCX Printer connection Via LPD Re: UCX Printer connection Via LPD Re: UCX Printer connection Via LPD Re: UCX Printer connection Via LPD Re: UCX Printer connection Via LPD wchar_t confusion Re: wchar_t confusion Re: wchar_t confusion Re: wchar_t confusion ---------------------------------------------------------------------- Date: Tue, 19 Jun 2007 23:33:12 -0800 From: glen herrmannsfeldt Subject: Re: 8086 vs patches Message-ID: Ron Johnson wrote: > And, being static, allows you to restart your program from where it left > off when the power failed. *Very* handy. It is not always handy. It caused a major hardware bug in one IBM computer, maybe the 704 or 709. If an execute instruction tried to execute itself, the processor went into an infinite loop. Power down and restart, it continued in the same loop. (The program counter is also in core.) The solution involves inserting magnets inside the core memory box. This is the reason the S/360 EX is not allowed to execute another EX instruction. -- glen ------------------------------ Date: Wed, 20 Jun 2007 09:19:54 +0200 From: Michael Kraemer Subject: Re: Amazing, two new articles on Computerworld.com actual mention OpenVMS! O Message-ID: JF Mezei schrieb: > > If that Livermore isn't an official announcement that HP intends to kill > VMS, I don't know what it. Relax. That's more or less what she said already 2 years ago, nothing new here. ------------------------------ Date: Wed, 20 Jun 2007 06:13:48 -0400 From: JF Mezei Subject: Re: Amazing, two new articles on Computerworld.com actual mention OpenVMS! Ope Message-ID: <632b3$4678fded$cef8887a$12760@TEKSAVVY.COM> Michael Kraemer wrote: > Relax. That's more or less what she said already 2 years ago, > nothing new here. Confirmation that HP has a "stay the course" policy and won't change it. It is perfectly happy to actively sink the VMS ship (along with itanic). Getting Cerner to drop VMS in favour of HP-UX as well as Livermore's sayings are no longer just passive "let VMS live happily in its corner". HP is now actively sending signals that VMS is on its way out. And what is new is that major ISVs have begun to abandon ship. I fear this is the last chance to save VMS. The next couple of months will be critical to try to reverse HP's policy. And out of principle, we must ensure this causes the biggest possible loss of customers to HP. If customers decide to remain with HP despite this action, it will set a very dangerous precedent for other vendors to handle customers in unesthical ways and lie to them about tru strategies for certain products. Our only hope is to convince Hurd that he has been mislead by Stallard/Livermore and to overrule their desire to act as pirates and plunder as many VMS customers as possible onto the HP-UX ship. ------------------------------ Date: 20 Jun 2007 13:47:49 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: Amazing, two new articles on Computerworld.com actual mention OpenVMS! Message-ID: <5dspg5F36dckuU5@mid.individual.net> In article , "Tom Linden" writes: > On Tue, 19 Jun 2007 13:58:39 -0700, JF Mezei > wrote: > >> Ken Robinson wrote: >> >>> Posted with no comment.... >> >> Well, did anyone think that *I* would be able to widthhold comments >> about this ? >> >> Ann Livermore: >> ## >> In particular for our OpenVMS customers, we are very focused on our >> installed base. We want our installed base to be happy, and if or when >> they ever want to migrate, we want them to migrate to another HP >> platform. >> ## > > From a bottom line perspective this is really stupid. Why would HP want > to migrate a > high margin VMS customer to a lower margin HP-UX customer, or even lower > Winodws. Can > you think of any other business that operates in this fashion? > >> >> This pretty much confirms the Stallard memo of May 7th 2002 not being a >> glitch but truly representing HP corporate policy. >> >> If that Livermore isn't an official announcement that HP intends to kill >> VMS, I don't know what it. > > It demonstrates a serious lack of business acumen. Well, over the summer I am taking a "Stat for Business" course. We are into "Decision Making" as of last night. Of course, the examples we work with are all business related. And. if what I am learning is really how they do it, I am not surprised that American businesses are so far in the crapper. If I applied the logic used in these methods to VMS it would have been killed long ago. for example, we had a thing on "Deicision making when no probability data is available." There are three methods. One is extremely pess- imistic. One is extremely optomistic. And the third looks totally random and could probably be done just as well with a dartboard with out all the paper and pencil work. Hmmmm...... Maybe I can ask some questions here and then see if I can lay it out according to the methods in my textbook. Then everyone can either laugh, cry or both. 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, 20 Jun 2007 08:12:47 -0700 From: AEF Subject: Re: Amazing, two new articles on Computerworld.com actual mention OpenVMS! Message-ID: <1182352367.904672.128230@q75g2000hsh.googlegroups.com> On Jun 19, 4:58 pm, JF Mezei wrote: > Ken Robinson wrote: > > Posted with no comment.... > > Well, did anyone think that *I* would be able to widthhold comments > about this ? > > Ann Livermore: > ## > In particular for our OpenVMS customers, we are very focused on our > installed base. We want our installed base to be happy, and if or when > they ever want to migrate, we want them to migrate to another HP platform. > ## > > This pretty much confirms the Stallard memo of May 7th 2002 not being a > glitch but truly representing HP corporate policy. > > If that Livermore isn't an official announcement that HP intends to kill > VMS, I don't know what it. C'mon. You're putting the worst possible spin on it. She says she wants their installed base to be happy, AND if or when, not "now". So that means if the customers want to stay with VMS, she wants them to be happy and they can stay with VMS. Now if she said "we want to move them off VMS", then you'd have a case. AEF ------------------------------ Date: Wed, 20 Jun 2007 13:22:17 -0400 From: Bill Todd Subject: Re: Amazing, two new articles on Computerworld.com actual mention OpenVMS! OpenV Message-ID: AEF wrote: ... > that means if the customers want to stay with VMS, she wants them to > be happy Of course: right up until the day that the same thing happens to VMS that happened to Alpha, and for the same reasons. She surely wouldn't want to preannounce VMS's demise while people are still willing to pay good money for it - far better to string things out until what they're paying is no longer sufficient to keep up even a pretense. As contrasted, of course, with actually *investing* in the future of a product that *has* an intended future. VMS is certainly not dead yet, and for some people that's enough. It's just the people for whom that is *not* enough who should delve behind the happy talk - not that this has been all that difficult since the Alphacide: the signs of senescence have been plentiful, and if it weren't for a continually diminishing but not yet completely irrelevant cadre of faithful friends the family would have pulled the plug long ago. - bill ------------------------------ Date: Wed, 20 Jun 2007 04:21:14 -0700 From: BaxterD@tessco.com Subject: Re: Anyone using HDS XP Series Arrays? Message-ID: <1182338474.658032.79150@u2g2000hsc.googlegroups.com> On Jun 19, 2:52 pm, p...@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) wrote: > > And in case of FC storage, you know SYSMAN IO FIND_WWID/LIST_WWID/REPLACE_WWID? > Peter, ------------------------------ Date: Wed, 20 Jun 2007 04:22:24 -0700 From: BaxterD@tessco.com Subject: Re: Anyone using HDS XP Series Arrays? Message-ID: <1182338544.315817.292610@w5g2000hsg.googlegroups.com> > > And in case of FC storage, you know SYSMAN IO FIND_WWID/LIST_WWID/REPLACE_WWID? > Hi Peter, Sorry about that last post, I was about to observe that these commands are for Tapes only. Dave. ------------------------------ Date: Wed, 20 Jun 2007 13:03:27 GMT From: "John E. Malmberg" Subject: Re: Anyone using HDS XP Series Arrays? Message-ID: Hal Kuff wrote: > Hi folks... answer via the now great Mr Wilson's folks is that you need LUN > 0 presented to the OpenVMS systems... nowhere in the DOC for XP or the > OpenVMS guide to XP Arrays... (Not that a mere mortal could find anything > through the HP web site search) A LUN 0 is required per the SCSI specification. Older versions of VMS required it to be a storage array controller, which it what it is on all supported storage arrays. Newer versions may allow it to be a storage device, however that has never been qualified, so is an undocumented feature that a hobbyist may find useful. There should be an e-mail address in the XP array documentation to send feedback to for corrections. If a storage array does not present a LUN 0, there is no way for a host to configure it, unless that host has a device specific driver that knows a non-standard way to do that configuration. Using brute force to find all LUNs in a fibre SAN with out querying LUN 0 for each potential LUN can take hours. The draft specifications for SCSI and Fibre are online at www.t10.org and www.t11.org. -John wb8tyw@qsl.network Personal Opinion Only ------------------------------ Date: Wed, 20 Jun 2007 12:11:57 +0200 From: "P. Sture" Subject: Re: Console monitoring programs Message-ID: In article <1182306343.314174.221480@w5g2000hsg.googlegroups.com>, "johnhreinhardt@yahoo.com" wrote: > On Jun 19, 9:38 pm, David J Dachtera > wrote: > > "johnhreinha...@yahoo.com" wrote: > > > > > Has anyone attempted to port Conserver (www.conserver.com) to > > > OpenVMS? Or is there any free or open-source or hobbyist licensed > > > replacement for the Polycenter Cluster Console application? > > > > Good question. > > > > Gotta lookup whether Tecsys Development has a hobbyist license for > > ConsoleWorks... (http://www.tditx.com/) > > I looked after their product came up in a Google search, but I didn't > see any indication of one. They were offering a license as a prize in > contest during the Bootcamp this year, but I didn't attend so I don't > know who, if anyone, got it. > I had a ConsoleWorks Hobbyist license back in 2003. I asked for it via email. IIRC you need to be specific about what configuration you are going to use it with so that the version they give you is useful. -- Paul Sture ------------------------------ Date: Wed, 20 Jun 2007 01:54:00 -0700 From: "tomarsin2015@comcast.net" Subject: help backing up nfs drives (dnfs) Message-ID: <1182329640.621551.232890@i13g2000prf.googlegroups.com> Hello I'm trying to backup some dnfsXXX drives using VMS backup enstead of BackupExec. When I use this command, I get the following error OCTANS>backup/image/media=comp/ignore=interlock/noassist/log dnfs3: $99$mkb700:adonis.bck/sav %MOUNT-I-MOUNTED, ADONIS mounted on _$99$MKB700: (ZULU) %BACKUP-F-PROCINDEX, error processing index file on DNFS3:, RVN 1 -SYSTEM-W-NOSUCHFILE, no such file So I tried this OCTANS>backup/init/log/ignore=interlock/noassist dnfs3:[000000...]*.* $99$mkb700:adonis.bck/sav %MOUNT-I-MOUNTED, ADONIS mounted on _$99$MKB700: (ZULU) %BACKUP-E-OPENDIR, error opening directory DNFS3:[000000] -SYSTEM-F-BADATTRIB, bad attribute control list %BACKUP-S-COPIED, copied DNFS3:[000000]000000.DIR;1 Then backup drops to the command prompt. Is there a way to backup DNFS devices using VMS backup?? tks phil ------------------------------ Date: Wed, 20 Jun 2007 02:05:52 -0700 From: IanMiller Subject: Re: help backing up nfs drives (dnfs) Message-ID: <1182330352.733785.280410@c77g2000hse.googlegroups.com> you can't do /IMAGE backups of DNFS drives. Do [000000....]*.*;* instead. BACKUP/LOG/NOASSIST DNFS3:[000000...]*.*;* $99$MKB700:ADONIS.BCK/SAVE/ INIT /IGNORE=INTERLOCK is not a good idea either. You should also look at /BLOCK=65256/MEDIA=COMPACT too. ------------------------------ Date: Wed, 20 Jun 2007 11:14:56 -0400 From: Stephen Hoffman Subject: Re: help set host/scsi Message-ID: Peter 'EPLAN' LANGSTOeGER wrote: > ...Don't know what /BTS is... Batch Terminal Simulator. It's a classic-vintage IBM terminal scheme. If you poke around, you'll find a few other terminal-oriented commands (also) have /BTS qualifiers. -- www.HoffmanLabs.com Services for OpenVMS ------------------------------ Date: Wed, 20 Jun 2007 12:16:50 +0200 From: "P. Sture" Subject: Re: Installing C compiler from Hobbyist CD Message-ID: In article , "Craig A. Berry" wrote: > In article <1182305768.160083.279130@e9g2000prf.googlegroups.com>, > rtk wrote: > > > I asked a question about licenses a couple of days back and was very > > happy to see all the helpful replies! Thank you! > > > > I'm now trying to install the C compiler on the Alpha Hobbyist CD. I > > got the FORTRAN compiler installed with PRODUCT INSTALL FORTRAN / > > SOURCE=... but when I try to get C installed I don't seem to know the > > product name (nor does help reveal anything to me about how to find > > it) > > > > I'm assuming the directory in the KITS directory called CC065_KIT is > > the right directory to use but I've tried C, CC, CC065, etc as the > > name and it never works. I looked around online and in the newsgroup > > but no one seems to have run into the same problem. > > > > What am I missing? (this time :) > > The C compiler still supports the older, clunkier VMSINSTAL > installation method and, as far as I know, does not support the newer > PCSI method. I think it has to do that so it can still be installed in > truly ancient VMS versions. Basically you want to poke around in the > documentation for VMSINSTAL. The following example is the first thing > that came up in Google: > > http://h71000.www7.hp.com/DOC/82final/6398/6398pro_002.html > > Or, something like the following should do the trick: > > $ @sys$update:vmsinstal cc065 and in one go including the source directory: $ @sys$update:vmsinstal cc065 source_dev:[source_dir] (that's the device and directory where your CC065.A kit is) -- Paul Sture ------------------------------ Date: Wed, 20 Jun 2007 04:15:57 -0700 From: BaxterD@tessco.com Subject: Re: Is NOONE using XP10000 & OpenVMS??? Message-ID: <1182338157.694389.193040@c77g2000hse.googlegroups.com> On Jun 19, 9:35 pm, David J Dachtera wrote: > Rob Brooks wrote: > > > Baxt...@tessco.com writes: > > > We have an issue with displaying multipath devices presented from > > > an XP10000 to OpenVMS 7.3-2 (Alpha) and OpenVMS 8.3 (IA64). > > > > We would love to hear from ANYone who is running XP storage with > > > OpenVMS!! > To all who replied, thanks. The main reason for our post was the difficulty we were having actually getting anyone from HP who knew XP with OpenVMS, or finding anyone at Hitachi who even heard of OpenVMS. We finally managed to talk to a very knowledgible guy (from VMS Engineering, I think) called Brian, who talked us through the problem, and brought it to a resolution in probably less than 15 minutes. Problem was that the Console LUNs were not being correctly presented to the OpenVMS hosts. Executing a "show dev gg" showed no devices, and this identified the problem. Once this was resolved, the paths now show up (after the usual "sysman io auto", etc.) thanks Dave. ------------------------------ Date: Wed, 20 Jun 2007 02:43:49 -0700 From: IanMiller Subject: Re: Itanium hardware Q Message-ID: <1182332629.916033.212530@u2g2000hsc.googlegroups.com> On Jun 20, 12:00 am, Robert Deininger wrote: > > For V8.3, it's documented. (Look for the "errata" document for VMS on > blades.) I'll not try to repeat the instructions here. > > For V8.3-1H1, the process will be streamlined. > > -- Robert See http://h71000.www7.hp.com/doc/hardware.html and http://h71000.www7.hp.com/openvms/cclass_support.html ------------------------------ Date: Wed, 20 Jun 2007 06:28:22 -0700 From: Don Subject: Re: NFS - what am I doing wrong? Message-ID: <1182346102.001691.133410@k79g2000hse.googlegroups.com> On Jun 18, 4:15 pm, "P. Sture" wrote: > In article <1182194671.275544.165...@n60g2000hse.googlegroups.com>, > > Don wrote: > > I am trying to give NFS access to a directory on an ODS-2 disk. In > > testing the setup, I tried to NFS mount the volume, and got an error > > message that puzzles me: > > > GREGG> TCPIP > > TCPIP> MOUNT DNFS0: /PATH="/SCORECARD/USER.DIR.1"/HOST=127.0.0.1 > > %TCPIP$DNFSMOUNT-E-MOUNTFAIL, error mounting _DNFS8:[000000] > > -SYSTEM-F-NOSUCHOBJ, network object is unknown at remote node > > Try this instead: > > MOUNT DNFS0: /PATH="/SCORECARD/USER"/HOST=127.0.0.1 > > i.e. you don't want the .DIR;1 > > You might want a /SYSTEM on the end of that command too - I forget it > one is necessary. > > -- > Paul Sture Thanks for the response Paul. I tried all of the combinations of user, user.dir, and user.dir;1 and nothing worked. The netword traffic doesn't show the path string anywhere, so I am guessing that the failure happens before the path is validated. Every mount command I issue gives a "-SYSTEM-F-NOSUCHOBJ" error. I am an automation engineer being pressed into sysadmin service. I am fluent with VMS as a user (20+ years), but my admin experience is spotty. Suggestions are appreciated, Don ------------------------------ Date: Wed, 20 Jun 2007 13:44:17 GMT From: =?ISO-8859-1?Q?Jan-Erik_S=F6derholm?= Subject: Re: NFS - what am I doing wrong? Message-ID: Don wrote: > Every mount command I issue gives a "-SYSTEM-F-NOSUCHOBJ" error. Is the "object" you are trying to mount available as an NFS-mountable "object" ? Is NFS *itself* available as an "object" ? ("service", "server" or whatever...) Jan-Erik. ------------------------------ Date: Wed, 20 Jun 2007 07:46:10 -0700 From: Don Subject: Re: NFS - what am I doing wrong? Message-ID: <1182350770.557856.29900@n2g2000hse.googlegroups.com> Jan-Erik, I believe everything exists: TCPIP> SHOW SERV Service Port Proto Process Address State FTP 21 TCP TCPIP$FTP 0.0.0.0 Enabled MOUNT 10 TCP TCPIP$MOUNTD 0.0.0.0 Enabled NFS 2049 UDP TCPIP$NFS 0.0.0.0 Enabled PORTMAPPER 111 TCP,UDP TCPIP$PORTM 0.0.0.0 Enabled TELNET 23 TCP not defined 0.0.0.0 Enabled TCPIP> TCPIP> SHOW MAP Dynamic Filesystem Map Pathname Logical File System /SCORECARD DKA7: TCPIP> TCPIP> SHOW EXPORT File System Host name /SCORECARD/USER * TCPIP> TCPIP> MOUNT DNFS0: /PATH="/SCORECARD/USER"/HOST=127.0.0.1 %TCPIP$DNFSMOUNT-E-MOUNTFAIL, error mounting _DNFS20:[000000] -SYSTEM-F-NOSUCHOBJ, network object is unknown at remote node TCPIP> ------------------------------ Date: Wed, 20 Jun 2007 17:09:24 +0200 From: Joseph Huber Subject: Re: NFS - what am I doing wrong? Message-ID: Don wrote: > TCPIP> MOUNT DNFS0: /PATH="/SCORECARD/USER"/HOST=127.0.0.1 > %TCPIP$DNFSMOUNT-E-MOUNTFAIL, error mounting _DNFS20:[000000] > -SYSTEM-F-NOSUCHOBJ, network object is unknown at remote node > TCPIP> > If everything else is configured (tcpip$config) and started (tcpip$nfs_startup), it could be the "/HOST=127.0.0.1" which does not work. What purpose should it have to NFS mount on the local/loopback address ? Can You test the same mount from a different node, specifying the real IP address of the VMS server node ? -- Joseph Huber - http://www.huber-joseph.de ------------------------------ Date: Wed, 20 Jun 2007 06:53:54 -0700 From: IanMiller Subject: OpenVMS - When downtime is not an option Message-ID: <1182347634.410293.195610@m36g2000hse.googlegroups.com> See the video at http://www.hp.com/go/disasterproof and see which operating system recovers first. I wonder about the disaster tolerant fish though :-) ------------------------------ Date: 20 Jun 2007 14:10:27 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: OpenVMS - When downtime is not an option Message-ID: <5dsqqiF35ot2sU1@mid.individual.net> In article <1182347634.410293.195610@m36g2000hse.googlegroups.com>, IanMiller writes: > See the video at > > http://www.hp.com/go/disasterproof > > and see which operating system recovers first. I wonder about the > disaster tolerant fish though :-) Why does it require cookies just to watch the video? 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: 20 Jun 2007 15:36:24 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: OpenVMS - When downtime is not an option Message-ID: <5dsvroF36fkohU1@mid.individual.net> In article , koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: > In article <5dsqqiF35ot2sU1@mid.individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: >> In article <1182347634.410293.195610@m36g2000hse.googlegroups.com>, >> IanMiller writes: >>> See the video at >>> >>> http://www.hp.com/go/disasterproof >>> >>> and see which operating system recovers first. I wonder about the >>> disaster tolerant fish though :-) >> >> Why does it require cookies just to watch the video? >> > > I didn't have cookies on. (Firefox 2.0.0.4 Windows XP). Hey, I didn't make it up. Here's what I got: "To view the HP-ETV, you must have browser Cookies enabled, Macromedia Flash Player, and a supported Browser/Operating System/Media Player configuration." I had to go and look at it on an XP box. I have to admit, it was even funnier than the list of no longer needed skills. I wonder if this is HP's submission to that new Reality Show about finding Movie Producers? 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, 20 Jun 2007 12:20:11 -0400 From: "David Turner, Island Computers" Subject: Re: OpenVMS - When downtime is not an option Message-ID: Did notice VMS took 13 secs and Windows >100 seconds Makes a big difference in the stock market, non???? DT "Bill Gunshannon" wrote in message news:5dsvroF36fkohU1@mid.individual.net... > In article , > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: >> In article <5dsqqiF35ot2sU1@mid.individual.net>, bill@cs.uofs.edu (Bill >> Gunshannon) writes: >>> In article <1182347634.410293.195610@m36g2000hse.googlegroups.com>, >>> IanMiller writes: >>>> See the video at >>>> >>>> http://www.hp.com/go/disasterproof >>>> >>>> and see which operating system recovers first. I wonder about the >>>> disaster tolerant fish though :-) >>> >>> Why does it require cookies just to watch the video? >>> >> >> I didn't have cookies on. (Firefox 2.0.0.4 Windows XP). > > Hey, I didn't make it up. Here's what I got: > > "To view the HP-ETV, you must have browser Cookies enabled, > Macromedia Flash Player, and a supported Browser/Operating > System/Media Player configuration." > > I had to go and look at it on an XP box. I have to admit, it was > even funnier than the list of no longer needed skills. I wonder > if this is HP's submission to that new Reality Show about finding > Movie Producers? > > 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, 20 Jun 2007 17:20:31 +0100 From: Richard Subject: Re: OpenVMS - When downtime is not an option Message-ID: On Wed, 20 Jun 2007 06:53:54 -0700, IanMiller wrote: >See the video at > >http://www.hp.com/go/disasterproof > >and see which operating system recovers first. I wonder about the >disaster tolerant fish though :-) Good - but not as good as the Oracle video where Larry Ellison fires thunderbolts at an Oracle RAC cluster. (and conveniently ignores the fact that the blast and electric field would have wrecked half his data centre). I'd be interested in hearing from anybody who uses HP's 'Extended Distance Clustering' in conjunction with Oracle RAC for high availability databases. ------------------------------ Date: Wed, 20 Jun 2007 16:26:30 -0000 From: IanMiller Subject: Re: OpenVMS - When downtime is not an option Message-ID: <1182356790.811635.277740@q75g2000hsh.googlegroups.com> I also though it interesting that the non-stop system took longer than VMS to failover. The linux platform was the slowest [except for the fish] ------------------------------ Date: 20 Jun 2007 16:54:48 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: OpenVMS - When downtime is not an option Message-ID: <5dt4enF35qc8vU1@mid.individual.net> In article <1182356790.811635.277740@q75g2000hsh.googlegroups.com>, IanMiller writes: > I also though it interesting that the non-stop system took longer than > VMS to failover. > > The linux platform was the slowest [except for the fish] Probably because it didn't failover. They just changed the IP address and re-booted it to take the place of the other system. :-) Like I said, it was funny. It is so contrived as to be almost absurd. Like the "bulletproof" one. Let me pick the path of the bullet and I can shoot a hole in my home built RAID arrays without a hiccup, too. Or, go back to the original "bulletproof" add, Master Locks. Stood up to a 30.06 shell but I can open it with the nail file in my pocket. What they would really need are some verifiable, real world testimonials from people who's datacenters have survived disasters like floods or tornadoes or earthquakes. (I used to work for Martin Marietta. They had a "hardened" data center down in Orlando. They once had a huricane blow thru town. The people who worked nthe data center didn't know about it til they came out at the end of the day to find all the damage to their cars in the parking lot. No, no VMS. It was all IBM big iron.) Or, as I also said recently, they should be out here selling VMS to all the people who want the NYC-NEPA mirrored operation that is getting all the great press around here right now. Sell your product and ride the wave!! Like that is really going to happen. :-( 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: 20 Jun 2007 12:26:49 -0500 From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Subject: Re: OpenVMS - When downtime is not an option Message-ID: In article , koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: > In article <5dsqqiF35ot2sU1@mid.individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: >> In article <1182347634.410293.195610@m36g2000hse.googlegroups.com>, >> IanMiller writes: >>> See the video at >>> >>> http://www.hp.com/go/disasterproof >>> >>> and see which operating system recovers first. I wonder about the >>> disaster tolerant fish though :-) >> >> Why does it require cookies just to watch the video? >> > > I didn't have cookies on. (Firefox 2.0.0.4 Windows XP). I take that back. Firefox cookies did get turned on. I suppose that's what I get for letting someone force Windows on me. ------------------------------ Date: Wed, 20 Jun 2007 17:47:39 GMT From: VAXman- @SendSpamHere.ORG Subject: Re: OpenVMS - When downtime is not an option Message-ID: <%Kdei.14$DV4.12@newsfe12.lga> In article <1182347634.410293.195610@m36g2000hse.googlegroups.com>, IanMiller writes: > > >See the video at > >http://www.hp.com/go/disasterproof > >and see which operating system recovers first. I wonder about the >disaster tolerant fish though :-) To view the HP-ETV, you must have browser Cookies enabled, Macromedia Flash Player, and a supported Browser/Operating System/Media Player configuration. For additional system requirements and further help, Click Here... Which yeilds: Q: What do I need to view this site? A: Your computer must meet the following requirements to successfully view this site: Operating System Requirements Microsoft Windows XP (SP1, SP2), 2000, 98 (SE) OR Macintosh OSX ^^^^^^^^^^^^^-- got this Browser Requirements For PC users we recommend the latest versions of: Internet Explorer, FireFox, Mozilla OR Netscape For Mac users we recommend the latest versions of: Safari, FireFox OR Mozilla ^^^^^^^^^^^^^^^-----^^^^^^-- got this Media Player Requirements Windows Media Player 9+(for PC IE users) OR Macromedia Flash 7.0.25 *** Shockwave Flash 9.0 r28... does anybody know if this will do??? Cookies must be enabled for this site. For more information see How do I Enable Cookies? *** Yeah,, they're on. -- VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)COM "Well my son, life is like a beanstalk, isn't it?" ------------------------------ Date: 20 Jun 2007 13:12:45 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: Question for the Group Message-ID: <5dsnecF36dckuU2@mid.individual.net> In article , david20@alpha2.mdx.ac.uk writes: > In article , Michael Kraemer writes: >>AEF schrieb: >> >>> Pardon my ignorance, but didn't Apple do the same? What did they do >>> wrong? >> >>Certainly I do not have to repeat PC history here, or ? >>It was IBM+MS+intel who lay the foundation for PCs >>dominance in the 1980s. No chance for a small startup >>like Apple to change that, no marketing could have done that. >>They could be happy to stay at a few % of the market, >>and even for that they will have to work harder and innovate faster than >>the rest. >> >>> >>>>>Others advertise security. Why not VMS? Why not back it up with >>>>>something? What would it hurt? >>>> >>>>Nothing, but would it help ? Everybody claims to be "secure" >>>>these days. Now if VMS would have some security certificate >>>>from NSA or whoever issues such things, putting them five notches >>>>above the usual Unix crowd, that would be something to brag about. >>> >>> >>> And those who claim security are doing better than VMS. So, if VMS >>> claimed security, it should do better, too, no? >> >>Just think of a guy of those two or three academic generations >>who have left unis without ever having heard about VMS. >>For him, these letters would expand to "Video Management System", >>or, as in google.de, to "Verkehrsverbund Mittelsachsen" which >>is a public transport service in eastern germany. >>He might read an ad about super-secure VMS, but >>on the next page there's an ad about hyper-secure AIX >>and on the next page another one touting ultra-secure Solaris. >>Even if he hasn't heard about the latter two, how should he >>be able to differentiate ? If there would be some official >>certificate rating one high above the others, this would be at >>least some differentiator. >> >>> >>>>OTOH, "security" these days means to organize your IT so >>>>that it has minimum cross section to the evil internet, >>>>rather than the choice of a particular OS. >>>>Raise a firewall, hide business critical systems and >>>>important databases etc. >>> >>> Which isn't really enough. >> >>These are by far the most important measures, >>much more important than the choice of OS. >>I think most security paranoid will tell you so. >> > No you need multiple security layers. Firewalls cannot protect you from attacks > through ports you leave open to enable the server to do it's business. > Eg If you are running a publically accessible webserver on port 80 then you > need to allow traffic through to port 80 (either directly or published via a > proxy server). If your webserver or web served applications have > vulnerabilities then the firewall provides zero protection. > Firewalls are not magic shields. > To a large extent firewalls are superfluous. You can provide pretty much the > same protection at the host level by hardening the OS and turning off > unnecessary services. However when you have a large number of systems it is > much easier to control this at a firewall rather than on every single host. > > Unfortunately some businesses seem to think that the firewall really is a > magic shield and that they are therefore protected just by having one in place. > Hence they become lax on application and OS level security and put off > patching etc While a firewall is not a magic bullet, it is a big help. For one, it cuts down on all the extraneous traffic. Even if you don't have a web server running your network is probably constantly being scanned for open port 80's. And the same for other well known ports. You may not run Windows at all but you can bet people are looking for it. And let's not forget port 25!! And then, if you have an IDS in conjunction with your firewall it can be very good at detecting the difference between real users and attacks and can instruct your firewall accordingly. 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, 20 Jun 2007 07:53:54 -0700 From: AEF Subject: Re: Question for the Group Message-ID: <1182351234.502606.105300@q69g2000hsb.googlegroups.com> On Jun 20, 5:18 am, Michael Kraemer wrote: > AEF schrieb: > > > Pardon my ignorance, but didn't Apple do the same? What did they do > > wrong? > > Certainly I do not have to repeat PC history here, or ? > It was IBM+MS+intel who lay the foundation for PCs > dominance in the 1980s. No chance for a small startup > like Apple to change that, no marketing could have done that. > They could be happy to stay at a few % of the market, > and even for that they will have to work harder and innovate faster than > the rest. Sorry, but I responding to your comment: " > Sure it was also marketing, but Windoze, as buggy as it may be, > gave the masses what they wanted, some toy they could play games, > edit their letters and (later) surf the internet. > That those boxes crashed 10 times a day didn't matter much, > apparently. It was second rank. Pardon my ignorance, but didn't Apple do the same? What did they do wrong? " It seems to me that Apple must have also "given the masses what they want", so something must be different. That's what I was alluding to. I think it was the Apple didn't go after the business market and MS/ IBM/Compatibles did, and employees wanted the same thing at home, I think. So it's not just giving users "what they want". It was marketing to businesses, no? > > > > >>>Others advertise security. Why not VMS? Why not back it up with > >>>something? What would it hurt? > > >>Nothing, but would it help ? Everybody claims to be "secure" > >>these days. Now if VMS would have some security certificate > >>from NSA or whoever issues such things, putting them five notches > >>above the usual Unix crowd, that would be something to brag about. > > > And those who claim security are doing better than VMS. So, if VMS > > claimed security, it should do better, too, no? > > Just think of a guy of those two or three academic generations > who have left unis without ever having heard about VMS. > For him, these letters would expand to "Video Management System", > or, as in google.de, to "Verkehrsverbund Mittelsachsen" which > is a public transport service in eastern germany. So call it OVMS or OpenVMS. > He might read an ad about super-secure VMS, but > on the next page there's an ad about hyper-secure AIX > and on the next page another one touting ultra-secure Solaris. > Even if he hasn't heard about the latter two, how should he > be able to differentiate ? If there would be some official > certificate rating one high above the others, this would be at > least some differentiator. Fine, but at least VMS would be out there like all the others and get SOME sales. By not being out there at all is worse than just at least being out there and saying "I'm more secure". > > > > >>OTOH, "security" these days means to organize your IT so > >>that it has minimum cross section to the evil internet, > >>rather than the choice of a particular OS. > >>Raise a firewall, hide business critical systems and > >>important databases etc. > > > Which isn't really enough. > > These are by far the most important measures, > much more important than the choice of OS. > I think most security paranoid will tell you so. You can't isolate yourself from your Web pages' users. You have to allow access to something and then you're vulnerable, no? > > Make ads claiming so and that demonstrate > > how VMS does more than that. > > Yeah, not everyone needs clustering. But I think many would benefit > > from it. > > But not if it comes at the price of an obscure OS which > has little else to offer. I have seen other HA systems not do so well when a node goes down. Though that's just my experience. AEF ------------------------------ Date: Wed, 20 Jun 2007 13:04:55 +0200 From: "P. Sture" Subject: Re: Same system disk for multiple architectures ? Message-ID: In article , david20@alpha2.mdx.ac.uk wrote: > In article <5dq9ruF33f287U1@mid.individual.net>, bill@cs.uofs.edu (Bill > Gunshannon) writes: > >In article , > > Stephen Hoffman writes: > >> Tom Linden wrote: > >> > >>> I didn't see a date on this, but judging by the refs it would appear to > >>> be > >>> ca. 1991. > >>> > >>> So what happened to this effort? > >> > >> The paper was published. > >> > >> The Mach changes that were prototyped for OpenVMS and described in the > >> paper were not deployed into production. > >> > >> Interestingly, Apple did deploy a Mach kernel port for its Mac OS X > >> product, re-hosting the existing Mac OS 9 platform onto the Mach kernel. > >> > > > >I'll bet the guys doing FreeVMS never heard of this. Might have had an > >effect on where they chose to start from. > > > I seem to remember years ago when there was first discussions about creating > a > freeVMS there was talk about building it on a Mach kernel. > > It looks like FreeVMS is not doing too badly see > > http://www.systella.fr/~bertrand/FreeVMS/IndexGB.html > That didn't work, but backing up a level and selecting the English version gave this: http://www.systella.fr/~bertrand/FreeVMS/indexGB.html -- Paul Sture ------------------------------ Date: Wed, 20 Jun 2007 12:22:26 +0000 (UTC) From: david20@alpha2.mdx.ac.uk Subject: Re: Same system disk for multiple architectures ? Message-ID: In article , "P. Sture" writes: >In article , david20@alpha2.mdx.ac.uk >wrote: > >> In article <5dq9ruF33f287U1@mid.individual.net>, bill@cs.uofs.edu (Bill >> Gunshannon) writes: >> >In article , >> > Stephen Hoffman writes: >> >> Tom Linden wrote: >> >> >> >>> I didn't see a date on this, but judging by the refs it would appear to >> >>> be >> >>> ca. 1991. >> >>> >> >>> So what happened to this effort? >> >> >> >> The paper was published. >> >> >> >> The Mach changes that were prototyped for OpenVMS and described in the >> >> paper were not deployed into production. >> >> >> >> Interestingly, Apple did deploy a Mach kernel port for its Mac OS X >> >> product, re-hosting the existing Mac OS 9 platform onto the Mach kernel. >> >> >> > >> >I'll bet the guys doing FreeVMS never heard of this. Might have had an >> >effect on where they chose to start from. >> > >> I seem to remember years ago when there was first discussions about creating >> a >> freeVMS there was talk about building it on a Mach kernel. >> >> It looks like FreeVMS is not doing too badly see >> >> http://www.systella.fr/~bertrand/FreeVMS/IndexGB.html >> > >That didn't work, but backing up a level and selecting the English >version gave this: > >http://www.systella.fr/~bertrand/FreeVMS/indexGB.html > Yes sorry for the typo. David Webb Security team leader CCSS Middlesex University >-- >Paul Sture ------------------------------ Date: 20 Jun 2007 13:37:02 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: Top ten dead or dying computer skills (guess what's NOT on the list!) Message-ID: <5dsortF36dckuU4@mid.individual.net> In article <1182307693.937246.310980@n60g2000hse.googlegroups.com>, AEF writes: > > OK, here's the link: > > http://www.globalknowledge.com/training/customgo.asp?find=P38_FEATURE&id=18488&pageid=29&country=United+States > Thank you. I needed a good laugh and this was definitely it. Number 1 was COBOL. Anyone looked at the COBOL job market lately? A lot more COBOL jobs out there than VMS. Sadly, they are mostly IBM/CICS and I haven't touched an IBM box in nearly 30 years. Of course, there are some non-IBM COBOL jobs too but they are harder to find. Actually, I just had this discussion here when the last Professor to use COBOL in a course was pressured into changing it to Java. I gave him a bunch of articles from the web citing the continued use (and even hinting at growth) of COBOL programs. Much of it is hidden between Java Web Frontends and Oracle Backends, but stuck right there in the middle COBOL is still going strong. Number 6 was C. Yeah, with Linux and Solaris and AIX and HPUX we're really gonna see C go away!! Oh well, some of them are probably right. Even I have to admit to not seeing cc:Mail or Netware anytime recently. 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, 20 Jun 2007 13:08:03 -0400 From: Bill Todd Subject: Re: Top ten dead or dying computer skills (guess what's NOT on the list!) list!) Message-ID: AEF wrote: ... > OK, here's my question: Is this good or bad as regards VMS? You mean, is it good or bad that VMS is such an unimportant dead or dying technology that it didn't make the top-ten list? I guess that would depend upon how one felt about VMS's demise, but I suspect that most people here would prefer to have had it considered sufficiently important to be mentioned. - bill ------------------------------ Date: 20 Jun 2007 13:19:43 GMT From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: UCX Printer connection Via LPD Message-ID: <5dsnrfF36dckuU3@mid.individual.net> In article <467883FA.BCE8E9C6@spam.comcast.net>, David J Dachtera writes: > Bill Gunshannon wrote: >> [snip] >> Or, in other words, it may not be VMS's fault. > > VMS's fault? Certainly not! > > Now, UCX on the other hand... Sorry, I thought people would get the point. While not in the kernel, UCX is the VMS IP Stack in use and just because it is not working doesn't mean it's UCX's fault. Having used it extensively in the past, I would be surprised if it was. I think it is more likely that while the way UCX is doing it looks wrong it is actually acceptable but the receiving end is just not accepting it. That is apparently the problem between the LPD provided in Windows Services for Unix and Linux. Windows actually stuck to the spec and got it right but Linux didn't. "Be conservative in what you send and liberal in what you accept." 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, 20 Jun 2007 09:25:04 -0500 From: Chris Subject: Re: UCX Printer connection Via LPD Message-ID: Bill Gunshannon wrote: > In article <467883FA.BCE8E9C6@spam.comcast.net>, > David J Dachtera writes: >> Bill Gunshannon wrote: >>> [snip] >>> Or, in other words, it may not be VMS's fault. >> VMS's fault? Certainly not! >> >> Now, UCX on the other hand... > > Sorry, I thought people would get the point. While not in the kernel, > UCX is the VMS IP Stack in use and just because it is not working doesn't > mean it's UCX's fault. Having used it extensively in the past, I would > be surprised if it was. I think it is more likely that while the way > UCX is doing it looks wrong it is actually acceptable but the receiving > end is just not accepting it. That is apparently the problem between > the LPD provided in Windows Services for Unix and Linux. Windows actually > stuck to the spec and got it right but Linux didn't. "Be conservative in > what you send and liberal in what you accept." > > bill > This is occurring no matter what we put at the other end. We are concerned because the protocol sequence is so far out of the spec. (For your information though the printers are Ricoh {Who I work for}). The concern is that the LPD RFC [RFC1179] Clearly states that there should be ONE control file and ONE data file. There are TWO DATA files here, and NO control file. Which is not even close. Although the syntax of the data is LPD "like" the construct of the sequence is wrong. Actually I am a member of the IEEE PWG print working group that writes these protocols for print devices. We know well that the LPD RFC has been extended by many vendors and is nearly impossible to follow to the letter in programming, but there are some basic sequences that must be followed. These basic sequences are not present. What I really need is to know if there are any special settings that can cause UCX to go out of the LPD spec. I've worked with it for some time and have never seen this behavior. Is it possible that other softwares could produce this also? Like NorthLake Printkit? thoughts? Regarding your comments about Linux, you are correct in one perspective, that there were some issues early on with some print systems. Linux has changed quite a bit and the current de-facto printing system CUPS is much better than some of the problems created by LPRng and standard LPR. The great thing about open source software is that anybody can find and fix those annoyances. Thanks for the assistance. Chris Story ------------------------------ Date: Wed, 20 Jun 2007 11:48:14 -0400 From: "Ken Robinson" Subject: Re: UCX Printer connection Via LPD Message-ID: <7dd80f60706200848y58bfcdefmfe656d2e474e7d5b@mail.gmail.com> On 6/20/07, Chris wrote: > This is occurring no matter what we put at the other end. We are > concerned because the protocol sequence is so far out of the spec. (For > your information though the printers are Ricoh {Who I work for}). The > concern is that the LPD RFC [RFC1179] Clearly states that there should > be ONE control file and ONE data file. There are TWO DATA files here, > and NO control file. Which is not even close. Although the syntax of > the data is LPD "like" the construct of the sequence is wrong. > Actually I am a member of the IEEE PWG print working group that writes > these protocols for print devices. We know well that the LPD RFC has > been extended by many vendors and is nearly impossible to follow to the > letter in programming, but there are some basic sequences that must be > followed. These basic sequences are not present. > What I really need is to know if there are any special settings that > can cause UCX to go out of the LPD spec. I've worked with it for some > time and have never seen this behavior. > Is it possible that other softwares could produce this also? Like > NorthLake Printkit? thoughts? You could try using DCPS. The latest version have the capability of using LPD for printing. Ken ------------------------------ Date: Wed, 20 Jun 2007 11:02:39 -0500 From: Chris Subject: Re: UCX Printer connection Via LPD Message-ID: Ken Robinson wrote: > On 6/20/07, Chris wrote: >> This is occurring no matter what we put at the other end. We are >> concerned because the protocol sequence is so far out of the spec. (For >> your information though the printers are Ricoh {Who I work for}). The >> concern is that the LPD RFC [RFC1179] Clearly states that there should >> be ONE control file and ONE data file. There are TWO DATA files here, >> and NO control file. Which is not even close. Although the syntax of >> the data is LPD "like" the construct of the sequence is wrong. >> Actually I am a member of the IEEE PWG print working group >> that writes >> these protocols for print devices. We know well that the LPD RFC has >> been extended by many vendors and is nearly impossible to follow to the >> letter in programming, but there are some basic sequences that must be >> followed. These basic sequences are not present. >> What I really need is to know if there are any special >> settings that >> can cause UCX to go out of the LPD spec. I've worked with it for some >> time and have never seen this behavior. >> Is it possible that other softwares could produce this also? >> Like >> NorthLake Printkit? thoughts? > > You could try using DCPS. The latest version have the capability of > using LPD for printing. > > Ken Yea, If it were DCPS there would be no problem at all. Unfortunately they wont let us add DCPS for 30 days... Change management policy... Thanks for the assistance. Chris Story ------------------------------ Date: Wed, 20 Jun 2007 12:07:57 -0400 From: Paul Anderson Subject: Re: UCX Printer connection Via LPD Message-ID: In article <7dd80f60706200848y58bfcdefmfe656d2e474e7d5b@mail.gmail.com>, "Ken Robinson" wrote: > You could try using DCPS. The latest version have the capability of > using LPD for printing. Yeah, and we even send the right number of control and data files. But I find it hard to believe that TCP/IP Services is doing the wrong thing. What version are you running? Paul -- Paul Anderson OpenVMS Engineering Hewlett-Packard Company ------------------------------ Date: Wed, 20 Jun 2007 03:25:06 -0700 From: jarl@mimer.com Subject: wchar_t confusion Message-ID: <1182335106.273103.75850@k79g2000hse.googlegroups.com> First of all I have to admit I'm quite new to this Unicode/wchar stuff. I'm having serious problems with national wchar_t values, in the ASCII range 128 - 255. For example I expect the character '=E5' to have the hexcode value E5, but I get ffffffe5. I've included a small example program below: #include #include void w_as_hex(wchar_t*); void main () { wchar_t wstr[2] =3D L"=E4"; w_as_hex(L"a"); w_as_hex(L"=E5"); w_as_hex(wstr); } void w_as_hex(wchar_t* wch) { printf("w_as_hex: %x %d %u\n",wch[0],wch[0],wch[0]); } VMS> cc/deb/noopt lc.c VMS> link/deb lc VMS> run/nodeb lc w_as_hex: 61 97 97 w_as_hex: ffffffe5 -27 4294967269 w_as_hex: ffffffe4 -28 4294967268 VMS> Am I doing something wrong here? I'm completely lost... I haven't found any compiler switches making any difference. I've tried both /nounsigned_char and /unsigned_char, but they made no difference. TIA, Jarl ------------------------------ Date: Wed, 20 Jun 2007 11:50:12 +0000 (UTC) From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: wchar_t confusion Message-ID: In article <1182335106.273103.75850@k79g2000hse.googlegroups.com>, jarl@mimer.com writes: > First of all I have to admit I'm quite new to this Unicode/wchar > stuff. > > I'm having serious problems with national wchar_t values, in the ASCII > range 128 - 255. For example I expect the character '=E5' to have the > hexcode value E5, but I get ffffffe5. That would be ASCII (2-byte) E5. ------------------------------ Date: Wed, 20 Jun 2007 13:24:34 GMT From: "John Wallace" Subject: Re: wchar_t confusion Message-ID: Dig out your Kernighan+Ritchie or equivalent and refresh your memory re "widening" of parameters - a single byte is widened as it is passed to printf, which iirc will by default print it as though it were the width of an "integer" (whatever one of those is). If you *know* it's a single byte then you can use %02x in the format string to get the desired result in hex, or maybe other magick for other formats (eg to print it as unsigned decimal you might try (wch[0]&0xff)) ie { printf("w_as_hex: %02x %d %u\n",wch[0],wch[0],wch[0]&0xff); } > I'm completely lost... So was I when I recently started porting a DOS program to Win32 and came across some similar "learning experiences" related to Unicode etc. Maybe I still am (I haven't tried the code snippet above :)). hth John ------------------------------ Date: Wed, 20 Jun 2007 11:36:51 -0400 From: Stephen Hoffman Subject: Re: wchar_t confusion Message-ID: jarl@mimer.com wrote: > VMS> cc/deb/noopt lc.c > VMS> link/deb lc > VMS> run/nodeb lc > w_as_hex: 61 97 97 > w_as_hex: ffffffe5 -27 4294967269 > w_as_hex: ffffffe4 -28 4294967268 > VMS> > > Am I doing something wrong here? I'm completely lost... Standard C variable promotion sign-extension stuff. The rules for most stuff are fairly obvious, though the rules for argument promotion with unsigned values have always tended to make my head explode. (Unsigned values can get sign-extended, and can end up as signed values.) http://h71000.www7.hp.com/commercial/c/docs/6180profile_018.html Read bullet 6e in the list for the unsigned-becomes-signed case. This is standard C and how it is supposed to work, and not anything specific to HP C. -- www.HoffmanLabs.com Services for OpenVMS ------------------------------ End of INFO-VAX 2007.334 ************************