INFO-VAX Tue, 06 Nov 2007 Volume 2007 : Issue 607 Contents: Re: @SYS$MANAGER:UTC$TIME_SETUP SHOW : no output using sysman Re: AUDITing question (file creation failure) Re: AUDITing question (file creation failure) Re: AUDITing question (file creation failure) Re: AUDITing question (file creation failure) Re: Happy Anniversary VMS - 30 years young Re: Happy Anniversary VMS - 30 years young Re: Happy Anniversary VMS - 30 years young ---------------------------------------------------------------------- Date: 5 Nov 2007 23:21:16 +0100 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOeGER) Subject: Re: @SYS$MANAGER:UTC$TIME_SETUP SHOW : no output using sysman Message-ID: <472fa56c@news.langstoeger.at> In article , Dale Dellutri writes: >If I do the command locally, it works on each of my >non-clustered systems: >[UTC$TIME_SETUP.COM] >But if I do it using sysman to the systems, >there's no output: > >Why doesn't it produce output? Because the UTC$TIME_SETUP.COM does a DEASSIGN SYS$OUTPUT and kills the logical the SMISERVER sets to get the output of the called procedure. Not the only bug of the SMISERVER... -- Peter "EPLAN" LANGSTOEGER Network and OpenVMS system specialist E-mail peter@langstoeger.at A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist ------------------------------ Date: Mon, 05 Nov 2007 15:59:28 -0500 From: JF Mezei Subject: Re: AUDITing question (file creation failure) Message-ID: Syltrem wrote: > %RMS-E-PRV, insufficient privilege or file protection violation Some fancier applications test access prior to actually doing the file access. The error code you are seeing might just have been hardcoded into the program after it has tested the file access capability and decided it couldn't access it. One possible way, of the AUDIT doesn't work is to so a SET WATCH FILE/CLASS=ALL (with mighty privs enables, then disable the mighty privs and run the aplication). Lots of output. But the temtative file access failure should happen not far from where the RMS-E error message is issued. BTW, on my system, I use an older SET AUDIT: $ SET AUDIT /ALARM /ENABLE=(AUTHORIZATION,BREAKIN=ALL,FILE_ACCESS=FAILURE,ACL) $ SET AUDIT /ALARM /ENABLE=(LOGFAILURE=ALL,LOGIN=DIALUP,LOGOUT=DIALUP) $ SET AUDIT /ALARM /ENABLE=(MOUNT,AUDIT) This also generates opcom messages when Mozilla tries to delete a file it is still using. (for instance, in a RELOAD of a page). ------------------------------ Date: Mon, 05 Nov 2007 16:08:00 -0500 From: JF Mezei Subject: Re: AUDITing question (file creation failure) Message-ID: <1d360$472f8630$cef8887a$11349@TEKSAVVY.COM> Another thing you can do is to set an ACL alarm on any directory file you suspect the application needs access to. I don't recall the exact syntax, but you can add an alarm ACL on any file at which point any access to the file generates an opcom message. ------------------------------ Date: 5 Nov 2007 16:16:26 -0600 From: Kilgallen@SpamCop.net (Larry Kilgallen) Subject: Re: AUDITing question (file creation failure) Message-ID: In article <13iujs049d8gq60@corp.supernews.com>, "Syltrem" writes: > Currently the application only returns this: > > %RMS-E-PRV, insufficient privilege or file protection violation > So.... How can I audit a file creation failure ? Enable auditing (or alarming) for failed use of privilege. That should be an infrequent circumstance on a well-behaved system. ------------------------------ Date: Mon, 05 Nov 2007 18:58:26 -0500 From: "Richard B. Gilbert" Subject: Re: AUDITing question (file creation failure) Message-ID: <472FAE22.4060507@comcast.net> Syltrem wrote: > Hi > > I'm trying to find out what file some program is trying to create, and > where. > My guess is that is is trying to *create* a file, anyway. > > Currently the application only returns this: > > %RMS-E-PRV, insufficient privilege or file protection violation > > I tried setting: > $ REPLY/ENABLE > > $ SET AUDIT/ALARM/AUDIT/ENABLE=ACCESS=FAILURE /CLASS=FILE > > But that returns nothing I believe because the program is not attempting to > access a file but it tries to *create* one > > Well I could have add an error on the .DIR, but I get nothing. > > I also tried with this: > $ SET AUDIT/ALARM/AUDIT/ENABLE=CREATE /CLASS=FILE > > and although it does return something when a file is successfully created, > it doesn't give anything for a file that could not be created due to > protection violation. > > > So.... How can I audit a file creation failure ? > > Thanks ! > Have you considered getting a better program? A well written program would tell you what file it was trying to create! If the program has documentation, you might try reading it to see what files it expects to create, what logical names it expects to find, etc, etc. Based on the evidence presented, there's no guarantee that it's trying to create a file! It could be trying to write to an existing file or even trying to READ a file. If you don't mind risking a system crash, you might try the undocumented and unsupported SET WATCH command. From memory the command is SET WATCH /CLASSS=MAJOR FILE or something like that. ------------------------------ Date: Mon, 05 Nov 2007 11:07:05 -0800 From: "Tom Linden" Subject: Re: Happy Anniversary VMS - 30 years young Message-ID: On Mon, 05 Nov 2007 09:29:35 -0800, Bob Koehler wrote: > IIRC Intel puts IA-32 "compatability mode" into there current IA-64 > processors. No knowing when they might get out of that habbit. The line between emulated and native can be a bit fuzzy, these days. -- PL/I for OpenVMS www.kednos.com ------------------------------ Date: Mon, 5 Nov 2007 19:50:13 -0000 From: "John Wallace" Subject: Re: Happy Anniversary VMS - 30 years young Message-ID: <13iusvtbg6nln67@corp.supernews.com> "Main, Kerry" wrote in message news:C72D63EB292C9E49AED23F705C61957BDEBA1F095F@G1W0487.americas.hpqcorp.net... Another essay follows, the executive summary is in the last line, you may have seen it before somewhere. >Ummm, since when did Office become a server application? >How many people install Office on Xeon based x86 servers? >Did I miss something? Does "thin client" mean anything to you. Of course it does. How does that work with an IA64 as a server and traditional corporate desktop apps? It's a piece of cake with AMD64 in the server, together with VMware, etc. Seen the VMware share price in the last few months? Looks safe for now, regardless of whether "thin client" really is the next big thing. Anyway, thin client with IA64 servers isn't happening in any IT department's future my crystal ball knows about. The ability to run Win32 apps on Windows 64 is described by MS as a "personal productivity feature"; in other words, "use at your own risk". On AMD64, that's relatively low risk. On Win64 on IA64? That's purely for the brave. How many round here remember needing to be brave with FX!32 on NT/Alpha? >> And frankly, do you serously believe that someone with pure business >> goals would choose to run windows on some IA64 box ? >As Keith stated, the answer is yes. With the greatest of respect Kerry, your belief, Keith's belief, even Sue's belief and JF's belief, is irrelevant to the bigger IA64 picture, as I'm sure you and many others must realise. The people whose beliefs matter in the bigger IA64 picture generally don't take part in this discussion, not here anyway; they are the big-cheque-signers and spreadsheet-jockeys in HQ at HP and Intel, and MS and Oracle, and SAP and... And to a lesser extent, the beliefs of similar beancounters at end users, consultants, outsourcers, etc are important too. In any given strategic sales opportunity, *all* of those people have to be on the side of IA64 for IA64 to succeed. In the particular case of IA64 VMS, they must speak up in favour of VMS too. If any one of these folks in a sales cycle can demonstrate plausibly that IA64/VMS is not a commercially sensible strategic option, the sensible customer must look elsewhere, just as they frequently did when Alpha/NT and Alpha/OSF were the new kids on the block. You remember those days, right? They weren't easy even when DEC had a pedigree, and Alpha had a performance advantage, and OSF/Tru64 allegedly had a technology/standards advantage (at least vs Slowaris), and VMS was, well, VMS, same as it ever was. What benefits does IA64 bring to this discussion? For anyone with open eyes, the biggest thing IA64 brings is uncertainty, which is hardly a benefit. Today's potential IA64 customers who see through the IA64 Emperor's clothes and are lucky enough to have a choice may perhaps look to products with a demonstrable track record and a visible vendor commitment. Power and SPARC arguably still have those, x86-64 demonstrably has and will continue to have, so there's no shortage of choices for most buyers. Once that factor comes into the discussion, IA64 has lost the sale *unless* the customer is one of the tiny proportion who have no other option (eg they're a price-no-object NonStop diehard, in which case this discussion is largely irrelevant anyway, or they're an HPTC buyer, maybe with Intel marketing sweeteners to help the decision). So, in order to succeed, IA64 needs to bring something very very special to the table. Maybe VMS or HPUX are that something today, certainly VMS is still very special. Maybe IA64 servers are that something special today, I don't know. Or maybe they're not - I'd be quite surprised if there was much difference in technology between a low-end IA64 box and a comparably priced/configured Proliant (I've seen both and from what I've seen they're both decent well-engineered products, and on a good day hopefully they're both well supported). A blade server from HP probably doesn't change all that much whether it's IA64 or x86-64, so IA64's technical superiority will eventually be reflected in the relative sales figures (including the ones for VMS blades), all other things being equal, right ? At the high end, Proliant still has some way to go, but Proliant-class kit keeps a lot of the IT world surprisingly happy, even if running Windows on them does rather spoil the picture. Wouldn't it be nice if there was another Proliant OS choice too ? It's hard for outsiders to see any logical basis for how IA64 is *helpful* to the VMS market, but lots of sensible folk find it easy to see how it's generally a hindrance; it's a great shame that's not a comfortable message for the few HP folk that are willing to raise their heads over the parapet round here (ie, to those of you that do, THANK YOU for your continued support). Are there really enough potential customers in the middle that simply don't care either way wrt IA64 vs AMD64, folks that just want VMS regardless of any perceived challenges associated with the underlying platform choice? "I'm not anti-IA64, I'm anti-stupid" ------------------------------ Date: Mon, 05 Nov 2007 17:08:42 -0500 From: JF Mezei Subject: Re: Happy Anniversary VMS - 30 years young Message-ID: FredK wrote: > To be honest I seldom reply or pay attention to COV anymore because of a > couple posters (JF being high among them) who have made it their mission to > turn every positive into a negative, Well thank you. I apologize for not being able to spin positively the fact that HP hasn't even bothered to issue an HP press release to celebrate VMS' 30th anniversary. The people in COV know the efforts made by the VMS people, but in the end, if you try to pitch a VMS solution to someone, that someone will want to see something from HP, not some VMS group employee whose promised can be overturned anyday by HP corporate, as was shown on June 25 2001. There is a big difference between promising continued support for your current applications and promising continued development at market pace to bring whatever new technology pops up to VMS within reasonable time. In the end, to be competitive, you need to be able to implement new technologies rapidly. ------------------------------ End of INFO-VAX 2007.607 ************************