1 INFO-VAX	Fri, 15 Dec 2006	Volume 2006 : Issue 689       Contents: Re: /LGICMD suggestion Re: /LGICMD suggestion Re: /LGICMD suggestion Re: /LGICMD suggestion Re: /LGICMD suggestion+ Re: Cluster: Deleting CSB for System <node> + Re: Cluster: Deleting CSB for System <node> $ Re: FTP limitation on directory size$ Re: FTP limitation on directory sizeE Re: Info-ZIP UnZip on VMS v. directory attributes: Time for a change? E Re: Info-ZIP UnZip on VMS v. directory attributes: Time for a change? E Re: Info-ZIP UnZip on VMS v. directory attributes: Time for a change?  Re: The Hole in Cerner's Logic RE: The Hole in Cerner's Logic Re: The Hole in Cerner's Logic Re: The Hole in Cerner's Logic  F ----------------------------------------------------------------------  # Date: Fri, 15 Dec 2006 06:41:58 GMT   From: John Santos <john@egh.com> Subject: Re: /LGICMD suggestion * Message-ID: <Wsrgh.7772$bj5.5647@trnddc07>   Stephen Hoffman wrote: >  >   Some suggestions...  > K >   If the disk is offline and is disrupting the login process, then there  @ > are second-level problems likely to arise all over the place. J > Applications and tools and images and most anything that references the A > (failed) device will itself fail.  I'd tend to look at why the  I > particular disk dropped offline, and at what can be done to reduce the  H > exposure to that -- the failed and inaccessable login is probably the K > tip of a very large iceberg of problems that can arise here, and dealing  / > with those errors is going to be interesting.  > K >   If it's the SYSTEM username, then that's the system disk and there are  . > other problems with that disk being offline. > G >   If it's another username that's involved with this (failed) login,  H > then /NOCOMMAND (or the console, for access to the SYSTEM username or > > the SYSUAF is locked up or otherwise corrupt) is the way in. > G >   If you need remote emergency access, then a terminal server (or as  F > they are often now known, as a "console management processor") or a H > management processor is the way in.  If the box has the option, as do J > various of the Integrity servers, the manage ment processor can be very  > useful for these cases.   J I really hate it when I get an error message says "ask your system manager@ for assistance" and I am the system manager and have no clue :-(  C I think the OP is trying to log in to fix the myriad of problems or D at least determine which drive has broken so he can tell the on siteC personnel what drive to have replaced and which backup tape to load C on the tape drive while he drives in to work (at 3 in the morning.)   B Not some random Joe User trying to log in to a broken system, he's? the system manager trying to get a head start on fixing things.   E Certainly he could get system access via the console (opa0:) remotely H via a null-modem to a terminal server or a console management processor,H but if he can't log in, he's still stuck (short of ctrl/P; HALT; reboot,D which he might *not* want to do until he understands exactly what isA wrong.  For example, the disk with the user accounts, development C sources, etc. might have crashed, but the server application itself B is still fine, so there may be no need to reboot, just replace the drive and restore the backup.   A I think the "search-list" logical name for the login command file ? might do the trick.  This would probably not work for a captive C account, but might work for a restricted account.  (Some companies' ? security policies say all users must be in restricted accounts, ; which prevents bypassing login.com.)  But maybe it would be @ happy if it finds *any* login.com, not just the one in the first$ device/directory in the search list.  @ Also, be sure that you don't have an "on error then exit" (whichD is the default) in your login.com (and maybe in sys$sylogin as well)> so you don't get kicked out prematurely.  Either "set noon" or> appropriate "on error" commands to deal with whatever happens.   --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------    Date: 15 Dec 2006 08:44:37 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: /LGICMD suggestion 3 Message-ID: <UIfCq2bcnvSO@eisner.encompasserve.org>   p In article <9_-dnU_As6eZAR3YnZ2dnUVZ_rHinZ2d@comcast.com>, "Richard B. Gilbert" <rgilbert88@comcast.net> writes: > K > In some twenty-two years of working with VMS, I can't recall a situation  ' > in which this would have been useful!   H    It would be trivial to make the single filepoint to a .com file which    did the search.  B    IIRC through VMS 2.5 there was no concept of sylogin, so systemE    managers pointed UAF records at a system login file which did it's <    thing then looked for and executed the user's login.com .   ------------------------------    Date: 15 Dec 2006 09:38:47 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: /LGICMD suggestion 3 Message-ID: <ksQ7CcS4l6qe@eisner.encompasserve.org>   M In article <Wsrgh.7772$bj5.5647@trnddc07>, John Santos <john@egh.com> writes:  > L > I really hate it when I get an error message says "ask your system managerB > for assistance" and I am the system manager and have no clue :-(  ?    Bruce Ellis did a national DECUS session on just that topic.    ------------------------------  % Date: Fri, 15 Dec 2006 16:42:16 +0100 / From: Paul Sture <paul.sture.nospam@hispeed.ch>  Subject: Re: /LGICMD suggestion J Message-ID: <paul.sture.nospam-355A3C.16421615122006@mac.sture.homeip.net>  3 In article <UIfCq2bcnvSO@eisner.encompasserve.org>, =  koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote:   I > In article <9_-dnU_As6eZAR3YnZ2dnUVZ_rHinZ2d@comcast.com>, "Richard B.  + > Gilbert" <rgilbert88@comcast.net> writes:  > > M > > In some twenty-two years of working with VMS, I can't recall a situation  ) > > in which this would have been useful!  > J >    It would be trivial to make the single filepoint to a .com file which >    did the search. > D >    IIRC through VMS 2.5 there was no concept of sylogin, so systemG >    managers pointed UAF records at a system login file which did it's > >    thing then looked for and executed the user's login.com .  G Due to that heritage, we already had group login files which did their  6 stuff then looked for the user's login.com to execute.  E The end result once sylogin arrived was system, group and user login   files.   --  
 Paul Sture   ------------------------------  % Date: Fri, 15 Dec 2006 18:41:01 +0100 / From: Paul Sture <paul.sture.nospam@hispeed.ch>  Subject: Re: /LGICMD suggestion J Message-ID: <paul.sture.nospam-E2D633.18410115122006@mac.sture.homeip.net>  3 In article <ksQ7CcS4l6qe@eisner.encompasserve.org>, =  koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote:   O > In article <Wsrgh.7772$bj5.5647@trnddc07>, John Santos <john@egh.com> writes:  > > N > > I really hate it when I get an error message says "ask your system managerD > > for assistance" and I am the system manager and have no clue :-( > A >    Bruce Ellis did a national DECUS session on just that topic.    Any tips that you recall?    --  
 Paul Sture   ------------------------------    Date: 15 Dec 2006 01:36:20 -0800! From: "Ian Miller" <gxys@uk2.net> 4 Subject: Re: Cluster: Deleting CSB for System <node>B Message-ID: <1166175380.594570.31500@l12g2000cwl.googlegroups.com>  @ a CSB is a data structure in non paged pool in which a node in aG cluster records data about other nodes in the cluster. A CSB is deleted G (and that message issued) when it is not needed. I guess when the other  node goes away.   , Why was it output multiple times? I know not   ------------------------------    Date: 15 Dec 2006 08:35:56 -0800 From: mckinneyj@saic.com4 Subject: Re: Cluster: Deleting CSB for System <node>C Message-ID: <1166200556.619821.140700@j72g2000cwa.googlegroups.com>    JF Mezei wrote:  > Jim wrote:, > > > %CNXMGR  Deleting CSB for system WHEEL >  > > http://tinyurl.com/yacfuk  > >  > > (gotta love tiny url ;-) >  > > > Thanks. I had looked up google/deja for references to CSB on > comp.os.vms, but to no avail.  > J > However, that documentation doesn't explain what the "CSB" is and why it > needs to be deleted !   E CSB=cluster-system-block; when a system is removed from a cluster the ' associated data structure goes with it.    ------------------------------    Date: 15 Dec 2006 09:43:44 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) - Subject: Re: FTP limitation on directory size 3 Message-ID: <Lx8AO18R9oDx@eisner.encompasserve.org>   a In article <45814080_1@glkas0286.greenlnk.net>, "David Biddulph" <groups@biddulph.org.uk> writes:  > N > I seem to remember someone talking in the past about some limitation on the N > directory size for FTP to be able to see all the files.  Can anyone tell me M > what the limitation is, whether it is at the Windows client end or the VMS  L > end, and whether there is a solution?  [As a short term fix in the past I I > have copied files into a smaller directory, and the FTP could see them  M > there, but we'd like to know whether there is an option that can be fixed.]   I    This totally depends on the FTP client and server in use.  There is no B    provision in the RFC for dealing with fewer than all the files.  @    I had to work with a server on a Sun which simply gave up andF    returned a successfull completion code ater listing no files.  ThisF    was a public domain server which supported SSH when the server from5    Sun didn't, I don't recall the name of the server.    ------------------------------    Date: 15 Dec 2006 09:47:37 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) - Subject: Re: FTP limitation on directory size 3 Message-ID: <Owzr8MjQNtPI@eisner.encompasserve.org>   T In article <Lr96lEd4c3mV@eisner.encompasserve.org>, briggs@encompasserve.org writes: > > > I would not expect it to be a limitation on the server side.  F    That's going to depend on who coded and tested the server.  The oneE    I just discussed in a previous post behaved as if it malloc()'ed a D    buffer to hold all the file names and just gave up if it couldn't    malloc() one large enough.   B    The FTP client I worked on a few years ago simply used a buffer@    large enough to read one file name from the server.  I had noC    idea how many files the server was going to have and didn't want /    to pick an arbitrary limit.  KISS principle.    ------------------------------  + Date: Thu, 14 Dec 2006 23:49:33 -0600 (CST) * From: sms@antinode.org (Steven M. Schweda)N Subject: Re: Info-ZIP UnZip on VMS v. directory attributes: Time for a change?2 Message-ID: <06121423493370_2020028F@antinode.org>  $ From: "AEF" <spamsink2001@yahoo.com>  H > Well, the only potential problem I can think of offhand is that if theE > directories were originally set to RE for all four categories, then D > that could interfere with restoring files to said directories as WC > access is needed to add files to a directory. I've seen this very B > problem happen with BACKUP back in the 1980s (during a non-imageH > restore). Certainly if the protections are somehow set after all filesF > are in place I don't see a problem. Restoring with BYPASS priv would > also obviate the problem.   H    As I said, "the directory attributes are handled by a post-processing step".  A    I don't think that I'd want to use an UnZip which had BYPASS.    D    In the tentative code, if you say "-X" but you lack the privilegeH needed to set the ownership for a file or a directory, you (usually) getH a %SYSTEM-F-NOPRIV complaint, as might be expected.  Generally, I'd say,* "-X" is useful only for a privileged user.    3 From: David J Dachtera <djesys.no@spam.comcast.net>   M > My opinion is that if the directory is actually stored in the archive, then O > restore using the saved attributes. If the directory is created because it is M > needed and does not exist when restoring from an archive, then it should be 9 > created on the fly and the usual defaults should apply.   G    The directory itself is not stored, but some of its attributes are.  @ Currently, all directories are created as needed, using a simple? mkdir(), so you get the usual default permissions.  This sounds C harmless, but it's different from all other files, which have their ? permissions set to match their original permissions.  It's also G different from the UnZip behavior on all other operating systems, where D directory permissions are adjusted match their original permissions.  C    Strictly speaking, all the normal files are also created because 8 they're needed, and they get their permissions restored.    H    The tentative code seems to be fairly happy (so far), but there seemsG to be one quirk:  If a non-privileged user does "unzip -X" and restores H a directory which he did not own, the attempt to set the ownership failsD with %SYSTEM-F-NOPRIV, which is ok with me.  However, on a directoryH which he _did_ own, the attempt to set the ownership (to what it alreadyF is, I believe) fails with %SYSTEM-F-BADPARAM.  I consider this to be aE harmless NOP, and don't see why it should fail this way.  Anyone know  why?  A    Oh, and for the moment, "-X" requests restoration of directory E ownership, and "-XX" _adds_ restoration of directory date-times.  I'm D guessing that no one will actually want that last feature, as no oneG seems to have missed it enough to complain about it since the beginning ? of time.  (Of course, only one person seems to have noticed the 7 directory ownership problem in the same time interval.)   H ------------------------------------------------------------------------  3    Steven M. Schweda               sms@antinode-org 4    382 South Warwick Street        (+1) 651-699-9818    Saint Paul  MN  55105-2547    ------------------------------    Date: 15 Dec 2006 07:32:13 -0800$ From: "AEF" <spamsink2001@yahoo.com>N Subject: Re: Info-ZIP UnZip on VMS v. directory attributes: Time for a change?C Message-ID: <1166196733.042354.102160@l12g2000cwl.googlegroups.com>    Steven M. Schweda wrote:& > From: "AEF" <spamsink2001@yahoo.com> > J > > Well, the only potential problem I can think of offhand is that if theG > > directories were originally set to RE for all four categories, then F > > that could interfere with restoring files to said directories as WE > > access is needed to add files to a directory. I've seen this very D > > problem happen with BACKUP back in the 1980s (during a non-imageJ > > restore). Certainly if the protections are somehow set after all filesH > > are in place I don't see a problem. Restoring with BYPASS priv would > > also obviate the problem.  > J >    As I said, "the directory attributes are handled by a post-processing > step".  G Immediately after which you said "None of this is used on VMS, ..." and B then you referred to the problem of read-only directories (which I? admit I missed). So it wasn't clear to me what you were saying.    > B >    I don't think that I'd want to use an UnZip which had BYPASS.  B I didn't mean to have the UnZip program itself turn on BYPASS! The@ system manager (or prived user) would do it as needed, like when> running ANAL/DISK (which often doesn't work well without it!).   [...] J > ------------------------------------------------------------------------ > 5 >    Steven M. Schweda               sms@antinode-org 6 >    382 South Warwick Street        (+1) 651-699-9818 >    Saint Paul  MN  55105-2547    AEF    ------------------------------    Date: 15 Dec 2006 09:11:43 -0800$ From: "AEF" <spamsink2001@yahoo.com>N Subject: Re: Info-ZIP UnZip on VMS v. directory attributes: Time for a change?B Message-ID: <1166196951.297495.145440@79g2000cws.googlegroups.com>   Steven M. Schweda wrote:& > From: "AEF" <spamsink2001@yahoo.com> > J > > Well, the only potential problem I can think of offhand is that if theG > > directories were originally set to RE for all four categories, then F > > that could interfere with restoring files to said directories as WE > > access is needed to add files to a directory. I've seen this very D > > problem happen with BACKUP back in the 1980s (during a non-imageJ > > restore). Certainly if the protections are somehow set after all filesH > > are in place I don't see a problem. Restoring with BYPASS priv would > > also obviate the problem.  > J >    As I said, "the directory attributes are handled by a post-processing > step".  G Immediately after which you said "None of this is used on VMS, ..." and B then you referred to the problem of read-only directories (which I? admit I missed). So it wasn't clear to me what you were saying.    > B >    I don't think that I'd want to use an UnZip which had BYPASS.  B I didn't mean to have the UnZip program itself turn on BYPASS! The@ system manager (or prived user) would do it as needed, like when> running ANAL/DISK (which often doesn't work well without it!).   [...] J > ------------------------------------------------------------------------ > 5 >    Steven M. Schweda               sms@antinode-org 6 >    382 South Warwick Street        (+1) 651-699-9818 >    Saint Paul  MN  55105-2547    AEF    ------------------------------  # Date: Fri, 15 Dec 2006 10:41:42 GMT ' From: ChrisQuayle <nospam@devnul.co.uk> ' Subject: Re: The Hole in Cerner's Logic 6 Message-ID: <GZugh.7752$493.5725@newsfe4-gui.ntli.net>   Main, Kerry wrote: >>-----Original Message-----1 >>From: ChrisQuayle [mailto:nospam@devnul.co.uk]  ! >>Sent: December 14, 2006 4:11 PM  >>To: Info-VAX@Mvb.Saic.Com ) >>Subject: Re: The Hole in Cerner's Logic  >>% >>dave.baxter@bannerhealth.com wrote:  >> >>6 >>>In Healthcare, particularly with the advent of the  >> >>Electronic Medical >>; >>>Record, there are many other equally important criteria.  >>> 0 >>>See if you can figure out what they might be! >>>  >>>Dave  >>>  >>H >>Here in the uk, we have just had the rollout of a national it project F >>for the national health service. It's a nationwide, several billion = >>pound project that is late, but is slowly coming together,   >>according to  
 >>reports. >>3 >>Anyone know whose hardware and os it runs on ?...  >> >  > @ > Well, not sure if I would qualify this as "coming together .." > @ > http://www.theregister.co.uk/2006/09/29/accenture_nhs_penalty/ > < > http://www.theregister.co.uk/2006/08/01/nhs_data_recovery/ > < > http://www.theregister.co.uk/2006/09/08/mps_condemn_npfit/ > = > http://www.theregister.co.uk/2006/08/02/private_before_nhs/  > E > http://www.theregister.co.uk/2004/01/06/csc_accenture_win_nhs_care/  >   A None of which answers the question, conveniently sidstepping the  H question - who's hardware does it run on ?. An ideal candidate for vms, 3 but of course HP didn't "invent" vms, did they ?...    Chris    ------------------------------  % Date: Fri, 15 Dec 2006 07:48:55 -0500 ' From: "Main, Kerry" <Kerry.Main@hp.com> ' Subject: RE: The Hole in Cerner's Logic T Message-ID: <FA60F2C4B72A584DBFC6091F6A2B868401ED1109@tayexc19.americas.cpqcorp.net>   > -----Original Message-----3 > From: ChrisQuayle [mailto:nospam@devnul.co.uk]=20 ! > Sent: December 15, 2006 5:42 AM  > To: Info-VAX@Mvb.Saic.Com ) > Subject: Re: The Hole in Cerner's Logic  >=20 > Main, Kerry wrote: > >>-----Original Message-----5 > >>From: ChrisQuayle [mailto:nospam@devnul.co.uk]=20 # > >>Sent: December 14, 2006 4:11 PM  > >>To: Info-VAX@Mvb.Saic.Com + > >>Subject: Re: The Hole in Cerner's Logic  > >>' > >>dave.baxter@bannerhealth.com wrote:  > >> > >>: > >>>In Healthcare, particularly with the advent of the=20 > >> > >>Electronic Medical > >>= > >>>Record, there are many other equally important criteria.  > >>> 2 > >>>See if you can figure out what they might be! > >>> 	 > >>>Dave  > >>>  > >>A > >>Here in the uk, we have just had the rollout of a national=20  > it project=20 J > >>for the national health service. It's a nationwide, several billion=20A > >>pound project that is late, but is slowly coming together,=20  > >>according to=20  > >>reports. > >>5 > >>Anyone know whose hardware and os it runs on ?...  > >> > >=20 > >=20B > > Well, not sure if I would qualify this as "coming together .." > >=20B > > http://www.theregister.co.uk/2006/09/29/accenture_nhs_penalty/ > >=20> > > http://www.theregister.co.uk/2006/08/01/nhs_data_recovery/ > >=20> > > http://www.theregister.co.uk/2006/09/08/mps_condemn_npfit/ > >=20? > > http://www.theregister.co.uk/2006/08/02/private_before_nhs/  > >=20G > > http://www.theregister.co.uk/2004/01/06/csc_accenture_win_nhs_care/  > >=20 >=20E > None of which answers the question, conveniently sidstepping the=20 9 > question - who's hardware does it run on ?. An ideal=20  > candidate for vms,=20 5 > but of course HP didn't "invent" vms, did they ?...  >=20 > Chris  >=20  G Its not OpenVMS. There are a number of OS platforms (see above last url G for one of them) but when you do bids on large RFP's like this the Govt E in question will often break-it up into smaller chunks - and this one H was no different.  I think thjere was something like 5 Regions to bid on5 when it was first released about 2 or 3 years ago.=20   F The big differentiator for the initial phase for evaluators is not theF OS platform as that is way down on their list, but rather the businessG plan, the applications proposed, the organizaqtion, program management,  security processes etc.   F It is only later when down time and security issues starts hitting theD press when the original OS platform becomes an issue to be concerned about.   Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------  # Date: Fri, 15 Dec 2006 15:28:15 GMT ' From: ChrisQuayle <nospam@devnul.co.uk> ' Subject: Re: The Hole in Cerner's Logic 6 Message-ID: <jazgh.3732$1W1.1817@newsfe4-win.ntli.net>   Main, Kerry wrote:  I > for one of them) but when you do bids on large RFP's like this the Govt G > in question will often break-it up into smaller chunks - and this one J > was no different.  I think thjere was something like 5 Regions to bid on5 > when it was first released about 2 or 3 years ago.   > H > The big differentiator for the initial phase for evaluators is not theH > OS platform as that is way down on their list, but rather the businessI > plan, the applications proposed, the organizaqtion, program management,  > security processes etc.  > H > It is only later when down time and security issues starts hitting theF > press when the original OS platform becomes an issue to be concerned > about. >   F Thanks - it still doesn't answer the question, but did bit of digging F and it seems ibm hardware will support at least some of it. Some news I feature recently showed hp monitors, so I guess hp hardware is involved,  C at least on the desktop, but why are hp not pursuing this business  C agressively with vms for the back end ?. With it's scalability and  G resilience, I would have thought vms would be a natural for such work.  B Perhaps the applications aren't available and let's face it, with G hardware and os being commodity, it's the apps that drive the business   in the end.   L VMS always was a mainframe class os, so why aren't hp making more of it ?...   Chris    ------------------------------    Date: 15 Dec 2006 07:34:39 -0800- From: "Andrew" <andrew_harrison@symantec.com> ' Subject: Re: The Hole in Cerner's Logic B Message-ID: <1166196879.869106.320020@73g2000cwn.googlegroups.com>   ChrisQuayle wrote: > Main, Kerry wrote: > >>-----Original Message-----2 > >>From: ChrisQuayle [mailto:nospam@devnul.co.uk]# > >>Sent: December 14, 2006 4:11 PM  > >>To: Info-VAX@Mvb.Saic.Com + > >>Subject: Re: The Hole in Cerner's Logic  > >>' > >>dave.baxter@bannerhealth.com wrote:  > >> > >>7 > >>>In Healthcare, particularly with the advent of the  > >> > >>Electronic Medical > >>= > >>>Record, there are many other equally important criteria.  > >>> 2 > >>>See if you can figure out what they might be! > >>> 	 > >>>Dave  > >>>  > >>I > >>Here in the uk, we have just had the rollout of a national it project G > >>for the national health service. It's a nationwide, several billion > > >>pound project that is late, but is slowly coming together, > >>according to > >>reports. > >>5 > >>Anyone know whose hardware and os it runs on ?...  > >> > >  > > B > > Well, not sure if I would qualify this as "coming together .." > > B > > http://www.theregister.co.uk/2006/09/29/accenture_nhs_penalty/ > > > > > http://www.theregister.co.uk/2006/08/01/nhs_data_recovery/ > > > > > http://www.theregister.co.uk/2006/09/08/mps_condemn_npfit/ > > ? > > http://www.theregister.co.uk/2006/08/02/private_before_nhs/  > > G > > http://www.theregister.co.uk/2004/01/06/csc_accenture_win_nhs_care/  > >  > B > None of which answers the question, conveniently sidstepping theI > question - who's hardware does it run on ?. An ideal candidate for vms, 5 > but of course HP didn't "invent" vms, did they ?...  > G The bulk of the systems refered to in the Register articles are HP/Dell A and IBM servers running Windows. This is mainly because iSoft the 5 software chosen for the LSP projects is Windows only.    Regards  Andrew Harrison  > Chris    ------------------------------   End of INFO-VAX 2006.689 ************************