Re: **PLEASE HELP ** 4584E : in DSMSERV RESTORE DB:
This rings a bell. When I moved to a larger machine, on V3R7, it also failed, tho I can't remember the earror message. Support said there was a known problem here, and, like Jelle, I had to redefine my library as manual. I saved the definitions, but there on my desktop at the office. Support stepped me thru it in less then 15 minutes. On Sun, 20 Jan 2002, Jelle Komrij wrote: > I don't remember details but when i tried to restore TSM DB tape, > during the restore the library dismounted the tape. I resolved that by > define a manual tape drive in manual library, not an automated library. > Worked fr me. > > > - Origineel Bericht - > Van: Michel David <[EMAIL PROTECTED]> > Datum: Zondag 20 Januari 2002 16:18 > Onderwerp: Re: **PLEASE HELP ** 4584E : in DSMSERV RESTORE DB: > > > Thank you very much Paul > > > > 1) I format the db and log files. > > 2) Our DB-BACKUP is on ONE TAPE only. Small DB > > ~10Mb(because all the files are more than 1GB) > > 3) We make the commit=yes > > 4) We don't have the VOLUME HISTORY butg we found the > > tapes with DSMSERV DISPLAY function. > > 5) We work according to "the Administrator's Guide" > > 6) We tried other tapes from other library. It works > > fine. > > 7) All the DB-BACKUP tapes seems to be corrupted > > (how?) > > > > ---> Do you know how to recover without the DBBACKUP > > with only the DATA TAPES ? > > > > > > > It is difficult to help you. Many of us run V4 of > > > TSM and do not have the > > > messages manual for V3. The complete message text > > > for the 8337 would be > > > helpful. It sounds like it is trying to tell you > > > how the recovery log will > > > be handled. Are you sure you have the DEVCONFIG > > > setup correctly for a > > > restore. Did you format the files to receive the > > > Database and recovery > > > logs? Did you follow the Administrator's Guide for > > > recovery based on your > > > specific scenario? > > > > > > I am presuming that you do not have the VOLHISTORY > > > file to work with. Are > > > you specifying COMMIT=YES? See the example command > > > here. > > > > > > dsmserv restore db devclass=8MM > > > volumenames=tape01,tape02 commit=yes > > > > > > Are sure the DBBACKUP is on a single tape, that it > > > is not a multi-volume? > > > Are you positive your DB Backups have been running > > > correctly? > > > > > > I suspect folks are having trouble helping you > > > without this information. > > > > > > -Original Message- > > > From: Michel David [mailto:[EMAIL PROTECTED]] > > > Sent: Sunday, January 20, 2002 3:53 AM > > > To: [EMAIL PROTECTED] > > > Subject: **PLEASE HELP ** 4584E : in DSMSERV RESTORE > > > DB: > > > > > > > > > Hi everybody. > > > > > > Thank you, Richard who tried to help us. > > > > > > Here is the problem: > > > TSM3.7.4 > > > NT-SERVER SP5 > > > FULL DB-BACKUP on tapes 8MM AIT2. > > > We DON'T USE INCREMENTAL DB BACKUP. > > > > > > We have some tapes with DBBACKUP on them. > > > We try DSMSERV RESTORE DB on them > > > This falls on every TAPE ! > > > > > > The Library starts to work. > > > 1)It choose the tape. > > > 2)BEFORE the tape is INSIDE we get 8337:Recovery log > > > > > > 3) The Tape is inside. > > > 4) 4638I REstore Backup Series 129 > > > 5) 4583E Imcomplete volume list error. > > > 6) The recovery stops. > > > 7) The tape is ejected > > > > > > I think that the 2) is not normally. > > > ? Has someone an idea. ? > > > > > > Thank you very much! > > > Michel DAVID > > > MARNET > > > > > > > > > > > > __ > > > Do You Yahoo!? > > > Send FREE video emails in Yahoo! Mail! > > > http://promo.yahoo.com/videomail/ > > > > > > __ > > Do You Yahoo!? > > Send FREE video emails in Yahoo! Mail! > > http://promo.yahoo.com/videomail/ > > >
OPEN BSD?
I have a user who ants to box up a box running OPEN BSD. The closest I could find in the archives was a solution for FREE BSD. The user checked this out but found it used a cleint no longer available. Has anyone out there found a way to do this??
Re: Strange behaviour????
That's what I wound up doing. On Fri, 26 Oct 2001, Jerry Caupain wrote: > But why EVERY directory of EVERY system, even the ones that have nothing to do with >this managementclass? > > Why don't all other systems just use the default managementclass? > Do I really need to specify a dirmc for every system in order to get the policy I >need? > > Regards, > > Jerry Caupain > > > > > >>> [EMAIL PROTECTED] 10/26/01 12:05PM >>> > hi, > > it will be used for directories if the retonly value is the highest in this > DOMAIN, > > > > > -Original Message- > > From: Jerry Caupain [SMTP:[EMAIL PROTECTED]] > > Sent: Friday, October 26, 2001 11:12 AM > > To: [EMAIL PROTECTED] > > Subject: Strange behaviour > > > > Hello everyone, > > > > I have noticed something strange. In my policy domain I have two > > management classes. One is called STANDARD and the other is called > > LOG_POLICY-MC. I want to use the last one only for my log server so I > > included the following line in my include/exclude file: > > INclude /logs/.../* log_policy-mc > > > > Managementclass STANDARD is the default managementclass. > > > > Why is it that all my other systems also use the LOG_POLICY-MC > > managementclass? It seems that the directories on my other systems are > > all bound to this managementclass. This can't be normalcan > > it??? > > > > Jerry Caupain >
Re: SIMPLE 3590/3494 sharing with 2 AIX/TSM(3466) boxes.
Joe, That depends on the level of TSM you're on. I've got six drives shared with three servers on three different levels. The two on 3.7.4 still drop a drive on occasion, but I haven't seen that on the test machine on 4.1.2. I can't tell you exactly, since this is Sunday am at home, but I think there's a 3590shared option also. On Fri, 19 Oct 2001, Joe Faracchio wrote: > I'm looking for a second opinion, caveats and any "we've been there > (and its good/bad)" about tape sharing . Sharing that's NOT SAN stuff. > > In speaking with a level 1 (1.5-2!) TSM/3466 support person about > something else, we got around to the topic of 3590's cabled to two systesm > and he said I could have all four tape drives (3590-E1A's) online to both > 3466 NSMs and they will just go into POLLING mode until a drive becomes > available. AIX/TSM can play NICE-NICE with shared drives!!(??!) > > And all I need to do is put 3494SHared Yes in both server options files. > > Does this work? Work well? Any hang-ups? Crashes? Gothchas?? > > As long as I scheduled things so that there's little or no drive > contention does it work? > > thanks ... joe.f. > > Joseph A Faracchio, Systems Programmer, UC Berkeley > Private mail on any topic should be directed to : > [EMAIL PROTECTED] COUNTDOWN: 148 days to my *first* retirement!!! > (510)642-7638 (w) (209)483-JOEF (M) > 5633 >
Re: Cannot delete tape in empty status... media state is incorrect - UPDATE
I have seen instances where a cartridge will show up as empty when it is allocated to a large backup and will only show up as full when the session is complete. On Mon, 3 Sep 2001, Robin Sharpe wrote: > Dear colleagues, > > I didn't think this would resolve this way, but... I just looked at this > tape again, and it IS NOT EMPTY ANYMORE! I never saw it go into the > scratch pool, and it is still in the same storage pool (ARCHLOG_TAPE). > That storage pool is used for Oracle achived redo logs, and we back those > up every hour or two on every system with an Oracle database in archive > mode. Here's a clip from my activity log... > > 08/31/01 16:16:52 ANR1425W Scratch volume DBR086 is empty but will not > be deleted - volume state is "mountablenotinlib". > 09/01/01 10:32:56 ANR8337I DLT volume DBR086 mounted in drive DRIVE3 > (/dev/mt11). > 09/01/01 11:49:57 ANR8325I Dismounting volume DBR086 - 60 minute mount > retention expired. > 09/01/01 11:50:18 ANR8336I Verifying label of DLT volume DBR086 in > drive DRIVE3 (/dev/mt11). > 09/01/01 11:51:46 ANR8468I DLT volume DBR086 dismounted from drive > DRIVE3 (/dev/mt11) in library ATL. > 09/02/01 09:18:27 ANR1228I Removable volume DBR086 is required for > storage pool backup. > 09/02/01 09:18:27 ANR8324I DLT volume DBR086 is expected to be mounted > (R/W). > 09/02/01 09:20:00 ANR8337I DLT volume DBR086 mounted in drive DRIVE0 > (/dev/mt8). > > Looks like it was used yesterday morning for a storage pool backup. I'm > wondering now if another admin defined this tape as belonging to that > storage pool... although I would think I should have been able to delete > it. And there's still the media state discrepancy issue. > > Oh well, I guess we'll just move forward, and wath to see if it happens > again! > > Thanks to all for your suggestions. > > Robin Sharpe > Berlex Laboratories >
Re: Include Exclude Problem
The list works from bottom to top. I think your second version is what you want if you switch the two lines. On Sun, 22 Jul 2001, John Freeman wrote: > Fellow TSM's > > As a TSM newbie I am having some difficulty configuring an incl excl list. > > (Windows Client 4.1.3 to Windows Server 4.1.0) > > I want to backup a single directory on the D drive of a client and want to exclude >everything else on both C and D drives. > > My incl_excl file looks as follows: > > include "d:\data\*" > exclude "c:\*" > > I have also tried the following: > > include "d:\data\*" > exclude "*" > > The first combination backs up everything and the second one backs up nothing. > > Some help would be greatly appreciated. > > Rgds > John >
Re: Management Class
Yes, that is the procedure for all new mgmtc. Remember that for most agents, the retention/deletion is controlled by the agent. On Mon, 23 Apr 2001, Zosimo Noriega (ADNOC IS&T) wrote: > Hello All, > I created new management class for SQL and Exchange agents. Do I need to > validate and activate the policy domain from policy sets object or what are > the next steps after creating the new management class? > > Thanks, > Zosi >
Re: Hung Admin Sessions
Are you seeing any BA client sessions going thru? Is expiration running? It certainly sounds as if you're creating deadlock with the SQL quieries. On Fri, 23 Feb 2001, Jennifer I. Moore wrote: > Gentle Greetings, > > Has anyone noticed any situations where Admin sessions on the server get > hung and become "ghost" sessions that will not go away? Cancelling has no > effect. Opening a separate Admin session and cancelling from there has no > effect. In some cases, subsequent sessions get hung and a reboot is > necessary. And in other cases, there seems to be no adverse side effects. > > In this case, I was doing some SQL queries on the database, and two user > Admin sessions hung and two Query Admin sessions hung. (Fortunately, > everything else is continuing to run smoothly.) Below is the output from > "show threads" related to these 4 sessions. > > We are running 4.1.2 Server on an IBM M80 running AIX 4.3.3. > > Thank you in advance for any help and/or suggestions. > > Jennifer > > > Thread 68: SessionThread > tid=4437, ktid=34577, ptid=52, det=1, zomb=0, join=0, result=0, sess=31636 > Awaiting cond 5c355b04 (mutex 301a65b0) at 10009b38 (outGetNext) > > Thread 101: SessionThread > tid=653d, ktid=69811, ptid=52, det=1, zomb=0, join=0, result=0, sess=31768 >Awaiting cond 5c33a484 (mutex 301a65b0) at 10009b38 (outGetNext) > > Thread 105: SessionThread > tid=6963, ktid=72547, ptid=52, det=1, zomb=0, join=0, result=0, sess=30446 >Awaiting cond 5c286354 (mutex 301a65b0) at 10009b38 (outGetNext) > > Thread 108: SessionThread > tid=6acf, ktid=52079, ptid=52, det=1, zomb=0, join=0, result=0, sess=31704 >Awaiting cond 5c1c69d4 (mutex 301a65b0) at 10009b38 (outGetNext) >
Re: import node into a different domain
The only way I find around this was to recreate the old Domain on the new server, with a stgpool if necessary, and then update the node to the new domain after the import. If a stgpool is necessary, I make it so small that it migrates to the desired stgpool immediately. On Thu, 8 Feb 2001, Sheelagh Treweek wrote: > Hi, > > I have several node definitions I want to move from serverA to serverB, > also changing the name of the domains as I go. Seems plausible. > > I have done export node filedata=none and then import node with dom=new > which successfully does nothing at all and imports nothing. I can only > get it to import if I specify a domain exactly like on the original > server. I thought that specifying dom=xxx on import would override what > was in the export file? I have tried with/without replacedefs option. > > Am I missing something obvious here? > > [AIX 433 and TSM 3740] > > Thanks for any help, > Sheelagh > -- > Sheelagh Treweek > Oxford University Computing Services > Email: [EMAIL PROTECTED] > Phone: +44 (0)1865 273205 Fax:-273275 >
Re: tsm: what is dsmserv.dsk
I found the same thing. When I looked, my dsmserv.dsk was intact and correct. I cd'd to the appropriate bin directory for the dsmserv, repeated the command, and was on my way. The other error or oversight in the install for 3.7.0 was that COMMethod is now required in the server options file as well as the client file. That was news to the guy at Tivoli support as well as me. On Fri, 29 Sep 2000, Don Prince wrote: > Just my first attempt to get Tivoli Storage Manager for AIX-RS/6000 > Version 3, Release 7, Level 0.0 to work! > > This is what I get ... > > # cd /usr/tivoli/tsm/server/bin > # ./dsmserv > ANR7800I DSMSERV generated at 16:19:10 on Aug 24 1999. > > Tivoli Storage Manager for AIX-RS/6000 > Version 3, Release 7, Level 0.0 > > Licensed Materials - Property of IBM > > 5697-TSM (C) Copyright IBM Corporation 1999. All rights reserved. > U.S. Government Users Restricted Rights - Use, duplication or disclosure > restricted by GSA ADP Schedule Contract with IBM Corporation. > > ANR0900I Processing options file /usr/tivoli/tsm/server/bin/dsmserv.opt. > ANR0990I Server restart-recovery in progress. > ANR0212E Unable to read disk definition file dsmserv.dsk. > > >>> WHAT IS THIS "disk definition file dsmserv.dsk"? <<< > > > TIA > Don Prince >
Re: Import failure
Kelly, The README for 3.7.3.8 says that the problem with multitape imports is fixed, which is why I only posted my results from that release. Friday night from home I moved the export tapes back to the export machine and tried to import with preview from those tapes. The result is the same. So now I think it's time for the hardware guys to have a look. On Fri, 22 Sep 2000, Kelly J. Lipp wrote: > More or less a known problem. Check the PMRs as I don't know if there has > been a resolution. >
