Include/Exclude: UNCs vs. drive letters
Our TSM server is absorbing the backups of many W2K servers, raising some issues I haven't had to deal with. How well does TSM handle UNC's in include/excludes? Which are preferred or easier to work with, UNC's or drive letter and wild cards? Are there any pitfalls I should avoid? Also, what do I need to do differently when W2K servers use DFS in concert with replication to "share" files/directories between servers? Your advice is appreciated... Ken
Include/Exclude: UNCs vs. drive letters
Our TSM server is absorbing the backups of many W2K servers, raising some issues I haven't had to deal with. How well does TSM handle UNC's in include/excludes? Which are preferred or easier to work with, UNC's or drive letter and wild cards? Are there any pitfalls I should avoid? Also, what do I need to do differently when W2K servers use DFS in concert with replication to "share" files/directories between servers? Your advice is appreciated... Ken
urgent - 4.2.1 client scheduler frequently stops during backup
Hello all I have the 4.2.1.32 client running on a W2K server over a T1 WAN link. The backup of this client is generally less than 500M per day, but the TSM scheduler service on that client regularly terminates unexpectedly with the following error in the W2K event log: Event Type:Error Event Source: Service Control Manager Event Category: None Event ID: 7031 Date: 7/16/2002 Time: 8:04:38 PM User: N/A Computer: FILESERVER Description: The TSM Scheduler service terminated unexpectedly. It has done this 31 time(s). The following corrective action will be taken in 0 milliseconds: No action. This client uses the CAD to manage the scheduler and web client access. When the scheduler terminates, the CAD restarts it 10 minutes later, until the startup window for the backup expires. There are no errors in the dsmsched.log or dsmerror.log. The dsmwebcl.log file shows the restart after 10 minutes. The server activity log merely shows ANR0480W - the connection was severed. Because this is a WAN client, I've looked at COMMTimeout (600 seconds), IDLETimeout (30 minutes), client compression (ON), TCPBuffSize (128), TCPWindowSize (default=32), TCPNoDelay (NO). I can't tell what is causing the scheduler service to terminate. Any ideas? Would running the scheduler service full time, rather than using the CAD to manage it, remedy this problem? Thanks... Ken
versioning for user's files
Hello all... We're migrating our NetWare servers, currently backed up with ArcServe, to W2K servers which will be backed up with TSM. These NetWare servers contain programs, users' home directories, and shared directories. The existing NetWare server backup is a full backup every day, all to one DLT tape. The number of tapes in rotation provides for going about six weeks, or 30 backups, before a tape is reused. Also, at the end of each month the tape is removed from the rotation and held for two years. My challenge is to provide a reasonable level of backup availability with TSM. All I've backed up with TSM thus far are application servers where users (other than administrators) don't have access. We keep 7 versions on those servers. Our NetWare users sometimes ask for restores of files more than a year old, and their requests have been fulfilled. I'm struggling to come up with a version scheme which will provide a similar level of service. It seems a bit extreme, but my first thought is to keep 30 versions and do a monthly backupset of the W2K file server. But that 30 versions is having a nasty effect on calculations of the resulting TSM database size. What kind of versioning do you set for user files, shared documents, etc., which are, in my mind, more volatile than application server files because users are involved? Any examples of the model you use would be appreciated. Thanks... Ken
4.2.1.32 problem with systemobject on Citrix
Hello all... Backups of a W2K server running Citrix Metaframe XPe are failing, with the following error in the server activity log: 04/30/02 18:17:36 ANR0480W Session 9355 for node CITRIXADTEST (WinNT) terminated - connection with client severed. After 10 minutes, the server reconnects with the client and the backup restarts, only to fail again a few minutes later. The following error occurs in the dsmerror.log, but I often see such failures on non-Citrix servers without the backup stopping: 04/30/2002 18:17:35 ANS4987E Error processing '\\citrixadtest\m $\WINNT\system32 \Perflib_Perfdata_380.dat': the object is in use by another process Following the invaluable information on this forum, I looked at the solution for the "sysfiles not on C:" problem, which I thought was fixed in 4.2.1.32. To test, I changed DOMAIN from ALL-LOCAL to just "M: N:" and the backup succeeded. I reset DOMAIN to "M: N: SYSTEMOBJECT" and the backup failed again. I'm using 4.2.1.32 on a W2K domain controller, backing up the system objects without problems. What's up, am I missing something, or is backing up system objects on Citrix servers still broken? Thanks... Ken
Re: 4.2.1.32 problem with systemobject on Citrix
Bruce, lower the duration of the startup window to reduce the amount of retries for these failed backups. Mine is set to 1 hour, so it isn't retrying all day. Bruce Kamp <[EMAIL PROTECTED]To: [EMAIL PROTECTED] T> cc: Sent by: Subject: Re: 4.2.1.32 problem with systemobject on Citrix "ADSM: Dist Stor Manager" <[EMAIL PROTECTED] RIST.EDU> 05/01/2002 07:18 AM Please respond to "ADSM: Dist Stor Manager" Join the club. I have been working with Tivoli for a couple of months on this (even though they don't support Citrix). They had me upgrade to 4.2.1.30. There thinking is that it has to do with the CAD starting scheduler. When I stopped using managed services it stopped hitting my TSM server every 10 minutes. I'm still testing the rest of it. I'll let you know what I find out --- Bruce Kamp Network Analyst II Memorial Healthcare System E: [EMAIL PROTECTED] P: (954) 987-2020 x4597 F: (954) 985-1404 ---- -Original Message- From: Ken Long [mailto:[EMAIL PROTECTED]] Sent: Tuesday, April 30, 2002 7:42 PM To: [EMAIL PROTECTED] Subject: 4.2.1.32 problem with systemobject on Citrix Hello all... Backups of a W2K server running Citrix Metaframe XPe are failing, with the following error in the server activity log: 04/30/02 18:17:36 ANR0480W Session 9355 for node CITRIXADTEST (WinNT) terminated - connection with client severed. After 10 minutes, the server reconnects with the client and the backup restarts, only to fail again a few minutes later. The following error occurs in the dsmerror.log, but I often see such failures on non-Citrix servers without the backup stopping: 04/30/2002 18:17:35 ANS4987E Error processing '\\citrixadtest\m $\WINNT\system32 \Perflib_Perfdata_380.dat': the object is in use by another process Following the invaluable information on this forum, I looked at the solution for the "sysfiles not on C:" problem, which I thought was fixed in 4.2.1.32. To test, I changed DOMAIN from ALL-LOCAL to just "M: N:" and the backup succeeded. I reset DOMAIN to "M: N: SYSTEMOBJECT" and the backup failed again. I'm using 4.2.1.32 on a W2K domain controller, backing up the system objects without problems. What's up, am I missing something, or is backing up system objects on Citrix servers still broken? Thanks... Ken
Re: 4.2.1.32 problem with systemobject on Citrix
I haven't looked into the 5.x clients yet, as my server is still at 4.1.4 and will be for awhile until some old clients go away. Does the 5.1 client work with a 4.1 server? Also, can I assume the fix will be applied to the 4.2 client as well? Thanks... Ken Rejean Larivee/QuebeTo: [EMAIL PROTECTED] c/IBMcc: Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED] RIST.EDU> 05/01/2002 08:06 AM Please respond to "ADSM: Dist Stor Manager" Hello Ken, I believe you are suffering from a known defect identified by IC32636. This apar describes a problem with the client having problems backing system files if the root drive is other than C:. Per your note below, it appears the boot drive is M:. This is why the client works fine if you don't include the systemobject in the domain option. This apar is currently fixed in the 5.1 client so you might want to install the 5.1.0.1 client. Have a good day ! - Rejean Larivee IBM - TSM/ADSM Level 2 Support Ken Long <[EMAIL PROTECTED]To: [EMAIL PROTECTED] OM> cc: Sent by: "ADSM: Subject: 4.2.1.32 problem with systemobject on Citrix Dist Stor Manager" <[EMAIL PROTECTED] .EDU> 04/30/2002 08:41 PM Please respond to "ADSM: Dist Stor Manager" Hello all... Backups of a W2K server running Citrix Metaframe XPe are failing, with the following error in the server activity log: 04/30/02 18:17:36 ANR0480W Session 9355 for node CITRIXADTEST (WinNT) terminated - connection with client severed. After 10 minutes, the server reconnects with the client and the backup restarts, only to fail again a few minutes later. The following error occurs in the dsmerror.log, but I often see such failures on non-Citrix servers without the backup stopping: 04/30/2002 18:17:35 ANS4987E Error processing '\\citrixadtest\m $\WINNT\system32 \Perflib_Perfdata_380.dat': the object is in use by another process Following the invaluable information on this forum, I looked at the solution for the "sysfiles not on C:" problem, which I thought was fixed in 4.2.1.32. To test, I changed DOMAIN from ALL-LOCAL to just "M: N:" and the backup succeeded. I reset DOMAIN to "M: N: SYSTEMOBJECT" and the backup failed again. I'm using 4.2.1.32 on a W2K domain controller, backing up the system objects without problems. What's up, am I missing something, or is backing up system objects on Citrix servers still broken? Thanks... Ken
ODBC and TSM timestamp data type (Crystal Reports)
I know some of you are running Crystal Reports against the TSM database, hopefully you can steer me straight. I can't get Crystal to render any of the TSM timestamp fields - I always get 0's for the timestamp. For example, the Start_Time field on the Summary table. All times display as 12:00:00 am. Browsing the data for the data for the field shows: DateTime(0,0,0,00,00,00). Needless to say, this makes it difficult to just show results from yesterday. I'm using Seagate Crystal Reports 8.0.1.1 and the TSM ODBC driver 4.02.01.00 (part of the 4.2.1 Win32 client.) ODBCADSM.DLL is dated 8/3/2001. Any help would be appreciated. Thanks... Ken
select * from events - only today's events
Am I missing something? I have Eventretention set to 90 days. I can use "query event" to see event data from days past. But if I use "select * from events" I only get events from today, no historical data. Shouldn't the Events table have previous days results in it? Thanks... Ken
Re: select * from events - only today's events
Gabriel, the query event part works fine, I'm able to see historical event data. It is when using select, which is more useful than query when feeding data to scripts, that I can't get history. I assume that selecting from the events table should provide the same data as using query event. Thanks... Ken Gabriel Wiley <[EMAIL PROTECTED]To: [EMAIL PROTECTED] M.COM> cc: Sent by: Subject: Re: select * from events - only today's events "ADSM: Dist Stor Manager" <[EMAIL PROTECTED] RIST.EDU> 11/21/2001 01:21 PM Please respond to "ADSM: Dist Stor Manager" Ken, Try this "q ev * * begind=xx endd=today" xx = how many days back do you want to go If you want to see only failures add "ex=y" to the end of the statement. There is also begint = begin time endt = end time Gabriel C. Wiley ADSM/TSM Administrator AIX Support Phone 1-614-308-6709 Pager 1-877-489-2867 Fax 1-614-308-6637 Cell 1-740-972-6441 Siempre Hay Esperanza Ken Long <[EMAIL PROTECTED]>@VM.MARIST.EDU> on 11/21/2001 01:00:38 PM Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] cc: Subject: select * from events - only today's events Am I missing something? I have Eventretention set to 90 days. I can use "query event" to see event data from days past. But if I use "select * from events" I only get events from today, no historical data. Shouldn't the Events table have previous days results in it? Thanks... Ken
Re: ODBC and TSM timestamp data type (Crystal Reports) WORK AROUND
Backing off to the 4.1.3 ODBC driver works for me also. Hopefully I don't run into lost functionality by backing off to an older version, but for now this will work. Tivoli's first level support needs to look a little deeper at some of these reported problems before throwing up their hands. Thanks to all who responded. William Boyer cc: Sent by: Subject: Re: ODBC and TSM timestamp data type (Crystal Reports) WORK "ADSM: Dist AROUND Stor Manager" <[EMAIL PROTECTED] IST.EDU> 11/21/2001 02:31 PM Please respond to bill.boyer I uninstalled the 4.2.1.0 ODBC driver, and installed the previous 4.1.3.0 ODBC driver. Now things are working just fine. I'll stick with this configuration for the time being. I will update the PMR with this information to see if Tivoli would like to pursue it. Bill -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED]]On Behalf Of Andrew Raibeck Sent: Tuesday, November 20, 2001 1:46 PM To: [EMAIL PROTECTED] Subject: Re: ODBC and TSM timestamp data type (Crystal Reports) Bill and Ken, Please follow the instructions in the "Troubleshooting/Diagnostics" section of the README.ODBC.TXT file to obtain traces of the problem, then open a problem with IBM support and provide them with the trace documentation (I'm not so concerned. Crystal Reports is not formally supported (i.e. we don't test it), so support is on a "best effort" basis. However, we are interested in understanding problems like this with the ODBC driver and fixing them when possible, so you shouldn't be flat-out refused (it even says so in the README file!). Regards, Andy Andy Raibeck IBM Software Group Tivoli Storage Manager Client Development Internal Notes e-mail: Andrew Raibeck/Tucson/IBM@IBMUS Internet e-mail: [EMAIL PROTECTED] The only dumb question is the one that goes unasked. The command line is your friend. "Good enough" is the enemy of excellence. William Boyer <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> 11/20/2001 11:16 Please respond to bill.boyer To: [EMAIL PROTECTED] cc: Subject:Re: ODBC and TSM timestamp data type (Crystal Reports) I have the same problems. If I use Access or Excel to get the data via the same ODBC connection, the date/times are correct. Also, if you use a utility that Seagate distributes on the CD called SQLCON32.EXE with the same select statement I cut and pasted from my report, it displays correctly. I went through this with the email tech support and finally had to call in and open a problem. They sent me the latest patches for CR8.5, but they didn't have any effect. I just haven't had the time to call back in and break the new to them. If you come up with a solution, PLEASE post it?!?! Bill Boyer DSS, Inc. -Original Message- From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED]]On Behalf Of Ken Long Sent: Tuesday, November 20, 2001 12:04 PM To: [EMAIL PROTECTED] Subject: ODBC and TSM timestamp data type (Crystal Reports) I know some of you are running Crystal Reports against the TSM database, hopefully you can steer me straight. I can't get Crystal to render any of the TSM timestamp fields - I always get 0's for the timestamp. For example, the Start_Time field on the Summary table. All times display as 12:00:00 am. Browsing the data for the data for the field shows: DateTime(0,0,0,00,00,00). Needless to say, this makes it difficult to just show results from yesterday. I'm using Seagate Crystal Reports 8.0.1.1 and the TSM ODBC driver 4.02.01.00 (part of the 4.2.1 Win32 client.) ODBCADSM.DLL is dated 8/3/2001. Any help would be appreciated. Thanks... Ken
Re: select * from events - only today's events
Even if I boost the number of hours in your select to 60, I still only get today's events. Am I correct in assuming that when I use "query event" the data I receive is from the events table that I'm running select against? Mahesh Tailor cc: Sent by: Subject: Re: select * from events - only today's events "ADSM: Dist Stor Manager" <[EMAIL PROTECTED] RIST.EDU> 11/21/2001 03:23 PM Please respond to "ADSM: Dist Stor Manager" Ken, Try this . . . select node_name,status from events where status!='Completed' and domain_name!='' and \ cast((current_timestamp-actual_start)hours as decimal(8,0))<19 and \ '$DATE' < scheduled_start order by node_name In this case $DATE is a variable that stores the date (computed using DATE ="$MONTH/$DAY/$YEAR") and cast((current_timestamp-actual_start)hours as decimal(8,0))<19 catches all events in the past 19 hours. Hope this helps. >>> [EMAIL PROTECTED] 11/21/01 02:52PM >>> Gabriel, the query event part works fine, I'm able to see historical event data. It is when using select, which is more useful than query when feeding data to scripts, that I can't get history. I assume that selecting from the events table should provide the same data as using query event. Thanks... Ken Gabriel Wiley <[EMAIL PROTECTED]To: [EMAIL PROTECTED] M.COM> cc: Sent by: Subject: Re: select * from events - only today's events "ADSM: Dist Stor Manager" <[EMAIL PROTECTED] RIST.EDU> 11/21/2001 01:21 PM Please respond to "ADSM: Dist Stor Manager" Ken, Try this "q ev * * begind=xx endd=today" xx = how many days back do you want to go If you want to see only failures add "ex=y" to the end of the statement. There is also begint = begin time endt = end time Gabriel C. Wiley ADSM/TSM Administrator AIX Support Phone 1-614-308-6709 Pager 1-877-489-2867 Fax 1-614-308-6637 Cell 1-740-972-6441 Siempre Hay Esperanza Ken Long <[EMAIL PROTECTED]>@VM.MARIST.EDU> on 11/21/2001 01:00:38 PM Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] cc: Subject: select * from events - only today's events Am I missing something? I have Eventretention set to 90 days. I can use "query event" to see event data from days past. But if I use "select * from events" I only get events from today, no historical data. Shouldn't the Events table have previous days results in it? Thanks... Ken
Re: select * from events - only today's events
Thanks... I was just coming to that conclusion myself I finally got Crystal Reports to understand the timestamp fields (requires the TSM 4.1.3 ODBC driver, TSM 4.2.1 ODBC driver doesn't work with Crystal Reports 8.0). Crystal could see all of the data, where select * couldn't. I understand the desire to make it work like Q Events does, but for those of us new to using select/SQL it is unexpected. Select * should select all records, like it does for other tables. Thanks again to the extremely valuable resources using this forum! "Rushforth, Tim" <[EMAIL PROTECTED]To: [EMAIL PROTECTED] PEG.MB.CA> cc: Sent by: "ADSM: DistSubject: Re: select * from events - only today's events Stor Manager" <[EMAIL PROTECTED] U> 11/21/2001 04:49 PM Please respond to "ADSM: Dist Stor Manager" Ken: Select * from events only shows events for the current day. See http://msgs.adsm.org/cgi-bin/get/adsm0009/619.html for an explanation. Tim Rushforth City of Winnipeg To: [EMAIL PROTECTED] cc: Subject: select * from events - only today's events Am I missing something? I have Eventretention set to 90 days. I can use "query event" to see event data from days past. But if I use "select * from events" I only get events from today, no historical data. Shouldn't the Events table have previous days results in it? Thanks... Ken
Re: Script to CHECKOUT multiples tapes on SCSI library
How many of the various DR commands are usable if we haven't purchased the separate DRM module? Back in the ADSM v2 days, in lieu of purchasing DRM, we developed AIX shell scripts which manage the checkout process. Since then, I've ignored the DR commands as I thought they weren't available for my use because I didn't purchase DRM. Our current setup uses TSM server scripts to build a list of volumes to be checked out, then uses an AIX script to convert the list into a macro to execute the checkout commands. Farren Minns cc: Sent by: Subject: Re: Script to CHECKOUT multiples tapes on SCSI library "ADSM: Dist Stor Manager" <[EMAIL PROTECTED] RIST.EDU> 12/07/2001 09:14 AM Please respond to "ADSM: Dist Stor Manager" I use the command move drm * wherestate=mountable tostate=vault remove=bulk That should be the kind of thing you are after. Thanks Farren Minns Etienne Brodeur <[EMAIL PROTECTED]>@VM.MARIST.EDU> on 07/12/2001 14:13:59 Please respond to "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] cc: Subject: Re: Script to CHECKOUT multiples tapes on SCSI library Thanks for the help, From I have been reading the MOVE MEDIA command is to allow you to move onsite tapes to an overflow location. What I want to be able to do is to identify which tapes I need to send offsite and with one command check them out, instead of having to check the tapes out one at a time, ie one command per tape I want to checkout. Right now if I have 8 tapes to checkout to the offsite vault I need to execute 8 commands. I would like to be bale to use one command to do this such as specifying a list which is why I was trying to use the VOLFILE option. Is the MOVE MEDIA a command I could use to do this? Thanks again for the suggestion. Etienne Brodeur Martin Trcka cc: Sent by: Subject: Re: Script to CHECKOUT multiples tapes on SCSI library "ADSM: Dist Stor Manager" <[EMAIL PROTECTED] RIST.EDU> 2001-12-07 08:29 AM Please respond to "ADSM: Dist Stor Manager" Doesn't MOVE MEDIA command meet your needs ? Martin Trcka > -Original Message- > From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED]]On > Behalf Of Etienne Brodeur > Sent: Thursday, December 06, 2001 9:09 PM > To: [EMAIL PROTECTED] > Subject: Script to CHECKOUT multiples tapes on SCSI library > > > I have an LTO3583 connected to a Windows2000 TSM server 4.2.0 > > Is there a way with a script or the DSMADMC to check out multiples tapes > with one command? > > When I try the VOLLIST or VOLFILE options the sever tells me I can't use > this option with a SCSI library. Is there a similar options for a non > SCISI LIB? By writing a server scripts or something? > > Any help would be greatly appreciated. > > Etienne Brodeur
multiple buses to 3575-L18 on W2K using TotalStorage driver?
Hello all... I'm researching moving my TSM server from AIX to W2K. My tape library is a Magstar 3757-L18, six drives, each pair on its own SCSI bus. In the Installation and User's Guide for the IBM TotalStorage Tape Device Drivers, chapter 36 - W2K Device Driver Management, the last sentence in the Installation Overview states: "All drives accessible from a medium changer must be on the same physical SCSI bus as the changer." Am I misunderstanding this, or is it wrong? Is this saying I must put all six drives on one SCSI bus? Any advice would be appreciated. Thanks... Ken
Re: AW: multiple buses to 3575-L18 on W2K using TotalStorage driver?
Thanks, Thomas, this helps. I'll have to get familiar with RSM. Anyone know which driver, TotalStorage or RSM, is the preferred method for connecting TSM with this tape library? Ken Schoenleitner Thomas <[EMAIL PROTECTED] To: [EMAIL PROTECTED] SOZVERS.AT> cc: Sent by: "ADSM: Dist Stor Subject: AW: multiple buses to 3575-L18 on W2K using TotalStorage driver? Manager" <[EMAIL PROTECTED]> 03/06/2003 02:00 AM Please respond to "ADSM: Dist Stor Manager" hi, > Hello all... > > I'm researching moving my TSM server from AIX to W2K. My > tape library is a > Magstar 3757-L18, six drives, each pair on its own SCSI bus. > > In the Installation and User's Guide for the IBM TotalStorage > Tape Device > Drivers, chapter 36 - W2K Device Driver Management, the last > sentence in > the Installation Overview states: > > "All drives accessible from a medium changer must be on the > same physical > SCSI bus as the changer." this is true, if you want to use removeable storage manager (RSM) with tsm it should be no problem using different scsi buses. I hope this helps. thomas > Am I misunderstanding this, or is it wrong? Is this saying I > must put all > six drives on one SCSI bus? > > Any advice would be appreciated. > > Thanks... Ken >
Re: Inactive Files Rebound To New Managment Class?
Inactive versions of active files should be rebound. But files that have been removed for the client - where only inactive versions exist - should not rebind. "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> wrote on 11/10/2004 05:49:56 PM: > > The nodelete management class is set to nolimit across the board. I ran an > incremental backup thinking this would rebind only the active and new or > changed files. However, when I looked in the GUI under restore, I saw that > ALL the files, inactive included show up as using the "nodelete" > management class. Have I merely been here too long today, or am I seeing > something new here?
