Re: Disk->Tape migration, cache and maxnummp
Interpretation ought to be correct except the number. 98% used with 0.2% migratable means 0.2% real data and 97.8% cached. If backup requires space and file is not larger than maxsize/poolsize cached data ought to be purged. Verify for (order of probability): 1. maxsize limit on the diskpool 2. extremely large file in the backup 3. some of diskpool volumes being read-only of varied off 4. nasty bug with cached data As a workaround (and according to best practices) modify the schedules not to overlap. Zlatko Krastev IT Consultant "Prather, Wanda" <[EMAIL PROTECTED]> Sent by: "ADSM: Dist Stor Manager" <[EMAIL PROTECTED]> 20.02.2003 20:35 Please respond to "ADSM: Dist Stor Manager" To: [EMAIL PROTECTED] cc: Subject: Re: Disk->Tape migration, cache and maxnummp Backup to the diskpool does not require a mount point. But it does require adequate space in the disk pool. My interpretation would be that the backup client has something to back up that is larger than 2% of the diskpool. The client sees that it can't do the incremental backup to the disk pool, tries to switch direct to tape, but can't because it is already at maxnummp. -Original Message- From: Conko, Steven [mailto:[EMAIL PROTECTED] Sent: Thursday, February 20, 2003 9:50 AM To: [EMAIL PROTECTED] Subject: Disk->Tape migration, cache and maxnummp Question about how this scenario plays out Client starts an archive backup of a database to server at 1700 from cron job. Archive is straight to tape. Uses all mountpoints assigned to it (4). Client then starts a scheduled incremental "system-level" backup at 1800 while original cron'd backup is still running. Incremental goes to diskpool. Diskpool is at 98% utilized, but 0.2 % migratable. Cache is on. Incremental fails due to media mount not possible. Does TSM consider the connection to the diskpool as an extra mount point and therefore does not allow it because the client is already using its max number of mount points (4) on the original archive? TSM Server is AIX 4.3.3 ML10 running TSM V5.1.5 TSM Client is AIX 4.3.3 ML9 running TSM V5.1.5. Library is a 3494, 6 drives.
Re: Disk->Tape migration, cache and maxnummp
Backup to the diskpool does not require a mount point. But it does require adequate space in the disk pool. My interpretation would be that the backup client has something to back up that is larger than 2% of the diskpool. The client sees that it can't do the incremental backup to the disk pool, tries to switch direct to tape, but can't because it is already at maxnummp. -Original Message- From: Conko, Steven [mailto:[EMAIL PROTECTED]] Sent: Thursday, February 20, 2003 9:50 AM To: [EMAIL PROTECTED] Subject: Disk->Tape migration, cache and maxnummp Question about how this scenario plays out Client starts an archive backup of a database to server at 1700 from cron job. Archive is straight to tape. Uses all mountpoints assigned to it (4). Client then starts a scheduled incremental "system-level" backup at 1800 while original cron'd backup is still running. Incremental goes to diskpool. Diskpool is at 98% utilized, but 0.2 % migratable. Cache is on. Incremental fails due to media mount not possible. Does TSM consider the connection to the diskpool as an extra mount point and therefore does not allow it because the client is already using its max number of mount points (4) on the original archive? TSM Server is AIX 4.3.3 ML10 running TSM V5.1.5 TSM Client is AIX 4.3.3 ML9 running TSM V5.1.5. Library is a 3494, 6 drives.
Disk->Tape migration, cache and maxnummp
Question about how this scenario plays out Client starts an archive backup of a database to server at 1700 from cron job. Archive is straight to tape. Uses all mountpoints assigned to it (4). Client then starts a scheduled incremental "system-level" backup at 1800 while original cron'd backup is still running. Incremental goes to diskpool. Diskpool is at 98% utilized, but 0.2 % migratable. Cache is on. Incremental fails due to media mount not possible. Does TSM consider the connection to the diskpool as an extra mount point and therefore does not allow it because the client is already using its max number of mount points (4) on the original archive? TSM Server is AIX 4.3.3 ML10 running TSM V5.1.5 TSM Client is AIX 4.3.3 ML9 running TSM V5.1.5. Library is a 3494, 6 drives.