Backup schedules via API code gives FAILED status - eve=
2015-10-04 17:59:02
Hi,
I have two seperate installations where the backup schedules via ADSM A=
PI
code (on AIX 4.2 or 4.1.5) at V2.1.0.6 and above shows the following in=
the
activity logs......with the event status indicating FAILED and the
DSMACCNT.LOG entry showing the expected amount of DB data having been
backed up (in fact the OBACKUP LOGS and BAR_ACT_LOG entries shows a cle=
an
backup). This is for DB2/2, DB2/6000, Oracle 7.3.2 (using EBU and
ADSMConnect) or Informix DB's. Since ADSM did not post any succesfull
backup completion status...can I be rest assured that I now have a
guaranteed backup and that it is restorable? And is there anyone out th=
ere
that knows how to settle this integrity/accuracy problem? Here follows =
a
typical extract from the activitylog.
01:00:32 ANR0400I Session 367 started for node XYZ (AIX)
01:00:32 ANR0484W Session 367 for node XYZ (AIX) terminated -
protocol
violation detected.
01:00:40 ANR0400I Session 368 started for node XYZ (DB2/6000)
01:01:20 ANR1000I Migration process 2254 started for stgpool
BACKUPPOOL
01:01:20 ANR0403I Session 368 ended for node XYZ (DB2/6000)
01:01:21 ANR0400I Session 369 started for node XYZ (AIX)
01:01:21 ANR2576W An attempt was made to update an event record f=
or a
scheduled operation which has already been executed=
-
multiple client
schedulers may be active for node XYZ
01:01:21 ANR0403I Session 369 ended for node XYZ (AIX)
01:01:29 ANR1001I Migration process 2254 ended for storagepool
BACKUPPOOL
01:01:52 ANR0400I Session 370 started for node XYZ (AIX)
01:01:52 ANR0403I Session 370 ended for node XYZ (AIX)
I do not believe the protocol violation has any bearing on the event re=
cord
updates as some schedules did not show the protocol violation error but=
still had an incorrect event status.
I cross checked umpteen times and there are not multiple schedulers run=
ning
on any of the client nodes,
I use the same node to backup my DB/2 databases ("su - dba '-c adsm
command'" to the right user authorities) as I use for backing up my ROO=
T
files. Can this pose a problem as seen in the event records? I
cross-checked on AIX for the correct syntax for the command via 'man su=
'.
Any input is welcomed.
Campbell Esplin
=
|
<Prev in Thread] |
Current Thread |
[Next in Thread> |
- Backup schedules via API code gives FAILED status - eve=,
Campbell Esplin [SMTP:Campbell <=
|
|
|