Re: RMAN / controlfile / TDPO issue

On Wed, 02 Aug 2006 09:02:58 -0400, Evan Ehrenhalt <eehrenh_at_emory.edu> wrote:

>HELP!!!
>
>We are seeing an intermittent error when using RMAN to backup our
>archive logs.
>
>The archive logs are moved to tape and deleted, but there are errors on
>the controlfile backup. Whenever this happens, the logs themselves seem
>fine. Just the controlfile fails to backup.
>
>Can anyone tell me why the controlfile backup would cause this error,
>when the archivelog backup goes fine?
>
>
>Here is the background:
>Oracle 9.2.0.6 + security patches.
>RMAN is 9.2.0.6.0 (same ORACLE_HOME)
>Tivoli Data Protection for Oracle (TDPO) 5.2.0.0
>
>----------------- RMAN Logfile ------------------
>Starting backup at 28-JUL-06
>channel c1: starting archive log backupset
>channel c1: specifying archive log(s) in backup set
>input archive log thread=1 sequence=103986 recid=102783 stamp=578189102
>channel c1: starting piece 1 at 28-JUL-06
>channel c2: starting archive log backupset
>channel c2: specifying archive log(s) in backup set
>input archive log thread=1 sequence=103987 recid=102784 stamp=578189702
>channel c2: starting piece 1 at 28-JUL-06
>released channel: c1
>released channel: c2
>released channel: c3
>RMAN-00571: ===========================================================
>RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
>RMAN-00571: ===========================================================
>RMAN-03009: failure of backup command on c1 channel at 07/28/2006 00:20:02
>ORA-19506: failed to create sequential file,
>name="DB1P_ARCHIVE_20060728_578190000_s19709_p1", parms=""
>ORA-27028: skgfqcre: sbtbackup returned error
>ORA-19511: Error received from media manager layer, error text:
> ANS1017E (RC-50) Session rejected: TCP/IP connection failure
>
>RMAN>
>
>----------------- RMAN SCRIPT ------------------
>RMAN script and settings:
>connect target /
>connect catalog $logn
>configure backup optimization on;
>configure default device type to 'sbt_tape';
>configure channel device type 'sbt_tape' parms
>'ENV=(TDPO_OPTFILE=/opt/tivoli/tsm/client/oracle/bin64/tdpo.opt)';
>configure controlfile autobackup on;
>run {
>allocate channel c1 type 'sbt_tape' parms
>'ENV=(DSMI_ORC_CONFIG=/opt/tivoli/tsm/client/oracle/bin64/dsm.opt)';
>
>backup filesperset 10
>format '%d_ARCHIVE_%T_%t_s%s_p%p' tag '$ORACLE_SID Archive'
>(archivelog like '/u01/app/oracle/admin/${ORACLE_SID}/arch/%' delete input);
>
>release channel c1;
>}
>
>----------------- Typical Tivoli Log ------------------
>
>
>
>
>
>***** Each backup has a start, a number of bytes send, and an end.
>
>***** In this example, 205017,
>***** It starts,
>***** backs up R25PROD_ARCHIVE_20060731_597246624_s29484_p1
>***** Size 33554432 bytes,
>***** then ends.
>07/31/06 13:50:25 ANR0406I Session 205017 started for node URANIAORA
>TDP Oracle SUN) (Tcp/Ip 170.140.13.75(41834)). (SESSION: 2050
>07/31/06 13:50:35 ANE4991I (Session: 205017, Node: URANIAORA) TDP
>Oracle SUN ANU0599 ANU2535I File
>/adsmorc//R25PROD_ARCHIVE_20060731_597246624_s29484_p1 = 33554432 bytes
>sent (SESSION: 205017)
>07/31/06 13:50:35 ANR0403I Session 205017 ended for node URANIAORA
>(TDP Oracle SUN). (SESSION: 205017)
>
>
>***** Controlfile success
>***** Start backup controlfile by deleting the old copy (never exists).
>07/31/06 14:20:23 ANR0406I Session 205103 started for node URANIAORA
>(TDP Oracle SUN) (Tcp/Ip 170.140.13.75(42121)). (SESSION: 205103)
>
>07/31/06 14:20:23 ANE4994S (Session: 205103, Node: URANIAORA) TDP
>Oracle SUN ANU0599 ANU2602E The object
>/adsmorc//c-2375565327-20060731-00 was not found on the TSM
>Server(SESSION: 205103)
>
>07/31/06 14:20:23 ANR0403I Session 205103 ended for node URANIAORA
>(TDP Oracle SUN). (SESSION: 205103)
>
>***** Now do the actual controlfile backup:
>
>07/31/06 14:20:23 ANR0406I Session 205104 started for node URANIAORA
>(TDP Oracle SUN) (Tcp/Ip 170.140.13.75(42122)). (SESSION: 205104)
>
>07/31/06 14:20:24 ANE4991I (Session: 205104, Node: URANIAORA) TDP
>Oracle SUN ANU0599 ANU2535I File /adsmorc//c-2375565327-20060731-00 =
>1835008 bytes sent (SESSION: 205104)
>07/31/06 14:20:24 ANR0403I Session 205104 ended for node URANIAORA
>(TDP Oracle SUN). (SESSION: 205104)
>
>
>***** Control file failure, it looks like this for a failure
>07/31/06 13:50:35 ANR0406I Session 205020 started for node URANIAORA
>(TDP Oracle SUN) (Tcp/Ip 170.140.13.75(41837)). (SESSION: 205020)
>
>07/31/06 13:50:35 ANR0403I Session 205020 ended for node URANIAORA
>(TDP Oracle SUN). (SESSION: 205020)
>
>No action, just a start, then an end. While on the RMAN side I get the
>error above.
>
>Any thoughts?
>
>Thanks

The RMAN log file sugggest you allocated 3 channels in the actual backup (c1,c2,c3), whereas the RMAN-script only allocates one channel (c1).

Furthermore, the RMAN log suggests that the creation of "DB1P_ARCHIVE_20060728_578190000_s19709_p1" failed, which happens to follow the archivelog backup format, not the controlfile autobackup format.

My guess(es) :

1. You run another RMAN script than you think you do.
2. The script you actually run has incorrect TSM parameters or the TSM server is simply down, in both cases

>ORA-27028: skgfqcre: sbtbackup returned error
>ORA-19511: Error received from media manager layer, error text:
> ANS1017E (RC-50) Session rejected: TCP/IP connection failure

is frequently seen.

• Kenneth Koenraadt