ora 3113 while compiling partial resync [RMAN]

[abhijit]

Hi all!

I am getting following error while taking backup through rman catalog. My script backup the datafiles -> switch to current log -> take backup of the archives and then delete the archives. It backed up all the datafiles but not the archives and failed to partially resync with catalog.

Extract @ debug mode:

krmqgns: looking for work for channel t6
krmqgns: commands remaining to be executed:
krmqgns: no work found for channel t6

krmkomp: the following SET options are in effect:
1 JOBSET max corrupt=0 for datafile 1
2 JOBSET max corrupt=0 for datafile 2
3 JOBSET max corrupt=0 for datafile 3

EXEC SQL AT RCVCAT yProxy := dbms_rcvman . getAnyProxy ; :getCfBackup := dbms_rc
vman . getCfBackup ; :listCfCopy := dbms_rcvman . listCfCopy ; :listDfCopy := db
ms_rcvman . listDfCopy ; :listCfBackup := dbms_rcvman . listCfBackup ; :listDfBa
ckup := dbms_rcvman . listDfBackup ; :listAlBackup := dbms_rcvman . listAlBackup
; :listDfProxy := dbms_rcvman . listDfProxy ; :getRecovAction := dbms_rcvman .
getRecovAction ; :getAlBackup := dbms_rcvman . getAlBackup ; :listAlCopy := dbms
_rcvman . listAlCopy ; end ;
sqlcode=-3113
krmicomp: error 6004 signalled during compilation
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure during compilation of command
RMAN-03013: command type: partial resync
RMAN-06004: ORACLE error from recovery catalog database: ORA-03113: end-of-file
on communication channel
RMAN-06097: text of failing SQL statement: begin dbms_rcvman . resetAll ; :succe
ss := dbms_rcvman . SUCCESS ; :av
RMAN-06099: error occurred in source file: krmk.pc, line: 9397

Recovery Manager complete.
krmxcd: xc=1076186168
krmxcd: xc=1076195376
krmxcd: xc=1076196352

Env: Oracle 8.1.7 ED 64 Bits, HP-UX 11.00

I appreciate your valuable suggestion.

Thanks,

abhijit

[sambavan]

Could you post your script. The problem could be at the place where you do the log switch.

Sam

[abhijit]

Here is my script:

run {

allocate channel t1 type 'SBT_TAPE';
allocate channel t2 type 'SBT_TAPE';
allocate channel t3 type 'SBT_TAPE';
allocate channel t4 type 'SBT_TAPE';
allocate channel t5 type 'SBT_TAPE';
allocate channel t6 type 'SBT_TAPE';
setlimit channel t1 maxopenfiles 32 readrate 200;
setlimit channel t2 maxopenfiles 32 readrate 200;
setlimit channel t3 maxopenfiles 32 readrate 200;
setlimit channel t4 maxopenfiles 32 readrate 200;
setlimit channel t5 maxopenfiles 32 readrate 200;
setlimit channel t6 maxopenfiles 32 readrate 200;


set maxcorrupt for datafile 1,2,3 to 0;

#--------------- BACKUP FULL DATABASE LEVEL 0 --------------
backup incremental level 0
tag backup_db_level0
filesperset 10
format 'df_t%t_s%s_p%p' (database);

#------------------------ ARCHIVE LOG --------------------------#


sql 'alter system archive log current';

#-------------------- BACKUP ARCHIVE LOGS ------------------#

backup
filesperset 10 format 'arch_t%t_s%s_p%p'
(archivelog all delete input);

#--------------------- RELEASE ALL CHANNEL -------------------#

release channel t1;
release channel t2;
release channel t3;
release channel t4;
release channel t5;
release channel t6;
}

Waiting for your suggestion

abhijit

[sambavan]

1. Are you using dedicated connection to connect to the target and the catalog databases?

2. Try to release and reallocate the channels after the db backup.


Sam

[abhijit]

1. Are you using dedicated connection to connect to the target and the catalog databases?
= What do mean by that. I am not clear about the query. Pl. explain it elaborately...

2. Try to release and reallocate the channels after the db backup.
=Tried that but getting the same erorr..

Thanks for your reply...

abhijit

[sonofsita]

If the target database and catalog are two different entities during the backup you may verify the alert log file for the catalog database. the catalog database is getting shut down.
If the catalog database is sutting down by itself the reasons may have to be looked into. Otherwise"Do not shutdown the catalog database during backups of target database(s)."

[abhijit]

It's not the case with me. Catalog database is up while it is backing up the target one. Error code, lost of connection misleading us to getting shutdown of catalog databse.

It's freaky, ora 03113 didn't generate any trace file...

I appreciate your suggestion...

abhijit

[sambavan]

Does your database work on the MTS environment. To check do

lsnrctl services

Do you see the name dispatcher in the list. That means that your database is using MTS. On such case you want to make sure that your RMAN connectivity goes through the dedicated connection and not the MTS.

On the other hand ORA-3113 would occour in a wide range of scenarios...

Sam

[abhijit]

No, it's running on dedicated connection...

Will try to trace each channel today nite to see which channel is failing...

thanks,

abhijit

[sambavan]

Here are some possible causes:

1. Listener is not communicating properly. If so stop and start your listener.

2. Check to make sure that no other process/script that would take down the catalog database when the RMAN backup starts....

3. I would also check that I'm able to just open and close all those channels without doing any backup.

4. check sbtio.log under target db's udump

These are some I could think of ...

Sam