I have run into this before:
set COMMTIMEOUT 1800      ( or something large)
restart the ADSM server

Reasonable guess:
If you do a Q SESS during the backup you will notice that some sessions
appear to stop from time to time - this is something to do with the a delay
in Oracle accessing the tablespaces you wish to back up. The default value
of COMMTIMEOUT is 60 seconds so the ADSM server not having received anything
from the API thinks the connection has timed out - the 15 minute delay you
see before the session is cancelled is another default - forge which
parameter.

The only drawback in making COMMTIMEOUT too large is that if there is a real
error you have to wait for that value until you are notified that the backp
has failed - I just set it to 3600 as nobody was going to do anything about
a failed backup unitl the next working day

Hope this helps

"Rupp Thomas (Illwerke)" wrote:

> Hi ADSM-ers,
>
> after installing the "ADSMConnect Agent for Oracle Backup
> on Windows NT" on Oracle 8 I had the following situation:
> Our DB guru was able to start a backup, ADSM started a
> session, transferred about 500 Bytes and then went into
> status Idle Wait.
> After 15 minutes the session is cancelled (max. wait time).
> On Oracle the RMAN process has to be stopped manually
> to make the DB accessible again.
> No error messages are found in DSIERROR.LOG but our
> DB guru gave me the following log file.
>
> ------------------------------------------------------------------------
> -----------
> Job:           Backup0049
> Status:        Failed
> Destination:   SMO.illwerke.at
> Time:          28-01-1999 01.17.15 PM
>
> Recovery Manager: Release 8.0.5.0.0 - Production
>
> RMAN-06005: connected to target database: SMO
> RMAN-06008: connected to recovery catalog database
>
> RMAN> run {
> 2> allocate channel adsm type 'SBT_TAPE' format 'full_%d_s%s_t%t_p%p';
> 3> set command id to '0:2Backup0049';
> 4> backup incremental level 0
> 5>    ( database );
> 6> }
> 7>
> RMAN-03022: compiling command: allocate
> RMAN-03023: executing command: allocate
> RMAN-08030: allocated channel: adsm
> RMAN-08500: channel adsm: sid=14 devtype=SBT_TAPE
>
> RMAN-03022: compiling command: set
> RMAN-03023: executing command: set command id
>
> RMAN-03022: compiling command: backup
> RMAN-03025: performing implicit partial resync of recovery catalog
> RMAN-03023: executing command: partial resync
> RMAN-08003: starting partial resync of recovery catalog
> RMAN-08005: partial resync complete
> RMAN-03023: executing command: backup
> RMAN-08008: channel adsm: starting datafile backupset
> RMAN-08502: set_count=11 set_stamp=355928981
> RMAN-08010: channel adsm: including datafile 5 in backupset
> RMAN-08010: channel adsm: including datafile 6 in backupset
> RMAN-08010: channel adsm: including datafile 1 in backupset
> RMAN-08011: channel adsm: including current controlfile in backupset
> RMAN-08010: channel adsm: including datafile 4 in backupset
> RMAN-08010: channel adsm: including datafile 7 in backupset
> RMAN-08010: channel adsm: including datafile 2 in backupset
> RMAN-08010: channel adsm: including datafile 8 in backupset
> RMAN-08010: channel adsm: including datafile 3 in backupset
> RMAN-00569: ================error message stack follows================
> RMAN-00601: fatal error in recovery manager
> RMAN-03004: fatal error during execution of command
> RMAN-10039: error encountered while polling for RPC completion on
> channel adsm
> RMAN-10006: error running sql statement: select act from x$ksusex where
> sid=:1 and serial=:2
> RMAN-10002: ORACLE error: ORA-03113: end-of-file on communication
> channel
> child process exited abnormally
> ------------------------------------------------------------------------
> -----------
>
> Does anybody know what could cause this situation?
> Any help will be greatly appreciated!
>
> Kind regards
> Thomas Rupp
> Vorarlberger Illwerke AG
> MAIL:   thomas.rupp AT illwerke.co DOT at
> TEL:    ++43/5574/4991-251
> FAX:    ++43/5574/4991-820-8251