Tag: EDIT CONFIGURATION SET PROPERTY OperationTimeout

Process RSM0, PID = 23456, will be killed

Standby database DG log:

11/09/2022 11:53:34
Creating process RSM0
11/09/2022 11:53:37
Process RSM0 re-created with PID = 20918
11/09/2022 11:59:37
Process RSM0, PID = 20918, will be killed
11/09/2022 11:59:51

Primary database DG log:

Data Guard Broker Status Summary:
Type Name Severity Status
Configuration DG_CDRDB Warning ORA-16607
Primary Database PCDRDB Error ORA-16778
Physical Standby database SCDRDB Error ORA-16810

SOLUTION

  1. Check and resolve network issues between primary db server and standby server.
  2. Increase the OperationTimeout property value :
DGMGRL> show configuration OperationTimeout;
  OperationTimeout = '30'
 
DGMGRL> EDIT CONFIGURATION SET PROPERTY OperationTimeout=90;
Property "operationtimeout" updated

DGMGRL>  show configuration OperationTimeout;
  OperationTimeout = '90'
Advertisement

The dataguard broker keeps killing process RSM

Dataguard broker keeps killing RSM process overnight and oncall DBA is pagered by alerts.

DG 2014-10-04-04:17:17 0 2 0 DMON: killing process RSM0, pid = 26742
DG 2014-10-04-04:17:18 0 2 0 DMON: waiting for subscribers to disappear...
DG 2014-10-04-04:17:18 0 2 0 PMON: delete state object for RSM0
DG 2014-10-04-04:17:18 0 2 0 PMON: RSM0 died unexpectedly while processing request 1.1.833218860,
DG 2014-10-04-04:17:18 0 2 0 notifying DMON of RSM failure
DG 2014-10-04-04:17:18 0 2 0 DMON: Creating process RSM0
DG 2014-10-04-04:17:21 0 2 0 RSM0: Attach state object
DG 2014-10-04-04:17:21 0 2 0 DMON: Process RSM0 re-created with pid = 22898

WORKAROUND

DGMGRL> show configuration OperationTimeout;
OperationTimeout = '30'

DGMGRL> EDIT CONFIGURATION SET PROPERTY OperationTimeout=120;
Property "operationtimeout" updated

DGMGRL> show configuration OperationTimeout;
OperationTimeout = '120'
DGMGRL>

Reference Oracle Doc:
Data Guard: Server Hang And Crash Because RSM0 Keeps Re-spawning (Doc ID 1322877.1)
Cause : The Broker is killing the RSM0 process and restarting it because of an internal timeout that is too low.