I have a problem with my warm standby system. One of the transactions caused an error and stopped DSI thread for secondary database. After issuing "resume connection to <server>.<db> skip transaction" command replication agent in primary database won't resume processing of transaction log. All replication threads are active, replication agent in primary database is running. When I query syslogshold table, $replication_truncation_point remains on the same page. The only thing that helped so far was restart of replication agent & replication server. Has anyone encountered this problem before ? Sybase version 12.5.3/EBF 13325 ESD#7 on Solaris 8.
Yes, RA was hanging, but I'd like to find out why (unfortunately there were no errors in repserver or dataserver errorlogs regarding this issue). Also, I've noticed that RA can be stopped only with 'nowait' option.
Well, now I got this error in repserver log after stopping replication agent with 'nowait' option:
E. 2007/09/18 17:15:26. ERROR #60024 REP AGENT(<server>.<database>) - /execint.c(3508)
Synchronization model CONDITION VARIABLE, element Executor <repserver_name>_ra TSD CV 15229b8 was interrupted waiting for a lock. (4107)eneric/useful/sync.c.
After I tried to start RA again, I keep getting this error:
E. 2007/09/18 17:16:36. ERROR #14067 USER(<repserver_name>_ra) - /execint.c(1815)
Replication Agent for '<server_name>.<db_name>' is in the process of disconnecting. Possibly wait and try again.