Problems with replication's scheduled jobs

From: Omen <bf109e_at_teleline.es>
Date: 18 Jun 2002 04:32:51 -0700
Message-ID: <25edf32d.0206180332.32ba1bed@posting.google.com>

Hi group:         

        I've got some problems with scheduled jobs for replication. The scenario is shown behind:         

        We've got two Linux machines with both Oracle 8.0.6 instances. One of them is the replication's source and the other is the target. Both, source and target, have two schemas (two users) each one. The replication source has two replication source schemas and the replication target has the two respectives replication schemas. Both instances are stopped at 23:00 because we make a cold physical backup. When the backup has ended, we start both instances. Those tasks are executed through crontab.

	The problem is:
	When we restart instances, only one of the replication jobs works

fine, the other is wrong. If we force a refresh of the refresh group wich doesn't work, the replication works fine in both instances and the four schemas (until next dbshut-dbstart)

	I've added 
	JOB_QUEUE_PROCESSES = 2
	JOB_QUEUE_INTERVAL = 30
	lines to initinstance.ora file

	This is the output of  "select Job, log_user, priv_user, schema_user,

last_date, last_sec, next_date, next_sec, broken, interval, failures, what from dba_jobs" when the system doesn't work correctly. The 2th entries (after ',' ) are from to the refresh group wich doesn't work.

JOB: 41, 63
LOG_USER: CD1, PRODUC
PRIV_USER: CD1, PRODUC
SCHEMA_USER: CD1, PRODUC

LAST_DATE: 7/6/02 9.54.10	09:54:10, 6/6/02 22.47.57	22:47:57
LAST_SEC: 09:54:10, 22:47:57
NEXT_DATE: 7/6/02 9.59.10, 1/1/00	00:00:00
NEXT_SEC:  09:59:10, 

BROKEN: N,Y
INTERVAL: SYSDATE + 1/(24*12), SYSDATE + 1/96 FAILURES: 0, 16
WHAT: dbms_refresh.refresh('"CD1"."GRUPO_REFRESCO"');, dbms_refresh.refresh('"PRODUC"."GR"');         

        When we force a refresh NEXT_DATE and NEXT_SEC are changed to correct values.

        And this is the output of "Select ROWNER, RNAME, IMPLICIT_DESTROY, PUSH_DEFERRED_RPC, REFRESH_AFTER_ERRORS
FROM DBA_REFRESH" ROWNER: CD1, PRODUC
RNAME: GRUPO_REFRESCO, GR
IMPLICIT_DESTROY: N, N
PUSH_DEFERRED_RPC: Y, Y
REFRESH_AFTER_ERRORS: Y, Y         I've readed almost all group's messages about this issue and I haven't found a solution.         

        Anyone can help me?

Thanks a lot in advance.
Juanfra Received on Tue Jun 18 2002 - 06:32:51 CDT