Re: Pls Help: Recover snapshot after the refresh group is broken (Thanks and solution)

Hi Dave,

Many thanks for the info. I checked the user_jobs table and found that dbms_refresh.refresh does clear the broken flag on the associated job. But obviously there is sth else that needs to be done for the job to resume. (Platform dependent?) I tried dbms_job.broken and dbms_job.run and both made the job work again.

To solve the problem, I created a new job which is run every hour. It calls a stored procedure to find out all broken jobs and runs dbms_job.broken on them. It seems to work so far. Curiously, running dbms_job.run in the job would cause Oracle internal error. Any idea why?

In our particular application, the customer wants immediate replication on the snapshot site so we can't extend the refresh interval to much longer than 10s. On the other hand, not many users (a couple of) will be updating the master tables simultaneously and we can assume that the db link works when they update. The mentioned problem only occures when, e.g., the master site is down for maintenance.

Thanks again.

Cheers,
Jian Hu Received on Mon Jul 05 1999 - 08:18:50 CDT