RE: Materialized view logs
- From: "Goulet, Dick" <DGoulet@xxxxxxxx>
- To: <thump@xxxxxxxxxxxxxxxx>, <Stephen.Lee@xxxxxxxx>
- Date: Wed, 13 Oct 2004 14:44:45 -0400
From Metalink, & there's a whole lot more:
When performing these quereies please be connected as either internal or
the sys user.
select owner, name, master, last_refresh, error
from dba_snapshots;
select log_owner, master, current_snapshot
from dba_snapshot_logs;
When your materailized views are refreshed part of the process is to
purge the logs. Since the logs are not purging it appears that you still
have refrence to snapshots on databases that do not exist any more. We
need to verify this and then take action to preclude this from occuring
again in the future. We must fix this by clearing the missing
snapshots/materialized views so that they are no longer tracked by the
mlog$ table. This table is only used for fast
refreshes of the snapshot/materalized views. It will have no other
impact on performance.=20
Dick Goulet
Senior Oracle DBA
Oracle Certified 8i DBA
-----Original Message-----
From: David [mailto:thump@xxxxxxxxxxxxxxxx]=20
Sent: Wednesday, October 13, 2004 2:39 PM
To: Stephen.Lee@xxxxxxxx
Cc: oracle-l@xxxxxxxxxxxxx
Subject: RE: Materialized view logs
That was the case at one point..we were replicating to one node and have
shifted over replication to a different node.
Logs and MV's have been recreated since then and since noticing this
issue. I think the 8 million row count came from a couple of failed
refreshes, but what concerns me most is the mlog have a higher row count
than the table and it's only an issue on one log...the other row counts
are inline with the table have more rows.
--
..
David
>
> You have to wonder if there is a past replication client that, one=20
> day, was turned off, and nobody told the source database. Whatever it
> is, it looks like some kind of disconnect with the source and one or=20
> more clients.
--
//www.freelists.org/webpage/oracle-l
--
//www.freelists.org/webpage/oracle-l
Other related posts:
