Re: v$rman_status problem

alek wrote:
> Hi,
>
> we have in our administration a RAC 10.2.0.2 database and starting from
> yesterday morning we cannot use RMAN anymore. It simply becomes frozen
> after simple commands like: "delete archivelog all;" or "report
> schema;". Tracing the corresponding session we notice that the involved
> SQL-SELECT which hangs it up is:
>
> select round(sum(MBYTES_PROCESSED)) ,round(sum(INPUT_BYTES))
> ,round(sum(OUTPUT_BYTES)) from V$RMAN_STATUS start with (RECID=:b1 and
> STAMP=:b2) connect by prior RECID=parent_recid
>
> The waiting event is: "control file sequential read".
> Querying the v$rman_status view to see how many rows are within it
> reveals the fact that there are 2256 records but we obtained this
> result after 4506 seconds. I have to mention that I've collected
> dictionary and fixed table statistics but this didn't help at all. Does
> anybody know what the cause of this behaviour might be and how can I
> fix it?
>
> Many thanks in advance.

We had this problem recently after we performed a dataguard switchover.  Oracle's repsonse - recreate the controlfile. I was flabbergasted. We even uploaded a binary backup of the controlfile, and a dump of the block in the controlfile on which it was spinning. Their response - recreate the controlfile. I would check with them first, though, as YMMV. Regards,

Steve Received on Mon Jul 10 2006 - 08:01:53 CDT