db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mike Matrigali <mikem_...@sbcglobal.net>
Subject Re: Question on why indexStat deamon is being triggered
Date Sun, 01 Apr 2012 00:00:40 GMT
Bergquist, Brett wrote:
> I will log a JIRA for this.
> 
> The SYSCS_UTIL.SYSCS_COMPRESS_TABLE does not fix the problem.  Backing up the database
does not fix the problem either although I did not expect it to.
> 
> The database cannot be taken down as it is being updated continuously 24x7 and is quite
large so export/import or copying to a temporary is not possible.  It is not only one table
as well, as there are about 50 that have this problem.
> 
> So I guess making indexStat work around this going to have to be done.    I am going
to start looking at that now.
> 
I don't have time to look at this more for now.
One possibility would be to make the code path that updates
the statistics on all indexes for the conglomerate to delete
all rows in systatistics first, making it find the orphan
ones also, i think it currently just does it based on existing
indexes and one at a time and thus misses the orphaned one.
There is also a code path that just updates the
statistics for a single index, so that one should not be
changed.  I don't know which one the stat daemon calls.

/mikem


Mime
View raw message