Closed GoogleCodeExporter closed 8 years ago
I can't replicate this with a small client population; it deletes multiple
clients fine. This suggests that
the root cause might be a limit in the ADO database layer, which only appears
if the list of clients exceeds
some undocumented internal limit.
Presumably this means I'll need to devise a system for capping the number of
clients processed in one query,
but it would help to be able to zero in on the limit. I don't have access to a
suitable cient population to
experiment on at the moment but what might help me with this is to get a copy
of your database (taken using
the backupdb.wsf script) along with a partial query that you know will provoke
this error. If you can get
this and e-mail me I can arrange a secure upload facility for this.
[ Just FYI, I'll be travelling without internet access for a couple of days
since it's coming up to a holiday
weekend here in New Zealand, so give me a few days. ]
Original comment by nigel.bree@gmail.com
on 28 May 2009 at 10:02
Sounds good, I appreciate your quick response. I have the info ready to upload
to
where ever, so just let me know when you get back and I'll be ready to get it
done.
Original comment by rodneybe...@gmail.com
on 28 May 2009 at 1:08
Thanks to Rodney's help it's become apparent that the underlying cause is when
the
script cleans out data in the WMIData table, which can have 10,000 or so rows
per
machine. The ADO error seems to result from having too many rows being hit in
one
statement as a result of this.
I've posted two updates to try and resolve this, and rodney has tried r10 and
it
appears to resolve the problem for him; change r11 fixes a bug in that which is
actually what made it work, and building on the insight into the root cause
that gave
now fixes the problem with the WMIData table specifically.
Original comment by nigel.bree@gmail.com
on 12 Jun 2009 at 12:01
Original issue reported on code.google.com by
rodneybe...@gmail.com
on 27 May 2009 at 9:21Attachments: