I installed the SQL Server 2005 SP2 update 2 rollup on my 64-bit
server and the performance has tanked!


I installed rollup 3 on some of them, but that did not seem to help.


I thought it was just a linked server performance issue, but my
optimization started running today on one of the "update 2" instances
and so far it's been running about 10 hours longer than it normally
does.


The rollup 3 fixed our stack dumping issues, but we NEED to have this
performance thing fixed!


I saw that MS has come out with update 4 last week - doesn't say
anything about fixing this, though.


Has anyone else experienced this?


I'm not necessarily expecting anyone to have a fix for this, just
want
to know I'm looking in the right place before I call MS.

traceable1 (thhamlin (AT) gmail (DOT) com) writes:
Exactly what performance has dropped? Queries in general? Specific
queries or actions? You mention "my optimization", is that the reindexing
job.

You first say "my server" in singular, and then continued with "some of
them" and talks about "instances". Am I right to conclude that this is
one machine with several instances, and all are slow?

Sorry for asking questions, but there is not really much information to
work from in your post.


--
Erland Sommarskog, SQL Server MVP, esquel (AT) sommarskog (DOT) se

Books Online for SQL Server 2005 at
http://www.microsoft.com/technet/pro...ads/books.mspx
Books Online for SQL Server 2000 at
http://www.microsoft.com/sql/prodinf...ons/books.mspx

sorry for not including enough info.

Yes, we have several servers/instances.

I have 2 4-node clusters, 3 of each box running only 1 instance, the
4th is the spare. These have rollup update 3.
I have 1 4-node cluster, 3 of which are running 1 instance, the 4th is
the space. This cluster is running update 2.

It appears that the performance in the optimization (defrag + stats +
reindex) is slow now on the update 2 cluster, but the update 3 cluster
seems to be optimizing fine.

On the update3 servers, it appears that they are all using views which
point to an instance on one of the other boxes. 5 instances contain
the application data and they all point to the instance on the 6th box
which contains the shared data. So the linked server is slowing
things down.
However, these links have always been there and have just now slowed
to an unworkable speed since the update.

For a temporary band-aid, I am replicating this shared database to the
other instances and that seems to be getting us by.

I sure would like to see a fix for this, though!


Thanks!!


On Oct 25, 4:27 pm, Erland Sommarskog <esq... (AT) sommarskog (DOT) se> wrote:

traceable1 (thhamlin (AT) gmail (DOT) com) writes:
I'm still mainly in the dark. It appears that your main problem is that
the optimisation job has slowed down to a crawl on the update-2 cluster.
But then you start talking about a linked server. I can't see how the
optimisation job can be affected by the presense of a linked server.

Are these servers all part of the same application and are hosting
similar databases? Or are they completely different from each other?


--
Erland Sommarskog, SQL Server MVP, esquel (AT) sommarskog (DOT) se

Books Online for SQL Server 2005 at
http://www.microsoft.com/technet/pro...ads/books.mspx
Books Online for SQL Server 2000 at
http://www.microsoft.com/sql/prodinf...ons/books.mspx