That one is next on the list now as of this morning.

The Virtual guys are now discussing Cisco firmware updates... apparently (in the environment where nothing changed on 8/3), we...

• This reply was modified 5 years ago by  jchapman.
• This reply was modified 5 years ago by  jchapman.

Thank you for the suggestions Frederico.  We've done ALL of those already.

We just stood up a new server, pristine install on everything, same results.

I just upgraded the test environment to...

Here is the base fact set as of right now...maybe someone else will encounter this some day...

VMWare 6.7U3 HW level 15.  Suddenly last week... a well-behaved server with well known...

• This reply was modified 5 years ago by  jchapman.

Because at the moment... it's a baseline.  And the scenario is more than just 1 query... there are hundreds of otherwise (previously) well behaved queries that now either run for...

No Joy...

Back to forcing maxdop @ 1 for our sample query gives data in 47 seconds... maxdop 2, the query shows running, but there is NOTHING happening on activity monitor. ...

Grant... despite the pre-requisite "nothing has changed" answer from everyone involved in this environment...

I was discovered this morning that VMWare 6.7u3 was applied (upgrading from VMware 6.7u2) 5 days before...

Thanks for the answers guys; I had wondered about the row threshold, but was not sure if that could even be a possibility.

I did roll maxdop down to 1 at...

That's what my wall of shame dashboard is for. The thing that makes it difficult (in this scenario, however)...

Jeff, your argument is exactly why I haven't put the Kill statement in the middle of my foreign process detective code (that records all open connections, statements, run time, user...

Clever idea... I had not thought of that. The big question would be if that change kicked in immediately (changing the behavior of existing open connections, and possibly running...

Ah... an idealist in the group. 😉

You are exactly right.. they don't understand the data model, and to the institutions discredit... haven't spent enough time (and money) on training...

This sounds familiar to ETL performance gremlins I've been chasing lately. If we assume that the procedure and database are capable of better performance, I would assume the problem...

I will start reading up on indexes with partitions. On the brighter side... I have convinced the DBA's here to start working up a plan to migrate to '08.

Thank...

I am not 100% certain this...

RID Index: Yes my all in one partition is...

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

|--Nested Loops(Inner Join, OUTER REFERENCES:([Bmk1000], [Expr1007]) WITH UNORDERED PREFETCH) ...