May 23, 2005 at 1:20 pm
We have a sql server 2000 SP3 that we are getting these errors in Error log. It happened just after Sun morning maint (checkdb, reindex, etc). These errors happened during diskkeeper defrag of all HDs. We do not disable sql for defrags since all users are removed before maint runs.
Diskkeeper was scheduled and running at the same time? The
applications that use this server do not appear out of the ordinary speed wise. We have not gottent these errors before with diskkeeper running during the same times.
Error: 17883, Severity: 1, State: 0
The Scheduler 0 appears to be hung. SPID 6, ECID 0, UMS Context
0x03CE3440
I found this error which was added in Sp3 but not the same UMS context? I will reboot ASAP since some KB articles say it may slow down server.
The other thing when I run this query from KB 890637 it does NOT match the spid in the error (Error has SPID 6) but my log writer is SPID 2 (as it should be)?
SELECT spid FROM master.dbo.sysprocesses WHERE cmd ='LOG WRITER'
SPID=2 (not 6 as in my error)
I am working through KB 319892
Any ideas on causes? Is it a problem.
May 24, 2005 at 8:54 am
Hi pneth,
For the scenario you have given this mostly to be expected.
17883 errors are reporting that a thread executing in the Sql process space did not release the CPU for more than 60 seconds. Schedule 0 tells you it was on CPU 0, spid 6 may be the executing process or it may be the one that was effected by what ever had the CPU.
A User Mode Scheduler is a thread that keeps track of the other threads executing on a particular CPU. There is one per CPU on any windows system.
Rather than use the query as written in the article you might want to re-write it to tell you what spid 6 is doing. Something like
SELECT cmd FROM master.dbo.sysprocesses WHERE spid = 6
I expect it will be the Task Manager. If you can positively associate the time and duration of the 17883 errors with the OS defrag you are running you probably don't have anything to worry about.
However, if it persists without the defrag you may want to open a support incident with MS. Alternatively, if you aren't using AWE memory on your system you could apply SP4 for Sql Server. It provides enhanced error messages for this type of error.
Hope this answers your questions.
Richard L. Dawson
Microsoft Sql Server DBA/Data Architect
I can like a person. People are arrogant, ignorant, idiotic, irritating and mostly just plain annoying.
May 24, 2005 at 9:13 am
Richard,
spid 6 is "TASK MANAGER". I suspect some conflict between diskeeper and sql. We have diskkeeper set to run after db maint jobs runs every weekend. I think it has improved our end of month processing load GL posting (heavy writes). It is the first time we have seen this error that i know of. I may switch to halting sql, running defrag and restart sql instead.
Thanks I will keep an eye out for errors during non-diskkeeper hours
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply