June 17, 2008 at 2:00 pm
I tried affinitty, but it came back after 3 days.
June 17, 2008 at 2:57 pm
If HT is turned off you still need to check for power saving settings in your server. Sometimes the Frequency is reduced when low activity is detected and it causes this error to appear.
I have not been able to get rid of this on some of my servers though.
* Noel
June 17, 2008 at 3:00 pm
Is this AMD tech problem then?
i also have HP should i ask help there or wait till MS fixes it?
How do i know if its really influencing performance?
June 17, 2008 at 3:03 pm
Paparux (6/17/2008)
Is this AMD tech problem then?i also have HP should i ask help there or wait till MS fixes it?
How do i know if its really influencing performance?
My problems are actually with HP hardware. HP support has not been very helpful.
I have not seen performance issues with this and because the message is not frequent I have "overlooked" it for the time being.
* Noel
July 10, 2008 at 5:38 am
After a reboot the message disapears for a couple of days then it comes back and the longer the system is online the more frequent the messages become.
How can i tell if this is influencing performance?
There is very little information on Ms about this either.
July 11, 2008 at 11:58 am
We have the same (non)issue on 2 active passive clusters. They each have 16 dual core Xeons w/HT turned on. For us this is an 'undocumented' feature of the IBM NUMA architecture that we are utilising.
RegardsRudy KomacsarSenior Database Administrator"Ave Caesar! - Morituri te salutamus."
October 15, 2008 at 1:49 pm
Hi we have active\passive cluster. On the windows event viewer of cluster node I am keep seeing this message. 'The time stamp counter of CPU on scheduler id 3 is not synchronized with other CPUs' I read the some article but didn't make sense. Does any know how much this affect on performance and how to take care this? Thanks
November 3, 2008 at 9:04 am
Check the KB article
November 3, 2008 at 9:14 am
Cheers on the link but it has already been discussed by the OP.
Still have the issue and i´m ignoring it until a performance issue is barked by my dev team.
I would like to migrate the BDs to another non HP server to see if the problems still persist but i dont have any free HW at the moment.
Just to eliminate the problem being associated with HP /AMD software (i have upgradede BIOS(firmware of CPU etc as described in HP article).
Thx for the help anyway!
Anyone have any ideias not yet discussed here?
November 3, 2008 at 9:19 am
We have this issue on an active/passive IBM NUMA cluster ... still looking ... still no noticable performance issues ...
RegardsRudy KomacsarSenior Database Administrator"Ave Caesar! - Morituri te salutamus."
November 7, 2008 at 10:09 am
I am facing the same issue. Could anyone let me know what should be the setting for affinity mask and affinity i/O mask for 8 processors...
January 5, 2009 at 9:40 am
I also have just seen this message on our test server AND a drive failed over the weekend. Hmmm...
We'll swap the drive this morning and I'll post next week as to whether I'm still seeing the message.
Server: HP DL580 G4, Intel XEON MP CPU, 2GHz, 3.71 GB RAM, HT is on
System: MS 2003, Enterprise Edition SP2
DB: SQL Server 2005, SP2 Update 6 (3228)
Alan
Viewing 12 posts - 16 through 26 (of 26 total)
You must be logged in to reply to this topic. Login to reply