Viewing 15 posts - 1 through 15 (of 15 total)
Update: Reboot took care of the memory problem. Not the fix I was hoping for offcourse.
I will get back on this topic when plans start to get flushed again.
September 29, 2015 at 6:48 am
The screenshots are taken 1 second apart. The high number of usecounts are on a plan for the classifier function of RG.
This would imply that cache isn't completely cleared, but...
September 23, 2015 at 3:49 am
Hi Gail, I included some images of the plan cache just before the flush, and directly after the flush. Results are grouped on first three columns. As you can see...
September 22, 2015 at 12:13 am
Hi Bill, tried this just to rule it out. Set memory back to 56Gb. Intervals are getting larger now, but still at 2-4 minutes before cache gets cleared.
Bill Talada (9/21/2015)
September 21, 2015 at 2:07 pm
Already tried that, know for sure that no dbcc commands are executed.
nicholas.winstanley (9/21/2015)
September 21, 2015 at 2:01 pm
Jayanth, many thanks for your efford so far. Please tell me what information I have to collect, and how to do it.
What I think is so strange is that this...
September 21, 2015 at 7:17 am
Hi Wilfred, This server option is already disabled.
Wilfred van Dijk (9/21/2015)
is "Optimize for AdHoc workloads" enabled? If so, try disable that option (server option, no restart necessary)
September 21, 2015 at 6:51 am
Database is highly transactional, but it also happens late in the evening when database is hardly used. Interval is steady on 30 seconds when database is hardly used, during peaktimes...
September 18, 2015 at 8:03 am
IS service is running but not actively used as far as I know. AS is not running. RS isn't even installled.
Not using extended stored procedures or sp_oa... calls.
No Linked servers...
September 17, 2015 at 5:31 am
Set it back to 64Gb, now it's taking 60-90 seconds before cache gets cleared again.
Any other hints?
marksanders1976 (9/17/2015)
Just dropped it to 60Gb (with 72Gb available), but no improvement. Still...
September 17, 2015 at 5:03 am
Hi Anthony,
Just dropped it to 60Gb (with 72Gb available), but no improvement. Still getting cleared within 30-60 seconds.
anthony.green (9/17/2015)
September 17, 2015 at 4:54 am
Hi Jayanth,
It's not exactly 30 seconds, can also be 35 or 40 seconds. But it's going on all day long.
I find it very hard to believe that it could...
September 17, 2015 at 4:45 am
dm_os_memory_clerks shows that CACHESTORE_OBJCP is way to small.
On simular servers it's at least 2Gb.
What can I do to make it grow larger? Or how to stop getting cleared?
September 17, 2015 at 4:26 am
I don't get Cache flush messages in error log, I'm seeing the size of cached plans going to zero.
September 17, 2015 at 3:53 am
Total RAM is 72Gb, 64Gb is max memory for SQL.
September 17, 2015 at 3:47 am
Viewing 15 posts - 1 through 15 (of 15 total)