Around 7:30pm every evening we receive the following message and the server experiences an outage of 2 minutes.
A significant part of sql server process memory has been paged out. This may result in a performance degradation. Duration: 1868 seconds. Working set (KB): 2096696, committed (KB): 18715612, memory utilization: 11%.
The message appears in the log 4 times around a 5 minute period.
I have been investigating, and we do not have 'lock pages in memory' activated.
min server mem is 0 (default) and
max server memory set at around 26GB.
Nothing seems that unusual when looking at perfmon counters ie: Available MB, Free Pages, Page Life, Lazy Writes.
Its possible that there is memory pressure as a result of backing up our DBs around this time.
How can I better check what the problem is / rectify the situation?
Edit: Results of
select TOP 10 (sum(pages_kb)*128)/1024 as 'sizein_mb', type as 'clerkttype' from sys.dm_os_memory_clerks group by type order by (sum(pages_kb)*128)/1024 desc 1903846 MEMORYCLERK_SQLBUFFERPOOL 428635 CACHESTORE_SQLCP 148648 CACHESTORE_OBJCP 22371 MEMORYCLERK_SOSNODE 16802 USERSTORE_SCHEMAMGR 6834 OBJECTSTORE_LOCK_MANAGER 5494 CACHESTORE_PHDR 5246 MEMORYCLERK_SQLSTORENG 4513 USERSTORE_DBMETADATA 4168 MEMORYCLERK_SQLGENERAL