Replication was stopped for one subscriber database due to memory issue with following error,
Error messages: There is insufficient system memory to run this query. (Source: MSSQL_REPL, Error number: MSSQL_REPL27453)
It got resolved with following commands in that subscriber database server which flush the distributed query connection cache and all unused cache entries from all cache.
DBCC FREESESSIONCACHE DBCC FREEPROCCACHE DBCC FREESYSTEMCACHE('ALL')
Did you encountered such memory issue? and how did you resolved?
I knew it but its intersting to put it in Replication perspective
Thanks
Just curious to know whether you've read the following which indicates that there may be performance trade-offs when using a MSTVF. Bit difficult to compare as you haven't provided any code :-)
MSTVF Performance
The important question would be - why did your server run out of memory? I have seen production systems (using replication) run for as long as 6 months without the need of an outage as small as 10minutes.
Anyway, running the 3 DBCC statements mentioned would probably create more problems in terms of...guess what? performance!
@Jeff: Great link! Thank-you! I knew about the issues with MSTVFs, but did not know why they were not as good as ITVFs.