May 15, 2020 at 3:37 am
Hi,
We recently upgraded our production DB server from 2008R2 standard edition to 2017 Enterprise edition. we have noticed queries that would run for couple sends are taking long time to run and causing timeouts in applications. I have rebuild indexes and updated the statics. I checked the sql server logs to see if there are any IO error messages but I didnt see any. I also made sure the configuration settings were identical on both servers. they both have the identical CPU and Memory.
Please advise me on what I could do next to troubleshoot this better.
Thanks,
May 15, 2020 at 6:45 am
Sounds like you've run into the new cardinality estimator issues that everyone had who upgraded from 2014 onwards.
When you rebuilt your statistics did you rebuild with a fullscan or was it a sample?
You will need to look at the different options available, eg set compatibility level to 110 or less, use the traceflags for legacy cardinality or just go and tune all the queries which are slow.
Theres lots on the net if you search SQL cardinality estimator and the issues people faced.
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply