Sql MI
We migrated one of our biggest on prem sql servers to Azure a few months ago. We’ve hit the occasional high CPU flagged by a product owner whose app data lives in their DB that we host and created coding efficiencies, or in some cases we realized it was caused by MS maintenance. But now we are kind of seeing a pattern that everytime there is maintenance, new issues come out of the woodwork.
Perhaps more relevant is that we also recently upgraded from sql 2016 to latest sql version in azure sqlmi. A lot of the code was old. In hindsight, we should have used MS assessment tool to check what code will be impacted. I opened a ticket with Microsoft to investigate change on sqlmi level and also opened a ticket in devops for our DBAs to identify SPs and queries that are high CPU/slow/showing any performance issues at all. Then we will use one of the procs as a test case. I’m also going to set up a task to set up some monitoring/alerts.
Is there something I’m not thinking about that we should be thinking about? How would you handle?