Thread Previous • Date Previous • Date Next • Thread Next |
It seems to me the simplest thing is: - when generating an OOPS - do an explain (not analyze) of the top 2 slow queries - log that This isn't perfect, but it should be fairly fast and capture more information that we get today. We can certainly do more things beyond this, but doing this alone would be, I think, useful.
Thread Previous • Date Previous • Date Next • Thread Next |