Open miohtama opened 2 years ago
HA, We have continuous aggregate refreshes hanging and I wish they were more verbose!
I think they only warrant a notification or action after a certain timeout. Similar to slow query logging. Usually people are not logging every query either 😄
Would it be possible to have it changed to NOTICE and in case of any issue WARN?
ATM refreshing continuous aggregate prints cover 25-75% of my logs coming from TSDB. If CAGGs are working as expected I don't wanna see any logs =)
What type of enhancement is this?
User experience
What subsystems and features will be improved?
Other
What does the enhancement do?
I am extensively using TimescaleDB continuous aggregate jobs.
Currently, my PostgreSQL log is populated with lines like (several lines per minute):
This is because there is a log every time a job runs and I have few jobs running every minute. This logging is not useful - continuous aggregate view refreshing log is only important if it somehow fails.
According to StackOverflow discussion this log level may not be configurable.
I would like to make it possible to tune down this log line, so that logs are not polluted.
Implementation challenges
I am not an expert on PSQL logging subsystem, so I have no idea how this change could be implemented.