Lately we've been having repeated problems with pipelines hanging. Our pipelines are all tap-mysql to target-postgres, pipelinewise variants for both. The general pattern is some database error occurs (e.g. can't insert null into not null column, connection error, invalid timestamp format), target-postgres throws an exception, and then everything stops. I don't actually know if the target ever exits, unfortunately, but meltano remains running (perpetually, waiting for the target to exit, I assume) and has to be killed manually. My questions for the collective brain are: 1. Any suggestions on how to investigate this if/when it happens again? Is there anything I can check to figure out where the problem really is? 2. How likely is it that this is a problem that's been fixed already? We're running meltano 2.8.0 in most environments and 2.10.0 in one of our dev environments.

I don’t know if this is related at all, but our most moody SFTP-server often leaves our pipeline hanging with tap-sftp. I mostly see this when running in Docker, but it happens from the command line on my laptop as well.

@thomas_briggs seen any progress on error hunting? @aaronsteers @ken_payne do you have any insights here?

Unfortunately not @Sven Balnojan, thanks for asking. I haven't been able to reproduce it artificially and it hasn't happened naturally in a few days... which is good except it means I don't have any instances to investigate. 😉

Well, I think getting the environments set up with docker etc. might help to more easily do upgrades etc.