@staff is it just me or has the instance been 502'ing a lot for the last few weeks?
@whitequark there have been a few updates/maintenances to the instance the past week or so yeah
@lillian right now for example?
@whitequark ah, no, the last maintenance announcement in the discord was from the 18th.
I did previously have an issue where my browser cached the 502 and it kept showing the 502 popup when reloading notifications, but I assume you are seeing real 502s from the web service
@lillian tons of them yeah, there's at least one interval each day when the site is basically unusable due to it. sometimes (usually) shorter, sometimes for like 10-20 min
@whitequark wondering if this is when we’re running backups
@kouhai @whitequark Fixed. Postgres was periodically running out of memory due to an old excessive shared_buffers in postgresql.auto.conf (which was overriding the one I set in the main conf)

@lina @kouhai @whitequark

@SDF could you please check if we're experiencing similar issues? I'm getting constant 502s still while going through the timelines