Hi, we are randomly getting a
503 Service Unavailable error from time to time. It is relative infrequent and only lasts for a few minutes at most. Once when I checked it and the number of Read IOPS was unusually high for the time when it happened. Could it be connected? Or is there some other reason why this happens? Also, what can we do about it?
Thx for help!
Btw we are running Metabase in AWS on a db.t2.micro instance. Could it be that the instance doesn’t manage to handle the number of IOPS?
Seem others have “less than perfect” experiences on there as well: … maybe it needs something a bit beefier?
Is there anybody running metabase on AWS free tier (t2.micro) with reliable MySQL requests? I’m at the point where I need to decide if I will go from designing a full dashboard from scratch or use metabase but the running costs on a t2.small are 80 bucks per month.
So I have two questions:
Does it work with a t2.micro instance?
Can you ask Metabase to shut down when not used or to freeze? How do you reduce running speed?
Btw, Heroku seems free but seems to launch the metabase portal …
Thanks for the response. We already upgraded the instance once to “small”, but we are probably gonna try upgrading a bit more to see if that helps.
Did it help? We still get it randomly!
Which version of Metabase? How are you hosting Metabase?
And do you see any detailed errors in the Metabase log?
Metabase on AWS EBS/RDS
Haven’t checked detailed error log
Also the EBS environment keeps changing to severe to normal quite often, haven’t been able to inspect yet. Any leads would be appreciated
Upgrade to 0.32.8 - there has been several issues with the async web server in the earlier 0.32.x versions.