Redash memory consumption is too high

[aryanboss]

Deployed: AWS EC2
How: Dcoker-Compose
Ram: 64GB

Its frequently happening that memory usage suddenly steeps high upto max 64GB and EC2 Instance goes unreachable.

How can we prevent this issue?

[justinclift]

Ahhh, that's a bit too generic of a problem report for really figuring out what's wrong.

How often are you experiencing the problem? eg "once every few weeks" vs "once every few hours" (etc)

Generally the thing that causes high memory use is starting up too many worker processes. With current versions of redash, each worker process seems to take about 1/4 of a GB of unshared RAM.

So, if you're starting up too many then it can go badly.

From your description though, it sounds more like memory use is ok for a bit... then is going suddenly very large. That sounds more like a memory leak, or someone's running some kind of uncontained process. Not sure what though. 😦