Nodestore_node size

Greetings, we have an on-premise sentry, version 20.12.1 deployed using helm charts and it has roughly about 40k events per hour and has 20 days retention (using sentry-cleanup cronjob), Our PostgreSQL DB (specifically, nodestore_node table)is growing larger day by day and will soon reach the upper limit.

I went through related posts and it seems like PostgreSQL is not an ideal destination for nodestore and the hosted sentry use Riak for hosting nodestore. Is it possible to use Riak (or something else) instead of PostgreSQL for nodestore storage with on-premise sentry? if yes, could someone please point me to some config docs, I already tried using elastic by following GitHub - hellysmile/sentry_elastic_nodestore: Sentry nodestore Elasticsearch backend and it didn’t work, but that’s a pretty old one and probably might not work in the new sentry setup. Thanks!

There are already multiple threads about nodestore disk usage, please use those.

Thanks. I went through most of them, but unfortunately they don’t seems to speak about leveraging non postgreSQL solutions for nodestore.

Unfortunately there’s no solution at the moment. One user is using S3 as nodestore but I am not entirely sure if that is actually saving cost: https://github.com/kanadaj/sentry-s3-nodestore

2 Likes

Thanks Much… :+1: I will give it a try and use it with our Minio cluster which runs on low cost storage, will keep you posted. Thanks!

1 Like

Update: I checked out the S3 option with on-premise sentry, version 20.12.1 and it didn’t work. but it works with sentry 9.x. so it seems like there are no other solutions for now. Thanks!

This topic was automatically closed 15 days after the last reply. New replies are no longer allowed.