pgBackRest and s3 Panroni and operator

Hi!

  1. Is there a way to immediately deploy the cluster so that backups are made in s3 and archives are saved in s3. Those in the local storage did not get anything so that?

I want to do it as described by crunchy here Crunchy Data PostgreSQL Operator Documentation

pgo create cluster mycluster --pgbackrest-storage-type=s3 -n pgouser1

so that:
Now all archives for the cluster will be pushed to S3 storage only, and local storage will not be utilized for storing archives (nor can local storage be utilized for backups and restores).

The question is how to achieve this?

  1. We set up backups to s3, but what is done by jobs is done sequentially locally and after to s3, and the archives are similarly saved locally and in s3

  2. When you make a switchover by patronictl, a backup is made, but it is only done in the local repository and not in c3, is it somehow configured?

1 Like

Hello @Sergey_Zhuravlev ,

discussed it in private a bit.

Seems like it is not possible to disable local storage completely as pgbackrest still needs a place to store stanza. CrunchyData has the same prob.
We will see what can be done here.

1 Like

Can anyone share the correct config for s3 backup storage? I tried everything to setup s3 minio backups with no avail. I have edited both pgo values and pg-db values, configs are in secret, but it backups only to local storage. Thank you

1 Like

We managed to make a backup on s3 minio and on fs at the same time.

We messed around for a long time with the minio for applications, it worked via http, but here it was necessary to have https only (for pgbackrest).

2 Likes