You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In the older reactive charm there was an action to create local on disk backups. This was helpful for on-prem deployments without any S3 setup. This workflow to save backups this way is also used in a lot of places.
Is there a reason this was not implemented in the new charms? Can we get it implemented please?
The text was updated successfully, but these errors were encountered:
Dear @nishant-dash , thank you for the enhancement request!
During the last discussion in the Hague, we understood the necessity of the "local SQL dump" for the maintenance reason.
Is t still a requirement? If so, it is smaller scope of work comparing to local backups.
If we are talking about the local backups, there are list of further questions:
do you also need restore capability from the local backups?
do you need a smart logic to check the local free disk space?
should charm use a separate 'Juju storage' for the local backups?
In case of K8s, should we store backup on the persistent volume?
do you need the logical backup or physical one? Should logical be archived?
do you need an extra metadata (e.g. wal segment filename and log sequence number, binlog+possition for MySQL)?
it is a PostgreSQL requirement or the same should be developed for Charmed MySQL? (other Data charms?)
what is the dump rotation strategy? Who should clean the old backups?
In the older reactive charm there was an action to create local on disk backups. This was helpful for on-prem deployments without any S3 setup. This workflow to save backups this way is also used in a lot of places.
Is there a reason this was not implemented in the new charms? Can we get it implemented please?
The text was updated successfully, but these errors were encountered: