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
What feature do you want to improve?
We offer a way to to automate backup of superset dashboards. however, the CI job we run will fill up with a backup every day. While handy - this is likely too many files which will never get used.
Describe the improvement you'd like
The CI job should not only make a backup, but also cull older backups from the GH repo. Something like keeping 1/day x 7 days, 1/week x 4 weeks, 1/month x 12 months.
Describe alternatives you've considered
manually cull or do nothing at all
Additional context
we're hoping to have national deployments run this and a more clean list of files in backup dir will be much easier when it comes time to restore.
The text was updated successfully, but these errors were encountered:
@mrjones-plip the CI job can be setup to run on whatever schedule maintainers decide, it doesn't have to be daily. I think automatic deletions are quite dangerous. Manually cull is ok, but that should only require documentation.
Fair point! While think that the old adage "backup is only as good as the restore" is very applicable here, I can agree that some docs on how to cull older backups would suffice!
What feature do you want to improve?
We offer a way to to automate backup of superset dashboards. however, the CI job we run will fill up with a backup every day. While handy - this is likely too many files which will never get used.
Describe the improvement you'd like
The CI job should not only make a backup, but also cull older backups from the GH repo. Something like keeping 1/day x 7 days, 1/week x 4 weeks, 1/month x 12 months.
Describe alternatives you've considered
manually cull or do nothing at all
Additional context
we're hoping to have national deployments run this and a more clean list of files in backup dir will be much easier when it comes time to restore.
The text was updated successfully, but these errors were encountered: