https://rancher.com/ logo
Join Slack
Powered by
This message was deleted.
# harvester
a
This message was deleted.
🐿️ 1
b
image.png,image.png
I'm happy to fill out a github issue, but thought I'd ask before just to make sure I'm not missing something obvious.
Cause let's face it, the chances of me just missing something are much much higher than 0. 😅
b
We have these job in longhorn to keep on top of this I also raised github.com/longhorn/longhorn/issues/10959 for scheduling them or at least trying to check there is enough space before taking one.
🦜 1
b
Yeah in some clusters, the backup cleanup would work, but there's nothing to be "aware" of what kind of snapshot is legitimately created by a user, and what was "generated by a user" but is just cruft left over from a machine backup schedule.
b
hmm maybe this setting in longhorn?
b
Oh maybe.
I also found that.
ah, that's not the link
there we go
b
so enable that and make sure your backups are nicely spaced apart to avoid IO issues?
b
If it works!
You would think the setting above that would trigger it though.
It was already checked
b
You would but I guess harvester counts as on demand rather than being an integrated longhorn schedule
b
Yeah, I'm guessing that it the longhorn schedule and not the harvester one.
3 Views
Open in Slack
PreviousNext
Powered by