zfs snapshot management

[Bartosz Stec]

W dniu 2011-06-06 04:32, Charles Sprickman pisze:
> Hello all,
>
> I've been using a few different tools to manage zfs snapshots in 
> different
> scenarios. For local use, I've found that Ralf Engelschall's set of
> scripts[1] that tie into the periodic(8) system work fairly well. I do
> not use the amd portion since I am only working with zfs snapshots and I
> don't see a need to actually re-mount the snapshots elsewhere for
> recovery. The only limitation I find with this system is that for use on
> a backups host the lack of a monthly or yearly retention period pretty
> much rules it out. For local "oops" stuff though, it's great.
>
> For hosts acting as backups servers, I've been using Snapfilter[2] and
> some cobbled together stuff that rsyncs a bunch of hosts and tries to
> detect and notify on errors. Snapfilter simply is the zfs snapshot
> "sweeper" that periodically deletes snapshots that are outside the 
> defined
> retention period(s).
>
> Since there seems to be a fair number of serious zfs users here, I was
> hoping for some further suggestions for use in either case. Any input is
> welcome...
>
> Thanks,
>
> Charles
>
> [1] - http://people.freebsd.org/~rse/snapshot/
> [2] - http://www.scottlu.com/Content/Snapfilter.html
I'm using periodic scripts from here: 
http://www.neces.com/blog/technology/integrating-freebsd-zfs-and-periodic-snapshots-and-scrubs 
as simple "set and forget" zfs snapshot management approach.
They are included into ports tree too: sysutils/zfs-periodic/
I found them much more convenient than Ralf S. Engelschall's solution.

Cheers!

-- 
Bartosz Stec