I've been experimenting with a series scripts that takes ZFS snapshots every
minute, eventually destroying the oldest so that only so many remain available
for a given window of time.

This may seem a trivial concern with hard drive sizes being what they are
these days, but after running a "zpool history" I started thinking that a ZFS
create and destroy being performed every minute would add up to a lot over
the course of a few years (current estimate is 62MB/year for the naming scheme
of my snapshots).

>From what I understand, ZFS compresses metadata by default, so this history
probbaly won't take up much space in the grand scheme of things.  However, I
was curious just the same about wether or not there was a way to prune down,
or limit the size of, the zpool history.

Also, is there any substantial performance penalty to having a huge history?

Thanks for any pointers.

-- Geoff
freebsd-questions@freebsd.org mailing list
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"

Reply via email to