Once upon a time, Ranjan Maitra <mlmaitra(a)gmx.com> said:
Thanks, this will be a fairly high uptime machine (not allowed to
call it a server here, because that is central IT's role to have and administer:-),
running lots of jobs at least a large part of the time, but the RAID will be on the /. It
is more to keep the machine going if one of the two / drives fail (and till such time as I
can get and put in a new one).
That's a good target for RAID (I just like to remind people RAID is not
backups, because double drive failures happen, filesystem corruption
happens, somebody deletes the wrong file happens, and so on).
I see, so your recommendation is to go for xfs?
It's what I use, in part because I also run RHEL and CentOS servers,
where XFS is the default and preferred filesystem by Red Hat for a while
now.
I think the pros and cons of XFS vs ext4 probably aren't that
significant in most use. XFS doesn't currently support any kind of
shrink operation (more of an issue if you are using LVM but not LVM thin
pools, and there is some work on adding this). ext4 can also journal
data (doesn't by default by can be enabled), which gives additional
protection (at an additional performance cost). XFS is higher
performance for some uses, but that probably gets into specifics about
your use cases to know if it really is (or if it matters). XFS supports
reflinks while ext4 does not, which again can be useful for certain
things.
If you are familiar and happy with ext4 though, there's no reason to
switch unless you see something in particular that XFS would do better
in your use. ext4 is not going away any time soon, and both ext4 and
XFS are mature and stable filesystems (and both are still getting
development).
--
Chris Adams <linux(a)cmadams.net>