> >> Bojan Smojver <bojan@xxxxxxxxxxx> writes:
>
> > Any idea when XFS is going to become an official part of the 2.4 kernel
> > tree?
>
> This is becoming a FAQ. (When reading this, please note I'm not
> involved with XFS beyond being an early adopter, and someone who's got
> a few systems using XFS for some time now)
>
> You have to consider two things: a) 2.4 isn't supposed to get any major
> patches (and it'd be nice if that was true for once); b) there's
> probably still some bug fixing to do on the XFS side. I have had some
> trouble with XFS on one NFS server[1], but other than that, it's rock
> solid. Do notice this is the same lame argumentation presented by
> people who wanted ReiserFS on the kernel, and you know what happened
> with that one. That said, it's very unlikely that XFS will ever go in
> the 2.4 kernel, but it will probably go on the 2.5 at some time.
We will see about that, XFS itself sits on the side, it is the core
kernel changes which have impact elsewhere, and these are a lot smaller
than some of the gyrations 2.4 has gone through so far. Most of Alan's
patches are probably bigger than the whole of XFS.
Steve
>
> HTH,
>
> Marcelo
>
> [1] I'm not 100% sure this was an XFS problem, but other people
> panicked and pulled the plug without proper investigation of the
> problem. The machine was doing a very small ammount of NFS serving
> (15-20 mounts tops scattered among 5-8 clients) out of a 150 GB
> external RAID (IDE) system hooked to a SCSI controller. The kernel
> oops'ed a couple of times, but because of a configuration error on
> my part, I couldn't get a decent call trace and the oops log alone
> was useless. Ever since the RAID was moved to another box
> (XFS/IRIX) I haven't been able to reproduce the problem. I suspect
> the problem was the SCSI controller (which is also gone), but I
> haven't been able to test this...
|