Gentoo Archives: gentoo-cluster

From: "John R. Dunning" <jrd@××××××××.com>
To: gentoo-cluster@l.g.o
Subject: [gentoo-cluster] examples of (large) Gentoo clusters
Date: Mon, 11 Dec 2006 13:39:43
Message-Id: 17789.24425.1900.186472@gs105.sicortex.com
In Reply to: Re: [gentoo-cluster] examples of (large) Gentoo clusters by Bryan Green
1 From: Bryan Green <bgreen@××××××××.gov>
2 Date: Fri, 08 Dec 2006 09:15:54 -0800
3
4 "John R. Dunning" writes:
5 > From: Bryan Green <bgreen@××××××××.gov>
6 >
7 > I was able to patch the kernel, but the server was somewhat unstable.
8 >
9 > Do you remember how it was unstable? That's the kind of thing I'd very much
10 > like to understand, as we're proposing to depend heavily on it. If there are
11 > issues, whether specifically tied to our patches or not, I'd love to know
12 > about them.
13
14 I remember the system was stable until we tried to shut it down.
15
16 Ah. I have observed lustre to get cranky when you try to boot the system out
17 from under it. In particular, if you go to all the servers and /sbin/shutdown
18 without first shutting down lustre, I've seen it hang. Given the nature of
19 lustre, that didn't surprise me a lot :-} The times I've shut down lustre in
20 the correct order (shut down clients, then oss's, then mds, then mgs) it's
21 always behaved itself.
22
23 It would lock up while
24 shutting down, possibly while unmounting filesystems. I also did not get to do extensive
25 testing of the system, so I don't know if it would have been stable under real use of the
26 Lustre filesystem.
27
28 Ok. Like I said, I've found a few bugs, but I've never seen it act unstable
29 in real use.
30
31 > I also tried to make a custom kernel for
32 > lustre 1.4, but ultimately hit too many roadblocks. I did learn a bit about how
33 > to use 'quilt' though.
34 >
35 > Hmmm. Maybe not. Our stuff ditches quilt.
36
37 I just used quilt when working with 1.4. I did not have an ebuild for that.
38
39 I could never get quilt to work so just ditched it. You don't need it anyhow,
40 if you ./configure blah-blah --disable-quilt, it works fine. I imagine if you
41 were doing core development on lustre, in particular trying to actually build
42 the large collection of patches they ship with it, quilt would be handy, but
43 for just trying to get the kernel patched, my scripts skip it.
44 --
45 gentoo-cluster@g.o mailing list