NFS serving seems to be broken in 2.4.22-pre6. I had 2 computers running
2.4.22-pre6 (x86, debian unstable current). Tried to acces them via NFS
(using am-utils actually) from a 3rd computer, IO error. Tried to
mount directly, mount: RPC: timed out. Rebooted one computer to 2.4.18
and NFS started to work.
No more details currently but I can test more thoroughly tomorrow.
--
Meelis Roos e-mail: [email protected]
www: http://www.cs.ut.ee/~mroos/
On Thu, 24 Jul 2003, Meelis Roos wrote:
> NFS serving seems to be broken in 2.4.22-pre6. I had 2 computers running
> 2.4.22-pre6 (x86, debian unstable current). Tried to acces them via NFS
> (using am-utils actually) from a 3rd computer, IO error. Tried to
> mount directly, mount: RPC: timed out. Rebooted one computer to 2.4.18
> and NFS started to work.
>
> No more details currently but I can test more thoroughly tomorrow.
Meelis,
Please report more details.
> > NFS serving seems to be broken in 2.4.22-pre6. I had 2 computers running
> > 2.4.22-pre6 (x86, debian unstable current). Tried to acces them via NFS
> > (using am-utils actually) from a 3rd computer, IO error. Tried to
> > mount directly, mount: RPC: timed out. Rebooted one computer to 2.4.18
> > and NFS started to work.
> >
> > No more details currently but I can test more thoroughly tomorrow.
>
> Meelis,
>
> Please report more details.
Seems to be a debian unstable problem with nfs-kernel-server:
http://bugs.debian.org/201598
--
Meelis Roos e-mail: [email protected]
www: http://www.cs.ut.ee/~mroos/
On Fri, Jul 25, 2003 at 11:30:49PM +0300, Meelis Roos wrote:
> > > NFS serving seems to be broken in 2.4.22-pre6. I had 2 computers running
> > > 2.4.22-pre6 (x86, debian unstable current). Tried to acces them via NFS
> > > (using am-utils actually) from a 3rd computer, IO error. Tried to
> > > mount directly, mount: RPC: timed out. Rebooted one computer to 2.4.18
> > > and NFS started to work.
> > >
> > > No more details currently but I can test more thoroughly tomorrow.
> >
> > Meelis,
> >
> > Please report more details.
>
> Seems to be a debian unstable problem with nfs-kernel-server:
> http://bugs.debian.org/201598
I just shut down 2.4.22-pre7 to test a 2.6-test kernel, and it hung trying
to shut down the kernel nfs server.
And I just checked my version and it is a reported good version, 1.0.3-1.
I only saw it once, and I seem to recall rebooting pre7 before without
trouble, so maybe it is hard to produce. :-/
Mike
On Fri, Jul 25, 2003 at 10:05:33AM -0300, Marcelo Tosatti wrote:
>
>
> On Thu, 24 Jul 2003, Meelis Roos wrote:
>
> > NFS serving seems to be broken in 2.4.22-pre6. I had 2 computers running
> > 2.4.22-pre6 (x86, debian unstable current). Tried to acces them via NFS
> > (using am-utils actually) from a 3rd computer, IO error. Tried to
> > mount directly, mount: RPC: timed out. Rebooted one computer to 2.4.18
> > and NFS started to work.
> >
> > No more details currently but I can test more thoroughly tomorrow.
>
> Meelis,
>
> Please report more details.
I'm also having trouble with NFS, with a 2.4.22-pre6-ac1 server.
I'm booting a diskless workstation (with a 2.4.21-ac1 client), and
it will boot and mount the root filesystem. When I try to mount any
other filesystems, the client reports
mount: can't get address for vlad
and loses the main NFS mount: attempts to access any file on the root
filesystem gives stale NFS file handle errors. The failure to mount
also breaks the server -- I can no longer boot the diskless client.
Restarting the NFS server allows me to boot the client once again.
Like Meelis, I'm also using Debian unstable, and I've tried with
both the 1.0.3-1 and 1.0.5 versions of the nfs-tools on both machines.
Both versions give the same error that I reported above.
Hugo.
--
=== Hugo Mills: hugo@... carfax.org.uk | darksatanic.net | lug.org.uk ===
PGP key: 1C335860 from wwwkeys.eu.pgp.net or http://www.carfax.org.uk
--- "Are you the man who rules the Universe?" "Well, I ---
try not to."
I don't know if this is related or not, but I noticed that on old
redhat-6.2 (glibc-2.1.3) systems, portmap dies instantly under 2.4.21+
kernels. Older 2.4.18 kernels worked fine. It exits with -EBADF. I can get
a complete strace if that's interesting. I tested the portmap binary from
redhat-7.2 (which works under redhat-7.2) in the 6.2 environment, and it
had the same problem, so it's not portmap itself.
-matt c
On Sat, 26 Jul 2003, Hugo Mills wrote:
> On Fri, Jul 25, 2003 at 10:05:33AM -0300, Marcelo Tosatti wrote:
> >
> >
> > On Thu, 24 Jul 2003, Meelis Roos wrote:
> >
> > > NFS serving seems to be broken in 2.4.22-pre6. I had 2 computers running
> > > 2.4.22-pre6 (x86, debian unstable current). Tried to acces them via NFS
> > > (using am-utils actually) from a 3rd computer, IO error. Tried to
> > > mount directly, mount: RPC: timed out. Rebooted one computer to 2.4.18
> > > and NFS started to work.
> > >
> > > No more details currently but I can test more thoroughly tomorrow.
> >
> > Meelis,
> >
> > Please report more details.
>
> I'm also having trouble with NFS, with a 2.4.22-pre6-ac1 server.
>
> I'm booting a diskless workstation (with a 2.4.21-ac1 client), and
> it will boot and mount the root filesystem. When I try to mount any
> other filesystems, the client reports
>
> mount: can't get address for vlad
>
> and loses the main NFS mount: attempts to access any file on the root
> filesystem gives stale NFS file handle errors. The failure to mount
> also breaks the server -- I can no longer boot the diskless client.
> Restarting the NFS server allows me to boot the client once again.
>
> Like Meelis, I'm also using Debian unstable, and I've tried with
> both the 1.0.3-1 and 1.0.5 versions of the nfs-tools on both machines.
> Both versions give the same error that I reported above.
>
> Hugo.
>
>