[tulip-bug] driver failure under high NFS load

Greg Wooledge greg@wooledge.org
Sat May 11 18:51:00 2002


--+HP7ph2BbKc20aGI
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

I'm running Linux 2.2.20 on a K6-2 333 MHz (320 MB RAM) with tulip.c:v0.93
(as a module).  My NIC is reported by lspci -v as:

00:08.0 Ethernet controller: Linksys Network Everywhere Fast Ethernet 10/10=
0 model NC100 (rev 11)
        Subsystem: Linksys: Unknown device 0574
        Flags: bus master, medium devsel, latency 64, IRQ 9

I'm loading the module with parameters "debug=3D1 options=3D13".

This machine is both an NFS server (kernel NFS) and NFS client, but it
does a lot more client operations than server.  Sometimes, when I'm
doing a lot of NFS reads and writes (e.g., ripping CDs and encoding
the resulting files to Vorbis on an NFS moutned file system), the NIC
will stop working altogether.  I can work around this by bringing the
interface down, removing the module, re-modprobe'ing, and then bringing
the interface up -- *EXCEPT* that the NFS file system which triggered
the problem (/music) is now completely inaccessible.  I can access other
NFS mounted file systems (/home) but not that one.  Any processes which
were using files on /music (cdparanoia) are frozen and cannot be killed.
The only way I can restore normal functionality is to reboot.

Here's a bit of my kernel's log:

May 10 19:02:12 jekyll kernel: hdb: ATAPI 32X CD-ROM drive, 128kB Cache
May 10 19:02:12 jekyll kernel: Uniform CD-ROM driver Revision: 3.11
May 10 19:02:12 jekyll kernel: VFS: Disk change detected on device ide0(3,6=
4)
May 10 21:27:10 jekyll kernel: eth0: Too much work during an interrupt, csr=
5=3D0xfc69c0d0.
May 10 21:27:10 jekyll kernel: eth0: Restarted Rx at 705859 / 705859.
May 10 21:45:06 jekyll kernel: VFS: Disk change detected on device ide0(3,6=
4)
May 10 22:23:58 jekyll kernel: eth0: Too much work during an interrupt, csr=
5=3D0xfc674050.
May 11 08:56:41 jekyll kernel: VFS: Disk change detected on device ide0(3,6=
4)
May 11 10:56:02 jekyll kernel: VFS: Disk change detected on device ide0(3,6=
4)
May 11 11:16:35 jekyll kernel: eth0: Too much work during an interrupt, csr=
5=3D0xfc674050.
May 11 14:19:27 jekyll kernel: VFS: Disk change detected on device ide0(3,6=
4)
May 11 14:20:56 jekyll kernel: VFS: Disk change detected on device ide0(3,6=
4)
May 11 14:36:57 jekyll kernel: VFS: Disk change detected on device ide0(3,6=
4)
May 11 16:46:25 jekyll kernel: nfs: server dwarf is not responding
May 11 16:46:40 jekyll kernel: nfs: server dwarf still not responding
May 11 16:47:12 jekyll last message repeated 2 times
May 11 16:47:24 jekyll kernel: tulip.c:v0.93 11/7/2001  Written by Donald B=
ecker <becker@scyld.com>
May 11 16:47:24 jekyll kernel: http://www.scyld.com/network/tulip.html
May 11 16:47:24 jekyll kernel: eth0: ADMtek Centaur-P rev 17 at 0xd48c9000,=
 00:20:78:1E:E9:BF, IRQ 9.
May 11 16:47:24 jekyll kernel: eth0: Transceiver selection forced to MII 10=
0baseTx.
May 11 16:47:24 jekyll kernel: eth0:  MII transceiver #1 config 2000 status=
 784d advertising 0081.
May 11 16:47:24 jekyll kernel: eth0:  Advertising 0080 on PHY 1, previously=
 advertising 0081.
May 11 16:47:27 jekyll kernel: nfs: server dwarf OK
May 11 16:47:28 jekyll kernel: nfs: task 1367720 can't get a request slot
May 11 16:47:28 jekyll kernel: nfs: task 1367721 can't get a request slot
May 11 16:47:28 jekyll kernel: nfs: task 1367722 can't get a request slot

The messages at 16:47:24 are from rmmod/modprobe of the tulip driver
module.  The "task NNNNNNN can't get a request slot" will occur hundreds
of times until I reboot.  Stopping and restarting the NFS-related daemons
makes no difference.

I tried searching Google for reports of this problem.  It appears that
<http://www.tux.org/hypermail/linux-tulip-bug/2000-Feb/0001.html> is a
similar problem -- 2 years ago, and no sign of a response to it.

--=20
Greg Wooledge                  |   "Truth belongs to everybody."
greg@wooledge.org              |    - The Red Hot Chili Peppers
http://wooledge.org/~greg/     |

--+HP7ph2BbKc20aGI
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org

iEYEARECAAYFAjzdoDEACgkQkAkqAYpL9t/dOACgxcSSAaAbDU7MWm1Ijilo9ecJ
EwcAn19XCeMMt/1iUt/dMxlsBTEvMBOy
=6R41
-----END PGP SIGNATURE-----

--+HP7ph2BbKc20aGI--