[vortex] 3C905C Hangs
Karyl F. Stein
kstein@xenos.net
Tue, 9 Oct 2001 09:31:00 -0400
Any ideas on what could be going on? Is it a flow control issue? Can I
offer any more debugging information? One thing it looks like I didn't
mention is that these are RedHat 7.1 machines with kernel 2.4.3-12. I'd
hate to have to unload these things after purchasing them all.
Thanks,
Karyl
-----Original Message-----
From: vortex-admin@scyld.com [mailto:vortex-admin@scyld.com]On Behalf Of
Karyl F. Stein
Sent: Friday, September 28, 2001 3:38 PM
To: vortex@scyld.com
Subject: [vortex] 3C905C Hangs
I have a problem similar to the one described here:
http://www.tux.org/hypermail/linux-vortex-bug/2000-Mar/0010.html
I did not see any resolution to this.
I recently put a 3C905C card into a backup server. During a large network
transfer of a remote disk, the network card will "hang". I do not have any
solid figures on when the transfer hangs, but anything under 200M seems to
go through fine.
If I run ping on the backup server to something else on the lan during the
"hang", I get some bursts of valid responses, but mostly "destination
unreachable". To fix the problem, I ifconfig eth0 down; ifconfig eth0 up.
After 20 - 30 seconds, the transfer continues at full speed. An ifconfig
does not show any errors, drops, etc.
The backup server has a 3C905C card, (tried two of them with same results).
All the client machines currently have RTL8139 based cards, (they retain
network connectivity). The switch is a D-Link DSS-8+. I've swapped out
cables.
I don't know what information would be useful, but below the output of
vortex-diag -v. There is no difference in this output and one performed
during a "hang", (except for the sixth set of numbers on the third row of
the transceiver registers, but that seems to change during normal operation
as well). Let me know if other information would be useful. Any help on
getting this card to work would be great. I just purchased a number of
these to replace the RTL8139 cards in the hopes that I would get more speed
and not load up the CPU as much.
Karyl
----- Output of vortex-diag -v -----
vortex-diag.c:v2.05 5/15/2001 Donald Becker (becker@scyld.com)
http://www.scyld.com/diag/index.html
Index #1: Found a 3c905C Tornado 100baseTx adapter at 0xe800.
Indication enable is 06c6, interrupt enable is 06ce.
No interrupt sources are pending.
Transceiver/media interfaces available: 100baseTx 10baseT.
Transceiver type in use: Autonegotiate.
MAC settings: full-duplex.
Station address set to 00:50:da:c2:4f:6a.
Configuration options 0052.
Saved EEPROM settings of a 3Com Vortex/Boomerang:
3Com Node Address 00:50:DA:C2:4F:6A (used as a unique ID only).
OEM Station address 00:50:DA:C2:4F:6A (used as the ethernet address).
Manufacture date (MM/DD/YYYY) 12/12/1999, division H, product BB.
Options: negotiated duplex, link beat required.
Vortex format checksum is incorrect (0026 vs. 10b7).
Cyclone format checksum is incorrect (0x9b vs. 0xb0).
Hurricane format checksum is correct (0xb0 vs. 0xb0).
MII PHY found at address 24, status 782d.
MII PHY 0 at #24 transceiver registers:
3000 782d 0040 6176 05e1 45e1 0003 0000
0000 0000 0000 0000 0000 0000 0000 0000
1000 0300 0000 0000 0000 02bb 0400 0000
003f 8d3e 0f00 ff40 002f 0000 80a0 000b.
Use '-a' or '-aa' to show device registers,
'-e' to show EEPROM contents, -ee for parsed contents,
or '-m' or '-mm' to show MII management registers.
_______________________________________________
vortex mailing list
vortex@scyld.com
http://www.scyld.com/mailman/listinfo/vortex