CentOS - Oct 2008 - tg3 ethernet card drops connection under heavy load

We have an HP DL360 server with dual on-board Tigon3 ethernet cards.
We are using eth0, eth1 is unused at the moment.  Sometimes when the
network interface is under heavy load, for example moving large file
transfers over rsync or NFS, the network interface stops working and
we lose all connection to the server.  The only solution at this point
is to jump on the console and restart the network interface.  I have
not found anything in the log files to indicate what is causing this.
Has anyone else experienced something similar?  Or perhaps you know
how I could troubleshoot this?

On Thu, Oct 09, 2008 at 07:44:41AM -0500, Sean Carolan
wrote:
> We have an HP DL360 server with dual on-board Tigon3 ethernet cards.
...
> how I could troubleshoot this?
why don't you start with the kernel version and architecture?
-> uname -a
-> /var/log/messages relevant lines?
-> /sbin/ifconfig -a
-> ethtool eth0 and ethtol eth1

Tru
-- 
Tru Huynh (mirrors, CentOS-3 i386/x86_64 Package Maintenance)
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xBEFA581B
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL:
<http://lists.centos.org/pipermail/centos/attachments/20081009/2ea1ff14/attachment-0003.sig>

> why don't you start with the kernel version and architecture?
> -> uname -a
This server is running centos 3.9
Linux server.domain.com 2.4.21-57.ELsmp #1 SMP Wed May 7 06:10:55 EDT
2008 i686 i686 i386 GNU/Linux
> -> /var/log/messages relevant lines?
There was nothing out of the ordinary in /var/log/messages.  The
logging just stops after the network card drops offline.  dmesg also
shows nothing out of the ordinary when the driver is loaded.  The
network card works fine until it is under heavy load.
> -> /sbin/ifconfig -a
eth0    Link encap:Ethernet  HWaddr FF:FF:FF:FF:FF:FF
          inet addr:10.100.1.200  Bcast:10.100.255.255  Mask:255.255.0.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:56261 errors:0 dropped:0 overruns:0 frame:0
          TX packets:30199 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:5969478 (5.6 Mb)  TX bytes:3305868 (3.1 Mb)
          Interrupt:26

(MAC address was changed by me)
> -> ethtool eth0 and ethtool eth1
Settings for eth0:
        Supported ports: [ MII ]
        Supported link modes:   10baseT/Half 10baseT/Full
                                100baseT/Half 100baseT/Full
                                1000baseT/Half 1000baseT/Full
        Supports auto-negotiation: Yes
        Advertised link modes:  10baseT/Half 10baseT/Full
                                100baseT/Half 100baseT/Full
                                1000baseT/Half 1000baseT/Full
        Advertised auto-negotiation: Yes
        Speed: 1000Mb/s
        Duplex: Full
        Port: Twisted Pair
        PHYAD: 1
        Transceiver: internal
        Auto-negotiation: on
        Supports Wake-on: g
        Wake-on: d
        Current message level: 0x000000ff (255)
        Link detected: yes

Settings for eth1:
        Supported ports: [ MII ]
        Supported link modes:   10baseT/Half 10baseT/Full
                                100baseT/Half 100baseT/Full
                                1000baseT/Half 1000baseT/Full
        Supports auto-negotiation: Yes
        Advertised link modes:  10baseT/Half 10baseT/Full
                                100baseT/Half 100baseT/Full
                                1000baseT/Half 1000baseT/Full
        Advertised auto-negotiation: Yes
        Speed: Unknown! (0)
        Duplex: Half
        Port: Twisted Pair
        PHYAD: 1
        Transceiver: internal
        Auto-negotiation: on
        Supports Wake-on: g
        Wake-on: d
        Current message level: 0x000000ff (255)
        Link detected: no

Sean Carolan wrote:
> We have an HP DL360 server with dual on-board Tigon3 ethernet cards.
> We are using eth0, eth1 is unused at the moment.  Sometimes when the
> network interface is under heavy load, for example moving large file
> transfers over rsync or NFS, the network interface stops working and
> we lose all connection to the server.  The only solution at this point
> is to jump on the console and restart the network interface.  I have
> not found anything in the log files to indicate what is causing this.
> Has anyone else experienced something similar?  Or perhaps you know
> how I could troubleshoot this?
Upgrade the driver? Back in my RHEL3 days(I noticed you
were running CentOS 3.9), we didn't even bother using the broadcom
NICs and instead installed e1000 on all the systems due to driver
issues with the broadcom chips(this was back in 2003-2005, DL360G2-G3).

Get the latest drivers at www.broadcom.com

nate