I had the same problem with both the cards you mentioned. And actually I
know of another person running Redhat 6.1 who has the same problem. Are u by
any chance running RedHat 6.1 ?. In each case of this phenomena I've seen
the OS was Redhat 6.1. You can "downgrade" to 6.0 ..or even better switch
to FreeBSD :).
Awais Riaz
-----Original Message-----
From: Chris Conn <cconn@abacom.com>
To: squid-users@ircache.net <squid-users@ircache.net>
Date: Monday, May 15, 2000 7:52 PM
Subject: Network stops responding
>Hello,
>
>Myself as well as another squid user have emailed this list in the past
>that after a variable amount of time the Linux 2.2.x box running squid
>would stop responding and the TCP/IP capability of the machine seemed
>hung. We were both using 3COM 3c905B cards, and I therefore changed to
>an Intel ExpressPro 100 to test to see if the 3com linux driver was at
>fault.
>
>However, once more the machine running squid was no longer accessible to
>the network. I this time however got logs from the Intel NIC driver:
>
>May 13 22:16:29 cache kernel: eth0: Transmit timed out: status 0050
>0090 at 155314080/155314094 command 000c0000.
>May 13 22:16:29 cache kernel: eth0: Trying to restart the transmitter...
>
>I had hundreds of these in my /var/log/messages. No errors were logged
>to squid.
>
>I have since changed my switch port to autodetect instead of forcing it
>to 100mbps FD. I hope that the switch is at fault. This problem
>usually does not occur in periods of high load, so I don't believe it is
>a congestion issue.
>
>Any ideas? Squid 2.3.STABLE-2.
>
>Thanks in advance,
>
>Chris
>
Received on Mon May 15 2000 - 14:25:48 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:53:29 MST