Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 27 Mar 2001 15:15:26 -0500
From:      "Deepak Jain" <deepak@ai.net>
To:        "Ted Mittelstaedt" <tedm@toybox.placo.com>, <freebsd-questions@FreeBSD.ORG>
Subject:   RE: Network lockups on fxp0?
Message-ID:  <MLEFIJIDEIDHLFKGNEILOEDFCFAA.deepak@ai.net>
In-Reply-To: <000101c0b681$206fad00$1401a8c0@tedm.placo.com>

next in thread | previous in thread | raw e-mail | index | archive | help

They are all on Cisco switches, each server is on a different port, and the
servers are on different switches in different cities.

The port graphs show no unusual traffic patterns in or out of the box. The
mbuf settings are pretty generous:

$ netstat -m
13033/16624/67520 mbufs in use (current/peak/max):
        2315 mbufs allocated to data
        10718 mbufs allocated to packet headers
2232/3184/16880 mbuf clusters in use (current/peak/max)
10524 Kbytes allocated to network (20% of mb_map in use)
0 requests for memory denied
0 requests for memory delayed
0 calls to protocol drain routines

Thanks for the advice,

Deepak Jain
AiNET

-----Original Message-----
From: owner-freebsd-questions@FreeBSD.ORG
[mailto:owner-freebsd-questions@FreeBSD.ORG]On Behalf Of Ted
Mittelstaedt
Sent: Tuesday, March 27, 2001 12:45 AM
To: deepak@ai.net; freebsd-questions@FreeBSD.ORG
Subject: RE: Network lockups on fxp0?


I'd start by looking at what is most common - your network
hubs.  You may have a failed hub that's trashing packets.
Otherwise maybe someone is hitting you with a DoS
attack?

Ted Mittelstaedt                      tedm@toybox.placo.com
Author of:          The FreeBSD Corporate Networker's Guide
Book website:         http://www.freebsd-corp-net-guide.com


>-----Original Message-----
>From: owner-freebsd-questions@FreeBSD.ORG
>[mailto:owner-freebsd-questions@FreeBSD.ORG]On Behalf Of Deepak Jain
>Sent: Monday, March 26, 2001 5:41 PM
>To: freebsd-questions@FreeBSD.ORG
>Subject: Network lockups on fxp0?
>
>
>
>This weekend we started seeing a bunch of new and stable machines (all
>4.2-RELEASE) with varying levels of traffic (3mb/s to 15mb/s) withdrawing
>their MAC addresses from the network layer:
>
>fxp0: SCA timeout
>fxp0: DMA timeout
>(repeating)
>
>or:
>
>fxp0: SCB timeout
>fxp0: DMA timeout
>(repeating)
>
>Reboots clear it, but the systems are completely responsive at the console.
>The strangest thing is that of 10 machines that showed this over the
>weekend, machines would lock up in pairs and singles. This, even though the
>users on the servers were completely unrelated. 5 locked up in one data
>center another 5 in a different one, no other servers anywhere had any
>issues.
>
>These systems are all Tyan Thunder motherboards, with dual integrated sym
>controllers (SCSI) and dual integrated fast ethernet (fxp) ports. 1GB RAM,
>single and dual hard drives. The kernel's buffers and things are known good
>well over 50mb/s and the kernel is identical across all of them as well as
>400 other 4.2 machines. No single server has gone down twice yet, so I have
>no idea of how long between occurrences. Occurrences don't seem to be
>related to traffic flows or system load.
>
>Any ideas of where to track down these issues?
>
>Thanks,
>
>
>Deepak Jain
>AiNET
>
>
>To Unsubscribe: send mail to majordomo@FreeBSD.org
>with "unsubscribe freebsd-questions" in the body of the message
>


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?MLEFIJIDEIDHLFKGNEILOEDFCFAA.deepak>