Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 15 Feb 2007 12:16:39 +1100
From:      "Dimuthu Parussalla" <dparussalla@baysidegrp.com.au>
To:        "'Glen Van Lehn'" <gvanlehn@ccsf.edu>, <freebsd-stable@freebsd.org>
Subject:   RE: Intermittent network issues with Freebsd 6.2
Message-ID:  <007401c7509e$f2d0e620$d801a8c0@dimuthu>
In-Reply-To: <s5d31619.015@bat-gwpo.ccsf.edu>

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

I did that already and it works for few days and dropping out. now I have
change network cards to Intel (em). Hopefully this will solve the issue.

Cheers,
Dimuthu

-----Original Message-----
From: Glen Van Lehn [mailto:gvanlehn@ccsf.edu]
Sent: Thursday, 15 February 2007 9:01 AM
To: dparussalla@baysidegrp.com.au; freebsd-stable@freebsd.org
Subject: Re: Intermittent network issues with Freebsd 6.2


>>> "Dimuthu Parussalla" <dparussalla@baysidegrp.com.au> 02/11/07 3:54 PM
>>>
Hi All,

I am having intermittent network issues with our IBM X236 dual xeon server.
Server comes with two bge network cards. At random intervals internal lan
drops out. And then few minutes later it comes back online.


So far I can only find some interface errors on netstat -i. Can anyone help
me to resolve this issue.

Please refer to attached files and below outputs for more information.


/etc/sysctl.conf

kern.ipc.maxsockbuf=8388608
kern.ipc.somaxconn=2048
net.inet.tcp.sendspace=3217968
net.inet.tcp.recvspace=3217968
net.inet.tcp.rfc1323=1
#net.inet.tcp.rfc3042=0
net.inet.ip.portrange.hilast=65535
net.inet.ip.portrange.hifirst=49152
net.inet.ip.portrange.last=65535
net.inet.ip.portrange.first=1024
net.inet.tcp.inflight.enable=0

/boot/loader.conf
kern.ipc.nmbclusters=32768


vmstat -i
interrupt                          total       rate
irq52: ips0                        75446         17
irq30: ahd0                           81          0
irq31: ahd1                           15          0
irq1: atkbd0                          69          0
irq3: sio1                            74          0
irq4: sio0                            67          0
irq6: fdc0                            90          0
irq14: ata0                           47          0
irq16: bge0 bge1+                1395326        315
irq19: uhci1                          15          0
irq23: ehci0                           1          0
cpu0: timer                      8845847       1998
cpu3: timer                            1          0
cpu1: timer                            2          0
cpu2: timer                      8830658       1995
Total                           19147739       4326


netstat -i

Name    Mtu Network       Address              Ipkts Ierrs    Opkts Oerrs
Coll
bge0   1500 <Link#1>      00:11:25:e9:7f:58   392325   880   449420     0
0
bge0   1500 192.168.1     lan                 342652     -
4175     -     -
bge1   1500 <Link#2>      00:11:25:e9:7f:59   369744  1648   363688     0
0


Regards
Dimuthu Parussalla
==============================

Do you have access to the network switch that your server connects to?  The
stats on the port may show CRC errors or the log may show carrier off/on
transitions indicating that 'Auto-negotiation' isn't stable.  I've seen this
a lot with 100FDx connections and a few times with GbE.  Sometimes it helps
to change the port speed/duplex settings from 'Auto' to <fixed whatever> [or
vice versa ;].    My preference has been to use 'Auto' if it works, and, if
it doesn't, to nail down the server settings, then see whether nailing the
switch port setting or leaving the switch port in 'Auto' is best.

ymmv,
glen van lehn
city college san francisco










Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?007401c7509e$f2d0e620$d801a8c0>