Date: Fri, 20 Jun 2008 16:45:51 -0700 From: "Support (Rudy)" <crapsh@monkeybrains.net> To: Jack Vogel <jfvogel@gmail.com> Cc: freebsd-net@freebsd.org Subject: Re: Seeking help understanding my "emX: watchdog timeout" messages Message-ID: <485C412F.5000204@monkeybrains.net> In-Reply-To: <2a41acea0806201355y3b123462wc37280f28a9f4216@mail.gmail.com> References: <20080516185813.H866@logos.sky.od.ua> <2a41acea0805160904g7dcf9f58rf69ca5d0612945cc@mail.gmail.com> <4853055C.2030303@MonkeyBrains.NET> <48535A11.4020003@monkeybrains.net> <48582C29.8030307@monkeybrains.net> <2a41acea0806191055w5e112b8bsa57a8db2b177adbe@mail.gmail.com> <485C0F07.7000408@monkeybrains.net> <2a41acea0806201355y3b123462wc37280f28a9f4216@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Jack Vogel wrote:
> On the 573 get me an eeprom dump: sysctl dev.em.4.debug=2
here are the dumps:
em4:
Jun 20 16:30:01 mango kernel:
Jun 20 16:30:01 mango kernel: Interface EEPROM Dump:
Jun 20 16:30:01 mango kernel: Offset
Jun 20 16:30:01 mango kernel: 0x0000 3000 6748 5014 0d30 f746 00f4 ffff ffff
Jun 20 16:30:01 mango kernel: 0x0010 ffff ffff 026b 108c 15d9 108c 8086 83df
Jun 20 16:30:01 mango kernel: 0x0020 0008 2000 7e14 0048 1000 00d8 0000 2700
Jun 20 16:30:01 mango kernel: 0x0030 6cc9 3150 0722 040b 0984 0000 c000 0706
em2:
Jun 20 16:31:07 mango kernel:
Jun 20 16:31:07 mango kernel: Interface EEPROM Dump:
Jun 20 16:31:07 mango kernel: Offset
Jun 20 16:31:07 mango kernel: 0x0000 1500 7817 7299 0424 ffff 50a2 ffff ffff
Jun 20 16:31:07 mango kernel: 0x0010 d473 1604 a42f 10a4 8086 10a4 8086 b165
Jun 20 16:31:07 mango kernel: 0x0020 0008 10a4 5800 0000 5001 0000 0000 0100
Jun 20 16:31:07 mango kernel: 0x0030 6cf6 37b0 07a6 8403 0783 0000 c303 0602
> If you have having TX timeouts using polling, perhaps your system
> is so busy that its not running the cleanup routine in time, can you
> switch and run the interface in non-polled, in fact if possible using
> MSI??
I don't think the system is all that busy dual core 2.8Gz CPU... CPU mostly idle, and
load is at 0.00. Here is the throughput on my busiest devices (lagg0 is em0 & em1):
dev out in
lagg0 67389 kbps 156781 kbps
em2 54342 kbps 14284 kbps
em4 93068 kbps 22433 kbps
vlan6 22784 kbps 122790 kbps
------------------------
-------------------------------------------------------
little script to run to monitor throughput on devices
-------------------------------------------------------
#!/bin/sh
DEVS="lagg0 em2 em4 vlan6"
# current_rate.sh
# This script print out the current bw on each link.
# Tue Dec 27 17:11:16 PST 2005, rudy
measure_device_traffic () {
# measure bytes of 2 seconds... bultiple by 4 to get bits per 1 second
BITS=`netstat -I $InterfaceToCheck 1 | head -3 | tail -1 `;
BITS_O=`echo $BITS | awk '{printf "%6d", $6 / 1024 * 8 }'`;
BITS_I=`echo $BITS | awk '{printf "%6d", $3 / 1024 * 8 }'`;
DEV_PAD=`echo $InterfaceToCheck | awk '{printf "%5s", $1}'`
}
### Measure traffic, print!
echo " dev out in"
while :
do
for InterfaceToCheck in $DEVS; do
measure_device_traffic
echo "$DEV_PAD $BITS_O kbps $BITS_I kbps"
done
echo "------------------------"
done
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?485C412F.5000204>