Date: Wed, 12 Dec 2001 11:51:47 -0500 From: Mike Tancsa <mike@sentex.net> To: Ted Sikora <tsikora@ntplx.net>, freebsd-stable@FreeBSD.ORG Subject: Re: dc0 TX underrun Message-ID: <5.1.0.14.0.20011212114951.03ed21e0@marble.sentex.ca> In-Reply-To: <3C178018.9129CD47@ntplx.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Yes, I see this on my 4port DLINK card. The strange thing is that it only
happens when I have it in a 2U box attached to a riser card. If I pop the
card in another box right into the PCI slot, I dont get these under runs,
nor do I get below average performance. After the card goes into this
mode, I see
ruby3# /usr/local/netperf/netperf -l 10 -H 10.2.2.1 -t TCP_STREAM -i 10,2
-I 99,5 -- -m 4096 -s 57344 -S 57344
TCP STREAM TEST to 10.2.2.1 : +/-2.5% @ 99% conf. : histogram
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
57344 57344 4096 10.01 66.64
ruby3#
as opposed to ~ 90 in another box that does not have the riser card.
---Mike
At 11:04 AM 12/12/01 -0500, Ted Sikora wrote:
>Did a make/buildworld on several machines yesterday. All the dc0 cards
>are getting
>the TX errors again ... not often but they appear to be struggling.
>
>Dec 12 10:47:48 dhcp-209-54-72-109 /kernel: dc0: TX underrun --
>increasing TX threshold
>Dec 12 10:47:54 dhcp-209-54-72-109 last message repeated 2 times
>Dec 12 10:47:57 dhcp-209-54-72-109 /kernel: dc0: TX underrun -- using
>store and forward mode
>
>--
>Ted Sikora
>tsikora@ntplx.net
>
>To Unsubscribe: send mail to majordomo@FreeBSD.org
>with "unsubscribe freebsd-stable" in the body of the message
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5.1.0.14.0.20011212114951.03ed21e0>