em0: watchdog timeout ...
Robert Noland
rnoland at 2hip.net
Wed Nov 14 08:09:01 PST 2007
On Wed, 2007-11-14 at 10:15 -0500, Mike Tancsa wrote:
> At 03:58 PM 10/31/2007, Mike Tancsa wrote:
> >To: John Baldwin <jhb at freebsd.org>, freebsd-current at freebsd.org
> >From: Mike Tancsa <mike at sentex.net>
> >Subject: Re: em0: watchdog timeout ...
> >
> >At 04:21 PM 10/31/2007, John Baldwin wrote:
> >> >
> >> > em1 at pci1:0:1: class=0x020000 card=0x115e8086 chip=0x105e8086
> >> > rev=0x06 hdr=0x00
> >> > vendor = 'Intel Corporation'
> >> > device = 'PRO/1000 PT'
> >> > class = network
> >> > subclass = ethernet
> >> > cap 01[c8] = powerspec 2 supports D0 D3 current D0
> >> > cap 05[d0] = MSI supports 1 message, 64 bit
> >> > cap 10[e0] = PCI-Express 1 endpoint
> >> >
> >> > its also in FastE mode and not gigE. Not sure if that makes a
> >> > difference or not.
> >>
> >>What if you use MSI?
> >
> >How do I enable that ?
>
>
>
> Just to follow up on this thread for the archives, enabling MSI does
> indeed seem to stop, or at least mitigate watchdog timeouts. Its
> been almost 2 weeks on this one box and we have had zero watchdog timeouts.
I'm still getting these with MSI...
em0 at pci0:2:0:0: class=0x020000 card=0x02f61014 chip=0x108c8086 rev=0x03
hdr=0x00
vendor = 'Intel Corporation'
device = '82573E Intel Corporation 82573E Gigabit Ethernet
Controller (Copper)'
class = network
subclass = ethernet
cap 01[c8] = powerspec 2 supports D0 D3 current D0
cap 05[d0] = MSI supports 1 message, 64 bit enabled with 1 message
cap 10[e0] = PCI-Express 1 endpoint
rnoland-ibm% vmstat -i
interrupt total rate
irq1: atkbd0 87456 0
irq4: sio0 2 0
irq6: fdc0 10 0
irq14: ata0 1896659 11
irq15: ata1 422846 2
irq16: uhci3 drm0 10788636 63
irq22: pcm0 15730204 92
irq23: uhci0 ehci0 258850 1
cpu0: timer 338437976 1999
irq256: em0 1363782 8
cpu1: timer 338430141 1999
Total 707416562 4180
FreeBSD rnoland-ibm.acs.internap.com 8.0-CURRENT FreeBSD 8.0-CURRENT
#112: Tue Nov 6 14:11:24 EST 2007
root at rnoland-ibm.acs.internap.com:/usr/obj/usr/src/sys/IBM i386
em0: watchdog timeout -- resetting
em0: link state changed to DOWN
em0: link state changed to UP
em0: watchdog timeout -- resetting
em0: link state changed to DOWN
em0: link state changed to UP
It seems to occur when a new connection is being created, typically from
some fairly heavyweight application. For example starting the acroread
plugin in firefox to download a PDF. I haven't had time to try and
investigate it thoroughly, so this is all anecdotal. But it is still
occurring with MSI.
robert.
>
> # vmstat -i
> interrupt total rate
> irq1: atkbd0 6 0
> irq4: sio0 59020 0
> irq17: em3 225204 0
> irq19: atapci1 243645 0
> cpu0: timer 1985129675 1997
> irq256: em0 1460563078 1470
> irq257: em1 2360550 2
> irq258: em2 1300707554 1309
> cpu1: timer 1985084642 1997
> Total 6734373374 6777
>
> However, it *seems* to increase the amount of missed packets. But,
> traffic patterns are a little higher, so I cant be certain.
>
> em0: Missed Packets = 5514
> em0: Receive No Buffers = 1950
> em0: RX overruns = 9
> em2: Missed Packets = 220073
> em2: Receive No Buffers = 29034
> em2: RX overruns = 1498
>
>
> ---Mike
>
> _______________________________________________
> freebsd-current at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-current
> To unsubscribe, send any mail to "freebsd-current-unsubscribe at freebsd.org"
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: This is a digitally signed message part
Url : http://lists.freebsd.org/pipermail/freebsd-current/attachments/20071114/f6441ba2/attachment.pgp
More information about the freebsd-current
mailing list