Re: if_em goes mum

From: Ian FREISLICH <if_at_hetzner.co.za>
Date: Wed, 20 Sep 2006 14:19:00 +0200
"Poul-Henning Kamp" wrote:
> In message <E1GQ0NC-000C5i-FM_at_hetzner.co.za>, Ian FREISLICH writes:
> >Poul-Henning Kamp wrote:
> >> 
> >> Every few minutes my if_em goes mum, it receives packets but
> >> does not transmit anything.
> >> 
> >> ifconfig em0 down ; sleep .1 ; ifconfig em0 up
> >> 
> >> makes it work again.
> >
> >Is the OACTIVE flag set? And is TSO enabled on the interface with
> >100M media?
> 
> em0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
>         options=38b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWCSUM,TSO4,TSO6>
>         inet 10.221.74.159 netmask 0xffff0000 broadcast 10.221.255.255
>         ether 00:0d:60:cf:26:7a
>         media: Ethernet autoselect (1000baseTX <full-duplex>)
>         status: active
> 
> 
> I've disabled tso and I still see the hang:
> 
> em0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
>         options=8b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWCSUM>
>         inet 10.221.74.159 netmask 0xffff0000 broadcast 10.221.255.255
>         ether 00:0d:60:cf:26:7a
>         media: Ethernet autoselect (1000baseTX <full-duplex>)
>         status: active

Interesting.  At the time of the hang, can you still establish
another connection to the box?  I'm guessing this is similar to
what I'm seeing on recent current.  I think the problem appeared
some time between (UTC+0200):

drwxr-xr-x  2 root  wheel   23552 Sep 19 11:38 kernel
drwxr-xr-x  2 root  wheel   23552 Sep 12 08:10 kernel.old

Some of my connections hang, but others still function and I can
make new connections.

Sorry, that this is just a "me too". 

Ian

--
Ian Freislich
Received on Wed Sep 20 2006 - 10:19:10 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:00 UTC