Solarflare LACP bug?

[aurfalien]

On Apr 17, 2014, at 11:09 AM, hiren panchasara <hiren.panchasara at gmail.com> wrote:

> On Wed, Apr 16, 2014 at 7:01 PM, aurfalien <aurfalien at gmail.com> wrote:
>> Hi,
>> 
>> I’ve a Solarflare SFN5162F dual port 10Gb ethernet adapter.
>> 
>> While the card works fine as individual ports, upon configuring LACP the machine suddenly reboots.
>> 
>> Here are my commands;
>> 
>> ifconfig sfxge0 up
>> ifconfig sfxge1 up
>> ifconfig lagg0 create
>> * ifconfig lagg0 up laggproto lacp laggport sfxge0 laggport sfxge1 10.0.10.99/16
>> 
>> * This is were the system reboots.
>> 
>> I believe this to be a bug, what other info can I supply to help determine if it is or simply user error?
> 
> I do not know much about LACP or solarflare but you can start with
> following and someone with more knowledge in that area will help you:

Hi Hiren and many thanks for the reply.

> - FreeBSD version you are using - uname -a

I’m using both the latest version of FreeBSD 9.2 and 10.

> - ifconfig output

lagg0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 9000
options=c07ab<RXCSUM,TXCSUM,VLAN_MTU,JUMBO_MTU,VLAN_HWCSUM,TSO4,TSO6,LRO,VLAN_HWTSO,LINKSTATE>
capabilities=c07ab<RXCSUM,TXCSUM,VLAN_MTU,JUMBO_MTU,VLAN_HWCSUM,TSO4,TSO6,LRO,VLAN_HWTSO,LINKSTATE>
ether 00:0f:53:08:44:7c
inet 10.0.10.99 netmask 0xffff0000 broadcast 10.0.255.255
nd6 options=9<PERFORMNUD,IFDISABLED>
media: Ethernet autoselect
status: active
supported media:
media autoselect
laggproto lacp lagghash l2,l3,l4
laggport: sfxge1 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>
laggport: sfxge0 flags=1c<ACTIVE,COLLECTING,DISTRIBUTING>

> - Any errors you see in /var/log/messages or dmesg.

No errors in logs as the system suddenly dumps and reboots uncleanly.  However I’ll look to enable crash dumps on these particular systems.

> - turn on any debugging knobs available with your card drivers

I’ll look into enabling that.  I don’t think any one in tech has the luxury of saying “I don’t know” and instead must say “I’ll look into that” :)

I was first made aware of this issue using FreeNAS and figured to test the upstream providers warez so to speak.  The problem persists as I expected and do have there crash dumps, just not FreeBSDs crash dumps.

Something about tripping an assert.  I don’t think FreeNAS can do much about it as the problem lies in the FreeBSD Solarflare driver.  Kind of kicking myself with going bleeding edge on the NIC but the Solarflare sounded soo good on paper.

I would say that if the drivers in FreeBSD were as mature as Intel, then Solarflare for the win for sure.  But its all about the drivers.

- aurf

"Janitorial Services"