10.1-BETA2 possible kernel memory leak in routing table

[Alexander V. Chernikov]

On 30.09.2014 20:52, Rumen Telbizov wrote:
> Hello,
>
>
>> Can you show route -Arn output ?
>>
> # route -Arn
> route: illegal option -- A
> route: usage: route [-46dnqtv] command [[modifiers] args]
Sorry. That meant to be "route -Arn".
It would be even better if you show something like
"route -Arn ; sleep 10; route -Arn"

>
> Can you disable carp and see if this changes anything?
> Unfortunately no. It is live and serving traffic. I will be rebuilding the
> backup to 10 as well and testing things today. Hopefully it will give me a
> bit more visibility.
Ok. Are there any (relevant) messages floating inside route socket?
e.g does "route -n monitor" shows a lot of output?
>
>
>
>> # Enable better lacp/lagg transmit distribution
>>> net.link.lagg.default_use_flowid=0
>>>
>> Are you sure this is really "better" ?
>>
> Absolutely. I have a lagg (lacp) interface between ix0 and ix1 and without
> this option outgoing traffic (transmits) were being sent only via one of
> the interfaces - ix0. With this option I have proper balancing on the
Very strange. Do you have "normal" ipv4/ipv6 traffic? (e.g. not 
tunneled, not ipsec, not all-fragmented, etc?)
ixgbe marks all packets by received queue id number, and queue 0 is 
always selected for "corner case" traffic.
I see more or less equal traffic distribution in you original post, so 
this looks very strange for me.

Can you show "sysctl dev.ix" output? (and any ixgbe tunables set if any)?
> outgoing traffic.
>
> Thank you,