[Coral-dev] crl_anf and crl_flow stall after 10 minutes

[James Grace]

Thanks Stephen,

I'll try that out.

James


On Wed, Apr 11, 2012 at 5:46 PM, Stephen Donnelly <
Stephen.Donnelly at endace.com> wrote:

>  Sounds like it could be a problem with either the application or the
> card.****
>
> ** **
>
> Is the card still seeing packets? Run dagconfig -Ci while coral is still
> running and see if the counters are incrementing. Is the card dropping
> packets (e.g. application is not consuming them)? Run dagconfig a few times
> and see if the stream drop counters are incrementing.****
>
> ** **
>
> Try capturing with a different application for a while and confirm that
> the card continues working for more than 10 minutes, e.g. dagbits -d0 -c,
> or tcpdump -v -i dag0 -w /dev/null****
>
> ** **
>
> I’m not that familiar with the specific tools, but the CAIDA folks may be
> able to help. It could be that the packet rate on the 10G link is too much?
> ****
>
> ** **
>
> Regards,****
>
> Stephen.****
>
> ** **
>
> *From:* coral-dev-bounces at caida.org [mailto:coral-dev-bounces at caida.org] *On
> Behalf Of *James Grace
> *Sent:* Thursday, 12 April 2012 1:06 a.m.
> *To:* coral-dev at caida.org
> *Cc:* Julio Ibarra
> *Subject:* [Coral-dev] crl_anf and crl_flow stall after 10 minutes****
>
> ** **
>
> Good Morning,****
>
> ** **
>
> I've encountered a problem when using either crl_flow or crl_anf on a
> Dag8.1SX with DAG 4.2.1 on a 10GigE link.  After about 10 minutes of
> collecting, both applications just stall.  The operating system still sees
> them as active and they are still utilizing the processor, but they are not
> generating and sort of output.  I've tried debug=2 to see if
> any useful information was generated from the application, but I didn't get
> anything.  Has this been an issue for anyone else?****
>
> ** **
>
> ** **
>
> Cheers,****
>
> James Grace****
>
> Network Engineer****
>
> AMPATH / CIARA****
>
> Florida International University ****
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://rommie.caida.org/pipermail/coral-dev/attachments/20120412/4018a5ee/attachment.htm