[evla-sw-discuss] Station board port mirroring

James Robnett jrobnett at nrao.edu
Mon Apr 28 18:25:31 EDT 2014


It's been more difficult than I'd anticipated to catch a failure in just 
the right places but I have a few incidents.

Indications are that the dropped delay models actually leave mccc but 
don't arrive at either s003-t-1 or the mirrored interface where d003-t-3 
used to be.  That implicates the intervening switches, I don't see how 
but the evidence certainly suggests it even if the switches claim otherwise.

Kevin turned on the forwarding of models to rack 8 again so hopefully I 
can catch an event between now and tomorrow that shows it occasionally
drops packets as well.

Jeff has the switch in rack 8 prepped for an IOS upgrade,  I think we 
should plan on upgrading that switch on Wednesday and see what if any 
effect it has on s008-t-1.

James

On 04/16/2014 02:49 PM, James Robnett wrote:
>
> Slight change of plans.  I accidentally pulled the connection for
> d003-t-3 rather than s003-t-3 (both for antenna 12), serendipitously
> Kerry said that he'd rather we used a deformatter connection than a
> station board anyway so I left it as is.
>
> I've set up a port mirror so that any packets destined for s003-t-1 are
> mirrored to d003-t-3's old port which is now connected to the spare
> ethernet jack on widar-boot-2.  I've tested and can see the packets so
> now that I'm back in the building I'll set up a tcpdump on both and
> hopefully catch a missing delay model over night.
>
> james
>



More information about the evla-sw-discuss mailing list