[j-nsp] Link with errors, how to fast detect ?

Alexander Arseniev arseniev at btinternet.com
Tue Oct 20 08:53:33 EDT 2015


If that's CRC errors which cause the OSPF across the link to flap, then 
You could configure a RMON event with appropriate thresholds, the OID is 
in the ifJnxTable:
aarseniev at labrouter> show snmp mib walk ifJnxTable ascii | grep crc

ifJnxInHslCrcErrors.1 = 0
ifJnxInHslCrcErrors.4 = 0
ifJnxInHslCrcErrors.5 = 0
ifJnxInHslCrcErrors.6 = 0
ifJnxInHslCrcErrors.7 = 0
ifJnxInHslCrcErrors.8 = 0
ifJnxInHslCrcErrors.9 = 0
ifJnxInHslCrcErrors.10 = 0
<skip>
http://www.juniper.net/documentation/en_US/junos14.1/topics/concept/rmon-for-monitoring-service-quality-junos-nm.html 

And then add 2 event-policies to proactively disable the LOGICAL unit in 
case of rising alarm, and undisable it in case of falling alarm.
Do not disable the main interface, it will extinguish the laser, and 
Your falling threshold won't work.
HTH
Thanks
Alex

On 19/10/2015 20:02, james list wrote:
> Hello
> Indeed it s not flapping the interface most of the times, just crc errors...
>
> Cheers
> Il 19/Ott/2015 20:49, "serge vautour" <sergervautour at gmail.com> ha scritto:
>
>> Hello,
>>
>> If you're worried about the link flapping, just set a Down-->Up hold-timer
>> directly on the interface.
>>
>>
>> http://www.juniper.net/documentation/en_US/junos15.1/topics/reference/configuration-statement/hold-time-edit-interfaces.html
>>
>> Serge
>>
>> On Mon, Oct 19, 2015 at 11:33 AM, james list <jameslist72 at gmail.com>
>> wrote:
>>
>>> Dear experts,
>>>
>>> I currently have a wan link between two MXs that sometimes is experiencing
>>> errors.
>>>
>>> I’ve on top of the link OSPF (timer 1/3)  as routing protocol and I’d like
>>> to find out any possible way to exclude the link when errors are in place
>>> and below to the 3 seconds of the OSPF timers.
>>>
>>>
>>> I’ve analyzed the following:
>>>
>>> 1)      BFD  with OSPF  but it doesn’t support hold-time (as BGP does), in
>>> order to avoid OSPF flapping and generate higher instability I cannot use
>>> it
>>>
>>> 2)      RPM event based script: probe interval is currently not below of 1
>>> second, hence to detect results with no false positive I need some
>>> sampling
>>> and we’re still at 3 seconds or more
>>>
>>> Any other possible options ?
>>>
>>> OAM ? does it support hold-time ?
>>>
>>>
>>> Cheers
>>>
>>> James
>>> _______________________________________________
>>> juniper-nsp mailing list juniper-nsp at puck.nether.net
>>> https://puck.nether.net/mailman/listinfo/juniper-nsp
>>
>>
> _______________________________________________
> juniper-nsp mailing list juniper-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/juniper-nsp



More information about the juniper-nsp mailing list