This really depends on where the packet is lost in the switch.  Might
just be a bug if the CPU is at 30+%.   Without more information, there
is very little to comment about here.

Regards
KK

On 1 March 2010 22:39, Martin Casado <[email protected]> wrote:
> OOB control doesn't help unfortunately (I forget that HP doesn't support
> inband control).  We need to be able to slice the control channel so that
> control packets are prioritized I think.
>
>> btw this is starting to look like a common issue as it's been reported a
>> few times in the past. I guess out-of-band control or control packet
>> priority would alleviate this
>>
>> On Mon, Mar 1, 2010 at 10:20 PM, Guanyao Huang <[email protected]
>> <mailto:[email protected]>> wrote:
>>
>>    I now use smaller traces and no such problem again. I think it is
>>    because of heavy load. ...
>>    Thanks.
>>
>>    On Mon, Mar 1, 2010 at 10:12 PM, Martin Casado <[email protected]
>>    <mailto:[email protected]>> wrote:
>>    > Actually nevermind, I don't think the HP switch support inband
>>    control so
>>    > you must be using out of band.
>>    >
>>    > Another option is to turn off the echo timer in Nox ...
>>    >
>>    >
>>    >> Sorry, my English is poor, what do you mean by "out of band
>>    control"? :)
>>    >>
>>    >> On Mon, Mar 1, 2010 at 10:07 PM, Martin Casado
>>    <[email protected] <mailto:[email protected]>> wrote:
>>    >>
>>    >>>
>>    >>> I wonder if this is a problem with lost echo requests under
>>    load ...
>>    >>>
>>    >>> Can you set up your environment to use out of band control?
>>    >>>
>>    >>>
>>    >>>>
>>    >>>> My program runs about 15 mins, it processes some traces. Datapath
>>    >>>> leaves at around 12 mins.
>>    >>>> Current CPU load is around 30%, and hit 35% occasionally.
>>    >>>>
>>    >>>> On Mon, Mar 1, 2010 at 9:59 PM, kk yap <[email protected]
>>    <mailto:[email protected]>> wrote:
>>    >>>>
>>    >>>>
>>    >>>>>
>>    >>>>> Just to be blunt.  What is the CPU load of your HP?
>>    >>>>>
>>    >>>>> Regards
>>    >>>>> KK
>>    >>>>>
>>    >>>>> On 1 March 2010 21:42, Guanyao Huang <[email protected]
>>    <mailto:[email protected]>> wrote:
>>    >>>>>
>>    >>>>>
>>    >>>>>>
>>    >>>>>> Which part defines the time interval for this echo request?
>>    I guess
>>    >>>>>> enlarger it will temporarily handle my problem.
>>    >>>>>>
>>    >>>>>> On Mon, Mar 1, 2010 at 9:38 PM, kk yap <[email protected]
>>    <mailto:[email protected]>> wrote:
>>    >>>>>>
>>    >>>>>>
>>    >>>>>>>
>>    >>>>>>> Hi Guanyao,
>>    >>>>>>>
>>    >>>>>>> NOX periodic checks if the switch is still alive by
>>    sending an echo
>>    >>>>>>> request.  Basically, your switch did not reply.  You
>>    should look at
>>    >>>>>>> the tcpdump to see if you can find the echo reply.  If
>>    yes, bug
>>    >>>>>>> people
>>    >>>>>>> in this list.  If no, debug your switch.
>>    >>>>>>>
>>    >>>>>>> Regards
>>    >>>>>>> KK
>>    >>>>>>>
>>    >>>>>>> On 1 March 2010 21:34, Guanyao Huang <[email protected]
>>    <mailto:[email protected]>> wrote:
>>    >>>>>>>
>>    >>>>>>>
>>    >>>>>>>>
>>    >>>>>>>> Hi
>>    >>>>>>>> Sorry to bother others.
>>    >>>>>>>> In my program the datapath leaves network again. This
>>    time it seems
>>    >>>>>>>> a
>>    >>>>>>>> little different:
>>    >>>>>>>>
>>    >>>>>>>> 00341|openflow|WARN:stream: no response to inactivity
>>    probe after 15
>>    >>>>>>>> seconds, disconnecting
>>    >>>>>>>> 00342|nox|WARN:stream: disconnected (Protocol error)
>>    >>>>>>>> Datapath leave, f7:ca:e4:00:01:90
>>    >>>>>>>> 00343|bindings_storage|ERR:NDB error on get_locnames_cb
>>    (ok if
>>    >>>>>>>> transient): Can't find specified row
>>    >>>>>>>> 00344|openflow|WARN:stream: no response to inactivity
>>    probe after 15
>>    >>>>>>>> seconds, disconnecting
>>    >>>>>>>> 00345|nox|WARN:stream: disconnected (Protocol error)
>>    >>>>>>>> Datapath leave, f7:ca:e4:00:01:2c
>>    >>>>>>>> 00346|openflow|WARN:stream: no response to inactivity
>>    probe after 15
>>    >>>>>>>> seconds, disconnecting
>>    >>>>>>>> 00347|nox|WARN:stream: disconnected (Protocol error)
>>    >>>>>>>> Datapath leave, f7:ca:e4:00:00:c8
>>    >>>>>>>>
>>    >>>>>>>>
>>    >>>>>>>> I am wondering what is this 15 second probe.
>>    >>>>>>>>
>>    >>>>>>>> Thanks.
>>    >>>>>>>>
>>    >>>>>>>> _______________________________________________
>>    >>>>>>>> nox-dev mailing list
>>    >>>>>>>> [email protected] <mailto:[email protected]>
>>    >>>>>>>> http://noxrepo.org/mailman/listinfo/nox-dev_noxrepo.org
>>    >>>>>>>>
>>    >>>>>>>>
>>    >>>>>>>>
>>    >>>>
>>    >>>> _______________________________________________
>>    >>>> nox-dev mailing list
>>    >>>> [email protected] <mailto:[email protected]>
>>    >>>> http://noxrepo.org/mailman/listinfo/nox-dev_noxrepo.org
>>    >>>>
>>    >>>>
>>    >>>
>>    >>>
>>    >
>>    >
>>
>>    _______________________________________________
>>    nox-dev mailing list
>>    [email protected] <mailto:[email protected]>
>>    http://noxrepo.org/mailman/listinfo/nox-dev_noxrepo.org
>>
>>
>
>
> _______________________________________________
> nox-dev mailing list
> [email protected]
> http://noxrepo.org/mailman/listinfo/nox-dev_noxrepo.org
>

_______________________________________________
nox-dev mailing list
[email protected]
http://noxrepo.org/mailman/listinfo/nox-dev_noxrepo.org

Reply via email to