Re: [foxboro] Running Manual Cable Test - Automatically?
- From: Barry Heap <barry.heap@xxxxxxxxxxxxxxxxxxxx>
- To: "foxboro@xxxxxxxxxxxxx" <foxboro@xxxxxxxxxxxxx>
- Date: Tue, 5 Feb 2013 20:15:27 +1030
Thanks to all ...
I have a few things to check ... Aim dead bands was not something I had thought
of .... But I suspect that they have not been set correctly ...
So far, we have pretty much run through the HH 'v7 slowdown checklist' - I'm on
my iPhone at the moment, so the number escapes me...
And yes, crossed A/B cables did get us on this site ... Even though we took
enormous care ... But you know in the heat of a diagnostic dilemma, things
happen.
I guess what is really annoying is that despite a due care, the v7.x systems I
have worked on still have the occasional slowdown.
Maybe it's noise on a cable, maybe just gremlins, but for no good reason one
cable must have been marked BAD ... And the Network Fault Detector is incapable
of realizing it really does have two perfectly good cables.
And the fix is so easy - cable test, then we have full speed back.
Maybe we will track if down one day ... Or maybe not ... For now if we could
somehow run the cable test in a script, we could get rid of the nasty side
effect.
Regards,
Barry Heap
On 05/02/2013, at 2:46 PM, Kevin Fitzgerrell <fitzgerrell@xxxxxxxxx> wrote:
> Barry,
> I'll echo Shawn and Brian's V7 experiences - in all the instances I've seen
> where V7 systems experienced slowdowns that were fixed by running the cable
> test (which is every V7 system I've worked with at some point) I've tracked
> the issue down to an underlying fault. Those faults have included crossed
> A/B cables between devices and switches, crossed A/B cables at NCNI's and
> "problem" RCNI's. At one site, crossed cable issues persisted for months
> after our fiber guys swore to me they'd confirmed everything until I gave
> up and personally checked each and every core (finding multiple crossed
> fibers). RCNI problems were harder - I had one where the only V7 part of
> the system was a single AW and RCNI - cables were perfect but kept having
> occasional slowdowns which a cable check would fix. Changing the RCNI
> fixed the problem. After finding that one I used that system to "confirm"
> good RCNIs.
>
> Does this V7 site have any V6.x pieces left or is it all V7?
>
> Have you set up nfd_log and nfd_debug to see what faults the system thinks
> it's getting?
>
> Cheers,
>
> Kevin
>
>
> On Mon, Feb 4, 2013 at 7:29 AM, Barry Heap
> <Barry.Heap@xxxxxxxxxxxxxxxxxxxx>wrote:
>
>> Hi List,
>>
>> We have a V7.x site running with NCNI/RCNI and Enterasys switches as our
>> Nodebus.
>>
>> We have experienced the V7.x slowdown that seems to be caused by Cable
>> faults that may be fleeting or longer term. Either way, the first
>> indication we get is the Operators informing the site engineers that
>> 'the screens are running slow again'.
>>
>> The fix is to run a Manual Nodebus Cable test and the fault clears
>> immediately.
>>
>> My question is:
>>
>> Is it possible to schedule a 'manual' Cable test to be run automatically?
>>
>> I know it may be an odd request, but the V7.x systems just don't recover
>> themselves the way the old V6.x systems do .. and the screen slowdown is
>> causing some grief on site.
>>
>> Oh, a V8.x upgrade is not an option at this time, or the near future...
>>
>> Thanks.
>>
>>
>> Barry Heap
>> Control Direction
>>
>>
>> _______________________________________________________________________
>> This mailing list is neither sponsored nor endorsed by Invensys Process
>> Systems (formerly The Foxboro Company). Use the info you obtain here at
>> your own risks. Read http://www.thecassandraproject.org/disclaimer.html
>>
>> foxboro mailing list: //www.freelists.org/list/foxboro
>> to subscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
>> to unsubscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave
>
>
>
>
> _______________________________________________________________________
> This mailing list is neither sponsored nor endorsed by Invensys Process
> Systems (formerly The Foxboro Company). Use the info you obtain here at
> your own risks. Read http://www.thecassandraproject.org/disclaimer.html
>
> foxboro mailing list: //www.freelists.org/list/foxboro
> to subscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
> to unsubscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave
>
_______________________________________________________________________
This mailing list is neither sponsored nor endorsed by Invensys Process
Systems (formerly The Foxboro Company). Use the info you obtain here at
your own risks. Read http://www.thecassandraproject.org/disclaimer.html
foxboro mailing list: //www.freelists.org/list/foxboro
to subscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
to unsubscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave
Other related posts:
