[j-nsp] Low power warning

[Richard A Steenbergen]

On Tue, Mar 23, 2010 at 11:27:22AM -0400, Chuck Anderson wrote:
> On Tue, Mar 23, 2010 at 03:48:21PM +0100, Wouter van den Bergh wrote:
> > Mar 22 16:14:20  chas[796]:  link 1 SFP receive power low  warning set
> > Mar 22 16:14:40  chas[796]:  link 1 SFP receive power low  warning cleared
> > 
> > Does anyone know how I can link this to the interface these messages come from? It doesn't seem to appear anywhere in the logfiles.
> 
> try:
> 
> show interfaces diagnostics optics ge-x/y/z
> 
> where x/y/z is where you have SFP interfaces, and look for low values 
> in the "Receiver signal average optical power" field.  Unfortunately, 
> you can't run the command without specifying each specific interface, 
> one at a time--oh wait!  That now works in 10.1R1!
> 
> show interfaces diagnostics optics
> 
> shows output for each of the interfaces that has DOM capabilities.
> 
> Not sure when they enhanced that, but it didn't do that in 9.5R2.

Hrm... The lack of ability to do "show interfaces diagnostic optics" and
see all interfaces has been on my bitch-list for the last 3+ years. I
had just given up hope that they were ever going to do anything to fix
it (or support the reverse order "show interface xe-0/0/0 diagnostics
optics" for that matter), so I had stopped even checking... But after 
reading this email I just went back and checked a bunch of boxes, and it 
actually IS working on MX on every version of code we're running (9.4R3 
being the oldest). Guess they slipped it in when we weren't looking.

I also just checked our lab EX4200 for the accuracy of the DOM readings
(currently running 10.1R1), and the news there is... less good. Both of
the interfaces are perfectly good working LR XFPs that work just fine in 
MX:

Physical interface: xe-0/1/0
    Laser bias current                        :  35.138 mA
    Laser output power                        :  0.5540 mW / -2.56 dBm
    Module temperature                        :  32 degrees C / 90 degrees F
    Laser rx power                            :  0.5460 mW / -2.63 dBm
    ...
    Laser output power high alarm             :  Off
    Laser output power low alarm              :  On
    Laser output power high warning           :  Off
    Laser output power low warning            :  On
    ...
    Tx data not ready alarm                   :  Off
    Tx not ready alarm                        :  Off
    Tx laser fault alarm                      :  Off
    Tx CDR loss of lock alarm                 :  On
    Rx not ready alarm                        :  On
    Rx loss of signal alarm                   :  On
    Rx CDR loss of lock alarm                 :  On

Apparently my transmit power of -2.56dBm (perfectly normal and in spec)
is setting off the low power alarm low, and even though I have link and
am passing packets, I have complete loss of signal too. The next one is
even worse:

Physical interface: xe-0/1/1
    Laser bias current                        :  42.375 mA
    Laser output power                        :  0.4670 mW / -3.31 dBm
    Module temperature                        :  37 degrees C / 99 degrees F
    Laser rx power                            :  0.0016 mW / -27.96 dBm

-27.96dBm is absurdly out of spec for LR, there is no possible way this 
link could be up if it was within even 10dB of being that low.

But that is still better than our EX8200 DOM under 10.1R1:

Physical interface: xe-0/0/0
    Laser bias current                        :  0.000 mA
    Laser output power                        :  0.0000
    Module temperature                        :  0 degrees C / 32 degrees F
    Module voltage                            :  0.0000 V
    Receiver signal average optical power     :  0.0000
    etc etc etc all 0's for everything...

-- 
Richard A Steenbergen <ras at e-gerbil.net>       http://www.e-gerbil.net/ras
GPG Key ID: 0xF8B12CBC (7535 7F59 8204 ED1F CC1C 53AF 4C41 5ECA F8B1 2CBC)