Unfortunately, the built-in optics in the PICs does not afford me this luxury.  
Levels are good, though, according to optical power meter readings.  On the end 
getting the errors, I am using an LH PIC back-to-back with an LH SFP on the WDM 
box connected to M20-A.  I am attenuating 20dB on the RX side of the PIC and 
the RX side of the SFP.  I thought this might be the problem early on, so I 
inserted an intermediary switch between the WDM node and the M20-A, using LX-LX 
between the switch and WDM node and LH-LH between the M20 and switch.  The 
errors followed the link coming directly from the WDM node, even on the LX-LX 
connection.  In other words, my short-run LH-LH connection seems to not be the 
problem.

-evt

From: Ricardo Tavares [mailto:[EMAIL PROTECTED]
Sent: Monday, August 25, 2008 2:21 PM
To: Eric Van Tol
Cc: juniper-nsp@puck.nether.net
Subject: Re: [j-nsp] Strange RX issue w/ GE PIC

Have you tried the "show interfaces diagnostics optics" to check the power 
levels?

Regards.


Ricardo.


On Mon, Aug 25, 2008 at 12:05 PM, Eric Van Tol <[EMAIL PROTECTED]<mailto:[EMAIL 
PROTECTED]>> wrote:
Hi all,
I'm experiencing a strange RX issue on a link and I need some more ideas on 
where to look.  Two routers, an M7i and M20, are connected back-to-back, 
sort-of (there's optical gear between them, obviously), over a WDM link.  Ping 
tests work perfectly from one to the other, using various packet sizes.  When 
enabling traffic from the M7i to the M20 by lowering an OSPF metric, the link 
works fine.  When enabling traffic in the opposite direction, M20 to M7i, I 
begin to get massive input errors on the M20 GE PIC.  I see no errors at all on 
the M7i side, ever.  Errors only start to accrue when traffic reaches a certain 
as-yet-undetermined level, when bi-directional traffic is enabled.

Done so far:
 - Cleaned every connector in the path.
 - Replaced both patch cables at either end of the link.
 - Installed intermediary switch to rule out PIC/SFP problems.
 - Failed over to redundant light path to rule out primary path problems.

In what situation would one see input errors accrue on one side, but only when 
bi-directional traffic is enabled?

Thanks,
evt
_______________________________________________
juniper-nsp mailing list 
juniper-nsp@puck.nether.net<mailto:juniper-nsp@puck.nether.net>
https://puck.nether.net/mailman/listinfo/juniper-nsp



--
__________________________________________
Vai Encarar? http://www.vaiencarar.com.br
_______________________________________________
juniper-nsp mailing list juniper-nsp@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp

Reply via email to