[j-nsp] Spanning tree RJ45 SFP on QFX5100

[Michael Loftis]

On Mon, Oct 20, 2014 at 7:41 AM, Richard Hartmann
<richih.mailinglist at gmail.com> wrote:
> Dear all,
>
> we are not done debugging yet, but as of right now, we are having a
> rather strange effect...
>
>
> Our setup looks as follows:
>
> QFX5100 = CStest
> sw1 = 2960g
> sw2 = 2960g
> sw3 = EX3300
> sw4 = 2960g
> sw5 = 2960S
>
> CStest <RJ45 1G> sw1
> CStest <SM 1G> sw2
> CStest <SM 10G> sw3
> CStest <RJ45 1G> sw4
> CStest <SM 1G> sw5
>
> The effect we're seeing is that Spanning Tree works just fine over the
> Single Mode links, but not across the RJ45 SFPs.
>
> Weirdly enough, LLDP and the like are being transmitted just fine.
>
> Yes, we tried all available JunOS releases, even a few beta images.
> Yes, we also tried 14.1X53-D10.
>
> I am kinda stumped as to how this is even possible. While we all hate
> RF45 SFP for various reasons, this is totally new and unexpected.
>

Might be interesting to know whose RJ45 SFP's?  If third party I'd
personally really tend to suspect/blame the third party...might be a
coding issue, might be the onboard hardware doing something silly.
I'd try a different vendor SFP.  It might be something Juniper is
making worse and is still willing to fix even if it's not necessarily
their problem - like there've been various i2c reading issues they've
dramatically improved behavior for.

>
> Yes, we're still having fun,
> Richard
> _______________________________________________
> juniper-nsp mailing list juniper-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/juniper-nsp



-- 

"Genius might be described as a supreme capacity for getting its possessors
into trouble of all kinds."
-- Samuel Butler