[c-nsp] SFP-10G-LRM in Nexus 5000

[Crist J. Clark]

On Mon, Sep 14, 2015 at 03:21:27PM -0700, Mike Hale wrote:
> Correct me if I'm wrong, but the LRM SFPs are designed to go over MMF,
> not SMF.  Isn't that going to be a problem?

No. As someone else already pointed out, Cisco says you can run LRM over
SMF for 300m, which is actually longer than they say you can run it over
MMF, 220m. You also don't need mode conditioning over SMF.

> We're using some LRM SFPs between floors to a 5548, and they function
> just fine.

What's on the opposite side of the 5548s? It may be similar to what we're
seeing. It looks like this works fine Nexus to Nexus, but it is flaky
Nexus to Catalyst.

> On Mon, Sep 14, 2015 at 3:15 PM, Crist J. Clark
> <cjc+cisco-nsp at pumpky.net> wrote:
> > On Mon, Sep 14, 2015 at 10:45:58PM +0100, Nick Hilliard wrote:
> >> On 14/09/2015 22:39, Crist J. Clark wrote:
> >> > We are running all of this over SMF.
> >>
> >> why are you using LRM transceivers then?
> >
> > I am not exactly sure of the design decisions behind the choice to use LRM over
> > SMF to orginally connect the distro to the core when the campus was built. But
> > given that it was working, and we already had almost enough spare LRMs on hand
> > to complete the project, It Seemed Like a Good Idea at the Time.
> >
> > We also have OM3 MMF between the buildings so replacing this all with SR is
> > the main plan right now, but whatever the problem is with the LRM modules
> > looks more like software/firmware than a problem with the fiber paths (the links
> > always do eventually come up with good light readings and no errors), and we
> > were just curious if anyone had seen this issue with 5ks and LRM before and
> > found a way to deal with it.
> > --
> > Crist J. Clark
> > _______________________________________________
> > cisco-nsp mailing list  cisco-nsp at puck.nether.net
> > https://puck.nether.net/mailman/listinfo/cisco-nsp
> > archive at http://puck.nether.net/pipermail/cisco-nsp/
> 
> 
> 
> -- 
> 09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0

-- 
Crist J. Clark