When making these connections, were the interfaces entering an up status? And negotiating speeds/duplex properly? # show interfaces detail gi1/0/1
While trying to make these connections, I would also take a look at the logs, see if there are any messages that might point you in the right direction. # show logging
The interface counters may also be a good place to look. See if the interface is producing some kind of errors. > show interfaces counters
Are there any firmware or driver updates available to these devices?
What devices do work with these transceivers on the N-series switch? And what are the differences between the devices that don't work and the ones that do? Maybe we can narrow down a specific brand of NIC, or common OS.
This is where it gets whacky. It looks like it may be random. We have two R630's with these SFP's plugged into the i350 Intel Gbit ports. One chassis is behaving, the other is not. Both are/were setup in the same manner (e.g. LACP bonded channels, linux with same kernel) but one had the deafness issue, the other is working fine.
The one I just switched the port on (which we did for the deaf R630) is a Cisco 3945 which I wouldn't expect to have issues.
That said, some of the transceivers are showing that they are Dell qualified, but some are not. I don't have the history (log or terminal scrollback) to confirm if these SFP's were qualified. AFAIK all of the SFP's have been obtained from Dell.
We have a bunch of other devices connected up to the copper SFP's on the stacked 3024f's (15 all up) and they are working fine.
If you move one of the working R630 connections to the port/transceiver of the R630 that was not working, does it continue to work? Or does it stop working?
Not at this point in time. The hardware and ports are in use in both instances so would have to wait for the next occurrence to see what may happen and hope that I have something spare that I can tinker with at the time (and be putting in two devices at the same time).
The serial numbers of the affected SFP's start with PVP & PSS whereas the working SFP's in the 3024's start with PTR or PX5 (PX5's are listed as not qualified). Don't know if this has anything to do with the functionality or not though.
It's not a high priority until the next hit so unless there is something else or someone else experiences similar issues and has more spare hardware to test with I'm going to idle on this.
The idea behind swapping connections around and testing a working server on the suspected transceiver, is to further identify the scope of the issue. Either the transceiver is working with some devices but not others, or the transceiver is not working any devices at all.