HP 1820-48G & J4858C SPF modules - What I am doing wrong here?

[Lanttu42]

I am pretty newbie with all these fibre optic stuff, but I hope that you have some time to help me out with this. I have easy customer with two HP 1820-48G switches installed. 

Current situation is this:

I have bought two J4858C modules (Both genuine HP) and installed them to replace a CAT6 copper cable between switches.  From the installed SPF modules I have wired an violet LC Multimode OM4 cable from module to panel.

• Floor 1: Half of the ethernet ports are used + SPF module is in port 49
• Floor 1: All ethernet ports are in use + SPF module is in port 49

The problem is, that there is no life at all. The switches wont even recognize those modules while hot swapping. Only indication of working module comes when I power cycle the swith. Then the down arrow on port 49 flashes once but then again -- nothing. 

Questions

1. Should I configure those switches somehow to get these SPF modules to work? Currently both of the switches are in default condition. No VLAN:n, no trunks. 
2. When I move my hand front of the installed SPF module  without cable or shield - should I see the light on my hand? Yes, I know - Stupid newbie question, but I just need to know this. :D
3. How important is the cable types (OM3-OM4) with this configuration?
4. Should the normal ethernet between switches removed before the link works?

[HP-Browniee]

HI

Can you see the red laser light through the sfp (when connected in the switch), you should see a red light.

I looked it up and this SFP module is supported by the switch so that can not be the problem.

Maybe you created a loop, when connecting the switches and maybe the port when in disable state. You should see this in the port details.

Kind regards

[parnassus]

1. No, just plug the SFP Transceiver in and that's all (Configuration on corresponding SFP Transceiver enabled port can be carried consequently at your convenience) <-- What Port Status the Switch reports about your port 49 (did you tried ports 50, 51 and 52 to see if the issue is slot-related?).
2. OK, but that can be dangerous...
3. Well the HP X121 1G SFP LC SX Transceiver (J4858C) accepts, AFAIK, 62.5/125 μm or 50/125 μm (core/cladding) diameters, graded-index, low metal content, multi-mode fiber optic complying with ITU-T G.651 and ISO/IEC 793-2 Type A1b or A1a, respectively...so if your Fiber Optic link end-to-end (so plus the rack optical patch cords on both ends) is all multi-mode 62.5/125 μm or all multi-mode 50/125 μm it should be OK...provided that its length is less than, let's say (to be restrictive at worst), 220 meters (and I think it is since you're connecting two adjacent floors) <-- se QuickSpecs of HPE X121 for other details.
4. Yes, otherwise you create a loop.

Are you two switches running PT.02.01 (latest available)?

[Lanttu42]

Thanks for your help so far!

I will reply to you both with this reply, to keep this thread little bit clearer.

The switches was on release firmware, so I updated those to latest, but that didn't solved this problem. 

Management says "Link down" same way as it says on those empty RJ-45 ports, so there aren't any indication of the SPF module if it is detected at all or not. My quess is, that it is not detected at all because there is no eye-piercing or hand cutting  Dr. Evilish "laser beam" coming out of the module (Yes, the hat is off and I checked laser by hand - not by watching. :D)

I had to leave the office suddenly, so I had no change to try those modules on other ports available. The only noticeable sign of life seems to be that bried down arrow light under the module when switch is restarted. 

At the moment switches are not connected together by those modules. Should I still see some activity light somewhere? 

I really thought that this is easy. :D 

[parnassus]

Would be interesting to see if those genuine SFP Transceivers work out of the box if you connect them directly using just an optical patch cable (Tx<-->Rx and Rx<-->Tx), say 2 or 3 meters long, provided that you can place both switches sufficiently near each other (especially if they aren't already in production)...this way you will eliminate (and, conversely, evidentiate) any optical trunk possible issue...and you will be able to diagnose Switch, SFP Transceivers and/or the optical patch cable used.

Switch Diagnostic should show you something...what switch diagonstic is reporting?

[AlSanawi]

Hi Parnassus.
1st of all I really and truly want to thank you so much for all of your posts and replays. Those are very helpful, as I am very new to the HP Aruba switches, I been all my life CIsco. And recently joined the HPE club. 

2nd, I just last night was cuting over a live network and had the same exact issue. I was replacing a Cisco 55K switch with a new HPe Aruba 2930M everything seem to be working just fine but that SFP is not supported by the switch. the only indecation I saw when I plugged that SFP the down arrow lighted green once for less than a secound. and thats it.

Unfortionatly, I had to roll back to the old CIsco 55k and putted everything back as it was before the cutover to get those people back to life. Because that fiber link was their only wayout to the other side of the network.  So, I had no enough time to collect much of loges or done any types of troubleshooting.

The only thing pop up in the configurations when I show interfaces is "Therd party transceiver". Which is really strange because this is an HPE SFP SX, J4858C, not a 3rd party transceiver. although I'm %100 sure its an HP SFP but I still issued the command "allow-unsupported-transceiver" and that changed nothing.

 I never had that issue before with Cisco products.

My question is, what I am doing wrong here?