1000BaseSX Connectivity trouble NEED HELP

[Zaxxon26]

I have 2 Cat2924's tied together with 2 1000SX GBICs.
I have a green light on both ends. When I first setup the switches I thought everything was OK. HPOpenview detected the far end switch but soon after installation it lost connectivity and was going up and down but more down than up. A client tried to place a device on the far end switch but could only ping at a ratio of 1 to 30.

I tested the fiber (62.5 MM). I was within 200ft. but the dB loss was 7.6 and 8.1
All mechanical ends....ick
I Spliced one of the barreled connection which brought the loss down to 4.6 and 5.6 which is within the 7.5 loss allotment for 62.5 MM on the SX GBIC.
BUT I am getting the same results.

I can't ping from the far end switch but I was able to get a DHCP address which was curious. So I turned on sniffer and could see the network from the far end switch but I still can't ping out.....

Near end switch 'sh int' of the gig mod shows:
5 minute input rate 0 bits/sec, 0 packets/sec
5 minute output rate 28000 bits/sec, 28 packets/sec
so nothing is comming from the far end switch, but the lights are still green??

My question is:
Is there a configuration step I am missing?
Everything is VLAN1 so I don't know what it could be.

Or am I off on the dB loss specs?
I am pretty sure it is 7.5 dB worst case for 1000SX over ~160ft of 62.5 MM

This is my first set of 1000SX GBICs but there are many more to come. So if you know of anything I am missing please let me know.

 

[BuckWeet]

What switches are you using? 2924's do not support any kind of gigabit interfaces... Unless its a 2924-M with some additional module that I'm not aware of..

Did you did a "show int gig 0/0" to see for any errors or port status?

BuckWeet

 

[baddos]

Make sure that you have an up to date IOS image installed. I had that problem with a pair of 3524's, and upgraded the IOS and that fixed the problem.

 

[Zaxxon26]

They are c2924M's with x2931 gig mods and 1000SX GBICS.
IOS is 12.0(5)WC3b which is pretty current.

Actually, I got it to work about an hour after my post.
I used to have a problem with giga stacking 2924Ms where if I added a switch to the stack or replaced a x2931 mod users would loose connectivity even though everything came up green. In that situation I would have to reboot the entire stack to get traffic to flow.

I didn't think of it at first because the switches are not geographically in the same stack. But after rebooting far and near end at the same time I got it to work.

It was kind of funky because if I telnet to the near switch I could then telnet into the far switch with no problem but I couldn't telnet directly to the far switch. While in the far switch all I could ping was the near switch and that switch was all that ws in the ARp table. After the syncronized reboot the far end switch could ping everything and the arp table was huge.

So, all is good. R E B O O T. (reload [confirm]) roger that!