10G link between switches does not work.

igano,

Nothing is blatantly standing out to me, so if possible I would suggest trying each of the ports without a port channel. and also looking at the port details, as well as the transceiver statistics.

Let me know what you see and we can go from there. 

Hello Chris,

Thank you for your reply. I will try your suggestion. I'm also thinking to update the firmware so it can be ruled out. However, when I go to download it from the DELL website (S5296-ON _ V3.40.5.1-24 ), I get a pdf file that describes how to update various things. Is the firmware update process different from OS6? I used to tftp the .stk file, copy it and load it. Is this not the case for S5296F?

Thank you 

I think I found it. I downloaded zip file, FW Updater_3.40.5.1-20-S5296F.zip I believe the .bin file I'm interested in is: onie-firmware-x86_64-dellemc_s5200_c3538-r0.3.40.5.1-20.bin

Hi

I have postponed the firmware update for now. Can someone please explain the following to me:

(config)# do show spanning-tree
Interface Designated
-------------------------------------------------------------------------------------------------------
port-channel1 128.2665 128 400 FWD 16 32768 8c47.be63.5c00 128.2665
port-channel2 128.2666 128 400 FWD 16 32768 8c47.be63.5c00 128.2666
port-channel3 128.2667 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2667
port-channel4 128.2668 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2668
port-channel5 128.2669 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2669
port-channel6 128.2670 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2670
port-channel7 128.2671 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2671
port-channel8 128.2672 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2672
port-channel9 128.2673 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2673
port-channel10 128.2674 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2674
port-channel11 128.2675 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2675

..........................................

..........................................

port-channel78 128.2742 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2742
port-channel79 128.2743 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2743
port-channel80 128.2744 128 200000000 BLK 0 32768 8c47.be63.5c00 128.2744

Would that be the reason, the link does not come up? Even if I unplug the SFP+, the STP state is still BLK. as far as I know STP will put a port to BLK state, only if a loop is detected. I don't understand why I see state: BLK on the relative port-channel even when nothing is patched in to it (members of port-channel80).

 Thank you

Igano,

If there is already a forwarding path it isn't going to change the state unless the whole other port channel goes down, so to me it would appear it is doing its job and preventing the loop from happening. Now you should be able to confirm this by showing which vlans are on the port channels.

I can confirm there is no loop for STP to block it. The following can be true for unplugged ports too:  

port-channel80 Disb 128.2757 128 200000000 BLK 0 AUTO No 

In my case the link does not come up. Even if STP was putting the port in BLK state, it would take 15-20 seconds to do so. During this time I should see some sort of traffic but ports just stay down. I think is configuration. I'm still investigating. 

VLT port channel 80 spans out on another switch (a second S5296, call it sw3). Do the member ports on that switch have to have the same config as the member ports on sw1? 

Member ports on sw3 have channel-group 80 in their config, not channel-group 80 mode active, as the ones in sw1. Does it matter?

@igano  wrote: Member ports on sw3 have channel-group 80 in their config, not channel-group 80 mode active, as the ones in sw1. Does it matter?

Hello,

no I don't think that this is an issue, I will ask to Chris if he has other ideas.

Thanks