07-23-2014 08:15 AM - edited 03-07-2019 08:10 PM
I have two Cisco 2960 and there is a trunk link between them of 100 mbps but the problem i am facing right now is that my trunk link goes down after 30 mins. i have no idea why this is happening after 29 mins of smooth run this things happens and port shows solid orange light. The rest of switch ports connected to end devices works perfect !!
please help me in this regards
Thanks
Solved! Go to Solution.
07-23-2014 08:37 AM
Hello,
I am afraid there is some error condition being detected on the port, and as a result, the switch puts the port into some kind of blocked state.
We definitely need to look into the logging messages because this is an untypical scenario. Please try to inspect the output of show logging for any messages related to the port in question. Also, when the port goes orange, be sure to pull out the following outputs (here, the fa0/1 is the port as an example; please replace with your true port name and number):
show int fa0/1 status
show int status err-disabled
show spanning-tree inconsistent
show spanning-tree int fa0/1
show spanning-tree int fa0/1 detail
These outputs together with the output of show logging should help us narrow the cause of the problem.
Thanks!
Best regards,
Peter
07-24-2014 01:26 PM
Hi Talha,
The switch reports that the link has been brought down because of flapping, i.e. repeatedly going down and back up. If you run the show errdisable flap-values on your switch you should get an output similar to this:
Switch# show errdisable flap-values ErrDisable Reason Flaps Time (sec) ----------------- ------ ---------- pagp-flap 3 30 dtp-flap 3 30 link-flap 5 10
A link is detected as flapping and the port is disabled if it flaps 5 times in 10 seconds. Your link must have been in this condition for the switch to have deactivated it.
I strongly recommend double and triple checking the entire cable run, ideally with a tester, or even replacing the cables. It would also be interesting to see the output of show interface f0/23 and show interface f0/23 counters errors to see if there is any error reported that could point us to the cause of the link flap.
my doubt is going on duplex and speed type. Does the mismatch of duplex and speed type causes the port to hang like this ??
It could potentially cause the link to not come up or to have excessive collisions on the half duplex end and excessive runts (undersized packets), alignment errors and FCS fails on the full duplex end if the duplex was mismatched. However, it seems that there is no mismatch because the switch reports the link as a-full (duplex) and a-100 (speed). Especially the duplex would not be reported as autonegotiated and in the full mode if the autonegotiation did not work. Of course, verify whether both switches report the same duplex and speed but I have a feeling that this is not the cause of the problem. I suspect a problem with the wiring.
Best regards,
Peter
07-23-2014 08:37 AM
Hello,
I am afraid there is some error condition being detected on the port, and as a result, the switch puts the port into some kind of blocked state.
We definitely need to look into the logging messages because this is an untypical scenario. Please try to inspect the output of show logging for any messages related to the port in question. Also, when the port goes orange, be sure to pull out the following outputs (here, the fa0/1 is the port as an example; please replace with your true port name and number):
show int fa0/1 status
show int status err-disabled
show spanning-tree inconsistent
show spanning-tree int fa0/1
show spanning-tree int fa0/1 detail
These outputs together with the output of show logging should help us narrow the cause of the problem.
Thanks!
Best regards,
Peter
07-23-2014 09:52 AM
Thank you so much Peter for you quick reply. i will definitely put the output from the following commands so that we can track down the issue ASAP.
Thank you once again for your expert advice.
Regards,
Talha
07-24-2014 09:11 AM
Hello Peter
well i have some of the output from following commands. firstly i want to show you something about the trunk.
English#show int status err-disabled
Port Name Status Reason Err-disabled Vlans
Fa0/23 err-disabled link-flap
when i used the show int fa0/23 status from the source switch (from where the trunk goes to my second switch) its shows me this output
English#show int fa0/23 status
Port Name Status Vlan Duplex Speed Type
Fa0/23 err-disabled 1 auto auto 10/100BaseTX
after that i shut this port and then i issued no shut command after that it starts working, then i again issued the same command as above
English#show int fa0/23 status
Port Name Status Vlan Duplex Speed Type
Fa0/23 connected trunk a-full a-100 10/100BaseTX
after doing all that i issued show cdp neighbors and i found my second switch in the neighbor table but my doubt is going on duplex and speed type. Does the mismatch of duplex and speed type causes the port to hang like this ??
Wait for your reply
Regards
Talha
07-24-2014 01:26 PM
Hi Talha,
The switch reports that the link has been brought down because of flapping, i.e. repeatedly going down and back up. If you run the show errdisable flap-values on your switch you should get an output similar to this:
Switch# show errdisable flap-values ErrDisable Reason Flaps Time (sec) ----------------- ------ ---------- pagp-flap 3 30 dtp-flap 3 30 link-flap 5 10
A link is detected as flapping and the port is disabled if it flaps 5 times in 10 seconds. Your link must have been in this condition for the switch to have deactivated it.
I strongly recommend double and triple checking the entire cable run, ideally with a tester, or even replacing the cables. It would also be interesting to see the output of show interface f0/23 and show interface f0/23 counters errors to see if there is any error reported that could point us to the cause of the link flap.
my doubt is going on duplex and speed type. Does the mismatch of duplex and speed type causes the port to hang like this ??
It could potentially cause the link to not come up or to have excessive collisions on the half duplex end and excessive runts (undersized packets), alignment errors and FCS fails on the full duplex end if the duplex was mismatched. However, it seems that there is no mismatch because the switch reports the link as a-full (duplex) and a-100 (speed). Especially the duplex would not be reported as autonegotiated and in the full mode if the autonegotiation did not work. Of course, verify whether both switches report the same duplex and speed but I have a feeling that this is not the cause of the problem. I suspect a problem with the wiring.
Best regards,
Peter
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide