CBS350 Port channel dropping randomly

[jasonn]

I have an issue with port channel members randomly going down, and therefore once all members are down the entire port channels will drop of course.
Then suddenly the port channel members will rejoin the port channel and all works until this happens again - anywhere from a random 1 minute to 8 minute interval.

All had been running well for approx 4 weeks, and then the issue suddenly appeared.

Switches are CBS350-24T-4X in a native stack configuration with ports te1/0/1 and te2/0/2 members of the port channel.
Firmware is version 3.2.0.84

The other end of the port channel (3rd party devices) have advised they do not see any logs or errors when this is occurring.  They have shut down the 2nd port on the port channel on their end its been solid ever since.

Log events on the CBS350 switches are as per below: (please ignore the incorrect date/time stamps)

14-Feb-2000 04:18:59 :%STP-W-PORTSTATUS: gi1/0/1: STP status Forwarding
14-Feb-2000 04:18:59 :%STP-W-PORTSTATUS: Po1: STP status Forwarding
14-Feb-2000 04:18:59 :%STP-W-PORTSTATUS: gi1/0/1: STP status Blocking
14-Feb-2000 04:18:57 :%TRUNK-I-PORTADDED: Port te1/0/1 added to Po1
14-Feb-2000 04:18:57 :%LINK-I-Up: Po1
14-Feb-2000 04:18:56 :%TRUNK-I-PORTADDED: Port te2/0/1 added to Po1
14-Feb-2000 04:18:56 :%LINK-W-Down: Po1
14-Feb-2000 04:18:56 :%TRUNK-W-PORTREMOVED: Port te2/0/1 removed from Po1
14-Feb-2000 04:18:56 :%TRUNK-W-PORTREMOVED: Port te1/0/1 removed from Po1
14-Feb-2000 04:18:56 :%TRUNK-I-PORTADDED: Port te2/0/1 added to Po1
14-Feb-2000 04:18:53 :%TRUNK-W-PORTREMOVED: Port te2/0/1 removed from Po1
14-Feb-2000 04:15:53 :%STP-W-PORTSTATUS: gi1/0/1: STP status Forwarding
14-Feb-2000 04:15:53 :%STP-W-PORTSTATUS: Po1: STP status Forwarding
14-Feb-2000 04:15:53 :%STP-W-PORTSTATUS: gi1/0/1: STP status Blocking
14-Feb-2000 04:15:51 :%TRUNK-I-PORTADDED: Port te1/0/1 added to Po1
14-Feb-2000 04:15:51 :%LINK-I-Up: Po1
14-Feb-2000 04:15:51 :%TRUNK-I-PORTADDED: Port te2/0/1 added to Po1
14-Feb-2000 04:15:50 :%LINK-W-Down: Po1
14-Feb-2000 04:15:50 :%TRUNK-W-PORTREMOVED: Port te2/0/1 removed from Po1
14-Feb-2000 04:15:50 :%TRUNK-W-PORTREMOVED: Port te1/0/1 removed from Po1

Configuration of the interfaces is:

interface TenGigabitEthernet1/0/1
lacp timeout short
channel-group 1 mode auto
!
interface TenGigabitEthernet2/0/1
lacp timeout short
channel-group 1 mode auto
!
interface Port-Channel1
flowcontrol on
no negotiation
no spanning-tree portfast
spanning-tree link-type point-to-point
switchport mode trunk

Has anyone seen this happen before or have any suggestions?

[marce1000]

 - Check port counters for the channel member ports , look a the error-indicating figures make sure that those number are idle (e.g.)

 M.

[jasonn]

Hi Marce,

All interface counters on channel members and the port channel itself are showing all good.

FCS Errors: 0
Single Collision Frames: 0
Multiple Collision Frames: 0
SQE Test Errors: 0
Deferred Transmissions: 0
Late Collisions: 0
Excessive Collisions: 0
Carrier Sense Errors: 0
Oversize Packets: 0
Internal MAC Rx Errors: 0
Symbol Errors: 0
Received Pause Frames: 0
Transmitted Pause Frames: 0

[marce1000]

 - If they claim that nothing is wrong at the other end, then  that claim should be validated , by providing the logs related to networking on the respective devices attached to the port channel. You may further more configure and use a syslog server on the CBS350 , then follow-up on logs send to it during normal times too , because sometimes information send can also be indicative for upcoming meltdown (e.g.)

 M.