Virtual chassis flapping | CRC errors on VC Port | VC adjacency lost
When troubleshooting a virtual chassis adjacency problem, you should always check the vc-port statistics to make sure this is not a Cyclic redundancy check (CRC) or framing errors issue on the virtual chassis port. When a virtual chassis adjacency issue occurs, you might observe syslog error messages like the following:
Aug 24 08:32:19 SWITCH vccpd[10591]: VCCPD_PROTOCOL_ADJDOWN: Lost adjacency to 4c6d.5876.068e on vcp-255/1/0.32768, reason: Interface down
Aug 24 08:32:19 SWITCH vccpd[10591]: Member 0, interface vcp-255/1/0.32768 went down
Aug 24 08:32:19 SWITCH kernel: DEVRT_TRUNK_TABLE_CHANGE: Trunk 5: 0/55(129,0) 0/54(128,0) ->
Aug 24 08:32:19 SWITCH chassisd[10585]: self ID 0 self mode 0 master ch 0 master de 0 back ch 0 back de 0 memb ch 0
Aug 24 08:32:19 SWITCH chassisd[10585]: CHASSISD_VCHASSIS_MEMBER_UPDATE_NOTICE: Membership update: Member 0->0, Mode Master->Master, 0M 1B, Master Unchanged, Members Unchanged
Aug 24 08:32:19 SWITCH jlaunchd: Registered PID 21846(disk-monitoring): exec_command
Aug 24 08:32:19 SWITCH jlaunchd: Registered PID 21847((null)): exec_command
These error messages are a report of a temporal lost of the virtual chassis adjacency or flap. One of the reason can be the existance of CRC alignment errors on one of the VC Ports.
To check for CRC alignment errors you can issue the command "show virtual-chassis vc-port statistics extensive", and you will see an output like this one:
root@switch> show virtual-chassis vc-port statistics extensive
fpc0:
--------------------------------------------------------------------------
RX TX
Port: vcp-255/1/0
Total octets: 485051116209 2530448696063
Total packets: 1725925753 2619383909
Unicast packets: 1583164007 2297860243
Broadcast packets: 33044217 318246310
Multicast packets: 109717529 3277356
MAC control frames: 0 0
CRC alignment errors: 550 <=================== Here you can see 550 CRC errors were reported
Oversize packets: 0
Undersize packets: 0
Jabber packets: 0
Fragments received: 15
Ifout errors: 0
64 octets frames: 48644312
65-127 octets frames: 1882514445
128-255 octets frames: 415895539
256-511 octets frames: 148417311
512-1023 octets frames: 148417311
1024-1518 octets frames: 124202328
Rate packets per second: 630 845
Rate bytes per second: 140971 496516
A (CRC) cyclic redundancy check is an error-detecting code used in digital networks to detect accidental changes to digital data. These errors can be caused due to a physical issue, regularly a cable issue, far end system physical issue or even a port or PIC issue. When CRC errors exist, you must perform a physical check of the ports and perform layer 1 troubleshooting to determine the errors in order to clear them.
1. Ensure the cable, smart jack, patch panel, port is well/tightly connected.
2. Ensure the line is clean enough for transmission requirements. Shield the cable if necessary.
3. Check the cable integrity by looking for breaks or other physical abnormalities in the cable.
4. Ensure the pinouts are set correctly. Replace the cable if necessary.
5. If all above steps were complete appropriately, try connecting the same cable to another VC port on the switch and check if the errors continue even using other cable, that means a physical problem with the port itself.
You can also clear the vc port statistics using the command "clear virtual-chassis vc-port statistics interface-name" to clear the statistics and then issue the command "show virtual-chassis vc-port statistics extensive" command again to check if the errors are still incrementing.Further information can be found on the next documentation:https://www.juniper.net/documentation/us/en/software/junos/virtual-chassis-fabric/topics/ref/command/show-virtual-chassis-vc-port-statistics.html
08/29/2023