Causes for Ethernet Discards and STP System Log Messages

[zimage]

We have a WS-12-DC deployed in the field and we've been having issues with it.

Code: Select all

Feb 13 20:53:10 STP: set port 9 to discarding
Feb 13 20:53:10 STP: set port 9 to learning
Feb 13 20:53:10 STP: set port 9 to forwarding
Feb 14 00:11:05 STP: set port 9 to discarding
Feb 14 00:11:05 STP: set port 9 to learning
Feb 14 00:11:05 STP: set port 9 to forwarding
Feb 14 00:31:00 STP: set port 9 to discarding
Feb 14 00:31:00 STP: set port 9 to learning
Feb 14 00:31:00 STP: set port 9 to forwarding
Feb 14 02:35:00 STP: set port 9 to discarding
Feb 14 02:35:01 STP: set port 9 to learning
Feb 14 02:35:01 STP: set port 9 to forwarding
Feb 14 05:01:09 STP: set port 9 to discarding
Feb 14 05:01:09 STP: set port 9 to learning
Feb 14 05:01:09 STP: set port 9 to forwarding
Feb 14 07:45:10 STP: set port 9 to discarding
Feb 14 07:45:10 STP: set port 9 to learning
Feb 14 07:45:10 STP: set port 9 to forwarding
Feb 14 09:02:51 STP: set port 9 to discarding
Feb 14 09:02:51 STP: set port 9 to learning
Feb 14 09:02:51 STP: set port 9 to forwarding


There shouldn't be any spanning tree loops on that part of the network. our SNMP graphing tool is showing ifInDiscards on most of the interfaces.

Code: Select all

IF-MIB::ifInDiscards.1 = Counter32: 0
IF-MIB::ifInDiscards.2 = Counter32: 68
IF-MIB::ifInDiscards.3 = Counter32: 0
IF-MIB::ifInDiscards.4 = Counter32: 1296436
IF-MIB::ifInDiscards.5 = Counter32: 15180
IF-MIB::ifInDiscards.6 = Counter32: 15005
IF-MIB::ifInDiscards.7 = Counter32: 15001
IF-MIB::ifInDiscards.8 = Counter32: 107379
IF-MIB::ifInDiscards.9 = Counter32: 306429
IF-MIB::ifInDiscards.10 = Counter32: 89129


I'm guessing that there's a bad ethernet connection and CRC corruption causing the discards, and occasional ethernet link state changes are causing the STP messages. We saw similar STP logs and discarded traffic on ethernet port 10 and disabling STP on that port last night fixed the packet loss. What steps do you recommend for troubleshooting.

[sirhc]

Yes, you have either a bad ethernet connection, a bad port in the switch, noise on the line, or a bad port in the device.

If the Ethernet link drops then STP goes through that process.

Another cause of this behavior is when people set the port speed / duplex to a fixed value instead of Auto so if there is any noise on the line the link drops and triggers this cycle through normal STP process of blocking/learning/forwarding.

I would check the Ethernet cable, Ends, and so on. Make sure the switch port and the device is set to AUTO on speed / duplex.