Switch error: %C4K_HWPORTMAN-4-BLOCKEDTXQUEUE

e.domingo · ‎08-26-2009

Hi there,

We've encounter an error on one of our device here in our company. Please see below error:

Aug 20 15:48:40.891: %C4K_HWPORTMAN-4-BLOCKEDTXQUEUE: Blocked transmit queue HwTxQId3 on Switch Phyport Fa5/35, count=181619

Aug 20 15:58:42.369: %C4K_HWPORTMAN-4-BLOCKEDTXQUEUE: Blocked transmit queue HwTxQId1 on Switch Phyport Fa5/35, count=181730

Aug 20 15:58:42.373: %C4K_HWPORTMAN-4-BLOCKEDTXQUEUE: Blocked transmit queue HwTxQId3 on Switch Phyport Fa5/35, count=181732

PCGSW2252>

The issue was previously occurred last May and the module was reset to resolve the issue. Can anyone advise on the cause of this issue and how to resolve this issue so that it will not re-occur again?

We've already tried to shut/no shut the port but still error occurred. Note that there is no device or PC connected to port 5/35.

Hoping for your help. Thanks! :-)

Joe Clarke · ‎08-26-2009

What version of code is this switch running?

e.domingo · ‎09-01-2009

Hi there!

Sorry for the late response. Here's what I got from the show version output.

--------- show version ----------

Cisco IOS Software, Catalyst 4000 L3 Switch Software (cat4000-I9K91S-M), Version 12.2(25)EWA4, RELEASE SOFTWARE (fc1)

Technical Support: http://www.cisco.com/techsupport

Compiled Fri 23-Sep-05 13:24 by ssearch

Image text-base: 0x10000000, data-base: 0x115EF984

ROM: 12.2(31r)SG2

Dagobah Revision 226, Swamp Revision 34

Thanks for the usual support! :)

Joe Clarke · ‎09-01-2009

The closest bug I could find based on your code is CSCsq94908:

Symptom:

C4K_HWPORTMAN-4-BLOCKEDTXQUEUE message appears in reference to a port on WS-C4948.

Conditions:

This message can appear when the port in question is receiving pause

frames from the peer (see "show int gi x/y flow" on this port and the peer port).

show platorm software int gi x/y all | inc Busy

shows an increasing BusyBitActivityCount as in

BusyBitActivityCount : 412

Workaround:

The message itself is not a problem. It indicates the current state of the port.

The port should recover and the message will not be seen when the

pause frames from the peer are stopped.

In rare cases, the port is stuck permanently and module reset is required to restore

normal operation.

Use of a policer in the output direction is likely to prevent a port from getting stuck permanently.

Example - For 1500 byte packets:

interface GigabitEthernet1/47

service-policy output p1

Policy Map p1

Class class-default

police 970 Mbps 500000

It is fixed in 12.2(31)SGA9 and 12.2(50)SG. However, it's not really clear that this is your bug. There were some other unreproducible issues reported. If you don't think the bug I mentioned matches, it might be a good idea to open a TAC service request, and have them analyze this further.

e.domingo · ‎09-03-2009

Thank you very much for the help! I appreciated it. We'll try our best to do the necessary actions. Thanks again! :-)