Stack port in STP incosistent state

Unanswered Question
Jan 5th, 2010

Hi All,

There are 3 switces stacked & spanning-tree root primary for one vlan is configured on it. One of the stack port showing as the STP inconsistance state.

Name                 Interface              Inconsistency
-------------------- ---------------------- ------------------
VLAN0001             StackPort3             Loop Inconsistent

Number of inconsistent ports (segments) in the system : 1

As I beleive the stack port should not be part of spanning tree calculation.

Can anyone help me how to troubleshoot this issue. Till now there is no issue but I fear that there may be a loop if it is not resolved.

Thanks & Regards

Sachi

I have this problem too.
0 votes
  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 0 (0 ratings)
Loading.
Mohamed Sobair Tue, 01/05/2010 - 03:10

Hi Sachi,

This message is normal and doesnt mean you have any loop inconsistency. Its just because STP makes the Stacking port asof the current state.

You dont have to worry about it, its normal.

HTH,

Mohamed

sachidananda panda Tue, 01/05/2010 - 03:27

Thanks Mohamed,

But the stack port should always be in forwarding mode right? Do we need to do any specific spanning tree configuration for the master switch.

Thanks

Sachi

Giuseppe Larosa Tue, 01/05/2010 - 04:07

Hello Sachi,

I've checked on one of our stacks and I have 0 incosistent ports

sh spanning-tree inconsistentports

Name                 Interface              Inconsistency
-------------------- ---------------------- ------------------

Number of inconsistent ports (segments) in the system : 0

sh switch
                                               Current
Switch#  Role      Mac Address     Priority     State
--------------------------------------------------------
1       Slave     0013.c35a.c900     10        Ready
2       Slave     0013.c4fc.4680     1         Ready
*3       Master    0013.80b5.b280     9         Ready
4       Slave     0013.c34c.3e00     8         Ready

so something has gone wrong in your stack.

Hope to help

Giuseppe

sachidananda panda Tue, 01/05/2010 - 04:18

Thanks Giuseppe,

I doubt on the stack cable (UDLD). Let me replace it when I get an outage & check. Otherwise there may be a problem with the stack port itself. In that case I may need to replace the switch.

Thanks

Sachi

Mohamed Sobair Tue, 01/05/2010 - 05:00

Guiseppe & Sachi,

The state is irrelevant here, this is Stack port and is used to sync STP information between Stack members.

HTH

Mohamed

ansalaza Tue, 01/05/2010 - 05:27

This could also be a bug:


CSCsx64925    Loopguard config should not be applied to stack ports
CSCee40722    Spanning-tree loopguard causes loops and multiple root ports

Could you attach a show version?

Do you have loopguard enabled globally?


Thanks,

sachidananda panda Tue, 01/05/2010 - 06:39

Hi,

Pls find the sh ver

QTR-TECH-TBDC-AS-001#sh ver
Cisco IOS Software, C3750 Software (C3750-IPBASE-M), Version 12.2(25)SEE3, RELEA
SE SOFTWARE (fc2)
Copyright (c) 1986-2007 by Cisco Systems, Inc.
Compiled Thu 22-Feb-07 15:04 by myl
Image text-base: 0x00003000, data-base: 0x00EE40E0

ROM: Bootstrap program is C3750 boot loader
BOOTLDR: C3750 Boot Loader (C3750-HBOOT-M) Version 12.2(25r)SEE4, RELEASE SOFTWA
RE (fc1)

QTR-TECH-TBDC-AS-001 uptime is 16 weeks, 16 hours, 4 minutes
System returned to ROM by error - Debug Exception (Could be NULL pointer derefer
ence) Exception (0x2000)!, PC 0x9060D8
System restarted at 20:11:19 DST Mon Sep 14 2009
System image file is "flash:c3750-ipbase-mz.122-25.SEE3.bin"

cisco WS-C3750G-48PS (PowerPC405) processor (revision F0) with 118784K/12280K by
tes of memory.
Processor board ID FOC1107Z5JM
Last reset from power-on
2 Virtual Ethernet interfaces
156 Gigabit Ethernet interfaces
The password-recovery mechanism is enabled.

512K bytes of flash-simulated non-volatile configuration memory.
Base ethernet MAC Address       : 00:1B:2A:F0:3E:80
Motherboard assembly number     : 73-10216-08
Power supply part number        : 341-0108-03
Motherboard serial number       : FOC11070C7Q
Power supply serial number      : DCA1104A0VR
Model revision number           : F0
Motherboard revision number     : C0
Model number                    : WS-C3750G-48PS-S
System serial number            : FOC1107Z5JM
Top Assembly Part Number        : 800-26853-01
Top Assembly Revision Number    : B0
Version ID                      : V05
CLEI Code Number                : CNMWN00ARC
Hardware Board Revision Number  : 0x09


Switch   Ports  Model              SW Version              SW Image
------   -----  -----              ----------              ----------
     1   52     WS-C3750G-48PS     12.2(25)SEE3            C3750-IPBASE-M
     2   52     WS-C3750G-48PS     12.2(25)SEE3            C3750-IPBASE-M
*    3   52     WS-C3750G-48PS     12.2(25)SEE3            C3750-IPBASE-M

Yes Loopguard is enabled globaly.

Thanks

Sachi

sachidananda panda Tue, 01/05/2010 - 06:49

I think the bug CSCsx64925 due to which stack ports are blocked by loopguard configured globaly. which first found in 12.2(46). My switch image is 12.2(25). It is fixed in 12.2(50)

Thanks a lot for the input I will upgrade the image.

Thanks & regards

Sachi

Francois Tallet Tue, 01/05/2010 - 10:24

The hardware of the stack port cannot block, and they cannot introduce a loop either. BPDUs are sent across the stack port to synchronize STP across the members of the stack, but the forwarding across the stack does not rely on STP at all. In fact, there are several normal scenarios where the state of the stack port will be blocking. In fact, you should not see them at all in your show spanning-tree output. It's there only  because you enable "debug spanning-tree event", the whole stack port should be hidden.

Now, I don't know of a case where a stack port ends up "loop inconsistent". That might well be a bug indeed, as mentioned earlier. All I can say is that your fear of loops across the stack are not realistic;-) Don't worry!

Regards,

Francois

ansalaza Tue, 01/05/2010 - 06:54

You could also try to remove global loopguard and enable them individually on interfaces except on the Stack ports

HTH,

Please rate useful posts!

jjenkins1 Tue, 07/12/2011 - 16:42

I experienced a spanning-tree meltdown related to this bug.  A link-flap triggered it, causing two different switches in the stack to assume root-bridge for all their vlans.  This resulted in a loop that took down the network.  I saw the loop inconsistent state recently on a 3-stack of 3750's, which we cleared by rebooting.  I'd recommend an upgrade to 12.2(50) to get around it. 

Actions

This Discussion