Switches, Hubs, and Modems
cancel
Showing results for 
Search instead for 
Did you mean: 

Procurve 8200zl STP issues

scam3r
Occasional Visitor

Procurve 8200zl STP issues

I have a core network which consists of three 8200 series switches all identical in firmware, hardware.

13.68M is the version of firmware.

We have some blade switches connected to all three of these switches and they are all linked together with 1Gb/s Fibre, we don't have any STP issues, and the network is stable.

Switches are named A,B and C.

When the command

show config

is run on switch B the blade switches report a STP topology change, but no STP change is reported in the CORE, but the blade switches don't report anything if this command is run on switches A or C. Have checked the CPU on switch B and it doesn't show as being over utilised. SSH isn't running and we only use Telnet to connect to the switches.

It seems strange that switch B does this and the other don't. Any help to try and diagnose this issue. As I am starting to suspect this maybe a hardware issue.
4 REPLIES
Jeff Carrell
Honored Contributor

Re: Procurve 8200zl STP issues

When you do a 'show span' on all of these switches, look for the values of:

Topology Change Count : 13
Time Since Last Change : 15 mins

This starts your STP baseline, then run this command again "after some amount of time" to see if anything changes. Keep looking at this.

This info will lead you to a direction to start troubleshooting. Generally a STP topology change indicates another STP device thinks it should be root, or another STP enabled switch came online and all the switches running STP must at least "re-calculate" their STP status (re-converge).

Make sure you have defined on each switch what its STP root state should be ('stp pri 1' and 2, and 3). This alone generally "solves" alot of STP issues. Remember the STP rules: when all else is equal (default STP configs), the lowest MAC address STP device will become the root, no matter where it is in the network - even a 5 level deep edge switch could become STP root - STP is not about operational topology, it's about lowest MAC address is root (when all else is equal). When you set bridge pri to less than 8 (default), that breaks the tie and its will become root (as long as another switch with same pri does not have a lower MAC addr).

btw, 'show config' only shows startup config and shouldn't be showing any STP "changes".

hth...Jeff
scam3r
Occasional Visitor

Re: Procurve 8200zl STP issues

I have collected the following outputs from the switches, and removed identifying information.
Switch A before

Switch A# sh spanning-tree

Multiple Spanning Tree (MST) Information

STP Enabled : Yes
Force Version : MSTP-operation
IST Mapped VLANs : 1-9,11-14,16-19,21,23,27-29,34,37-39,41-44,46-49,51-54,
56-59,61-64,66-69,71-74,76-79,81-84,86-89,91-94,96,99,
101-104,106-109,111-114,116-119,121-124,126-129,131-134,
136-139,141-144,146-160,163-231,233-265,267-300,303-310,
313-320,323-4094
Switch MAC Address : XXXXXXXXXXXXX
Switch Priority : 0
Max Age : 20
Max Hops : 20
Forward Delay : 15

Topology Change Count : 414
Time Since Last Change : 3 hours

CST Root MAC Address : XXXXXXXXXXXXX
CST Root Priority : 0
CST Root Path Cost : 0
CST Root Port : This switch is root

IST Regional Root MAC Address : XXXXXXXXXXXXX
IST Regional Root Priority : 0
IST Regional Root Path Cost : 0
IST Remaining Hops : 20

Root Guard Ports :
TCN Guard Ports :
BPDU Protected Ports :
BPDU Filtered Ports :
PVST Protected Ports :
PVST Filtered Ports :


Switch B Before

Switch B # sh spanning-tree

Multiple Spanning Tree (MST) Information

STP Enabled : Yes
Force Version : MSTP-operation
IST Mapped VLANs : 1-9,11-14,16-19,21,23,27-29,34,37-39,41-44,46-49,51-54,
56-59,61-64,66-69,71-74,76-79,81-84,86-89,91-94,96,99,
101-104,106-109,111-114,116-119,121-124,126-129,131-134,
136-139,141-144,146-160,163-231,233-265,267-300,303-310,
313-320,323-4094
Switch MAC Address : XXXXXXXXXX
Switch Priority : 4096
Max Age : 20
Max Hops : 20
Forward Delay : 15

Topology Change Count : 835
Time Since Last Change : 3 hours

CST Root MAC Address : XXXXXXXXXXXXX
CST Root Priority : 0
CST Root Path Cost : 0
CST Root Port : Trk1

IST Regional Root MAC Address : XXXXXXXXXXXXXX
IST Regional Root Priority : 0
IST Regional Root Path Cost : 2000
IST Remaining Hops : 19

Root Guard Ports :
TCN Guard Ports :
BPDU Protected Ports :
BPDU Filtered Ports :
PVST Protected Ports :
PVST Filtered Ports :


Switch C Before

Switch C# sh spanning-tree

Multiple Spanning Tree (MST) Information

STP Enabled : Yes
Force Version : MSTP-operation
IST Mapped VLANs : 1-9,11-14,16-19,21,23,27-29,34,37-39,41-44,46-49,51-54,
56-59,61-64,66-69,71-74,76-79,81-84,86-89,91-94,96,99,
101-104,106-109,111-114,116-119,121-124,126-129,131-134,
136-139,141-144,146-160,163-231,233-265,267-300,303-310,
313-320,323-4094
Switch MAC Address : XXXXXXXXXXXXXXXXXXXXX
Switch Priority : 8192
Max Age : 20
Max Hops : 20
Forward Delay : 15

Topology Change Count : 550
Time Since Last Change : 3 hours

CST Root MAC Address : XXXXXXXXXXXXXXXXX
CST Root Priority : 0
CST Root Path Cost : 0
CST Root Port : A21

IST Regional Root MAC Address : XXXXXXXXXXXXXXXXX
IST Regional Root Priority : 0
IST Regional Root Path Cost : 20000
IST Remaining Hops : 19

Root Guard Ports :
TCN Guard Ports :
BPDU Protected Ports :
BPDU Filtered Ports :
PVST Protected Ports :
PVST Filtered Ports :



I will run the command

"show run" from a telnet connection on switch B

The blade switches confirmed a topology change, and below the core switches also confirm this.


Switch A After

Topology Change Count : 472
Time Since Last Change : 26 secs

Switch B After

Topology Change Count : 895
Time Since Last Change : 12 secs


Switch C

Topology Change Count : 576
Time Since Last Change : 19 secs


But what I find disturbing is that just running a "show run" command on switch B causes a topology change when itâ s not even the route switch and it seems that it's too busy and drops BPDU's.
Is this an indication of a possible hardware problem with this switch?
Jeff Carrell
Honored Contributor

Re: Procurve 8200zl STP issues

scam3r said: "But what I find disturbing is that just running a "show run" command on switch B causes a topology change when itâ  s not even the route switch and it seems that it's too busy and drops BPDU's.
Is this an indication of a possible hardware problem with this switch?"

First I would say it looks like alot of topology changes in a short time, that is generally not good.

What is causing it, that's hard to say...could be a flapping interface, bad hardware, device going up-n-down (which looks like a flapping port), etc.

You could also enable BPDU filtering/protection and disable edge ports from listening to BPDU's and see if the topology changes drop.

If it were me, I'd schedule an outage an update your switch code to K.14.65 (or even up to K.15.xx). It may take a 2-cycle process in order to update the bootrom first (I don't remember off the top of my head, but you can read the release notes).

If you call support, they are going to say the same thing anyway.

I have not seen that doing a 'sh run' causes a topology change...VERY interesting.

It may be time for you to contact HPN support..if its hardware, they will know better, and then you can get it replaced.

hth...Jeff
scam3r
Occasional Visitor

Re: Procurve 8200zl STP issues

Thanks Jeff for the info, it seems that the I can only conclude it is a hardware issue, as spanning tree hasn't flapped for 3 weeks before today, because we haven't logged into and run that command.

So I think i will be raising a case with HP. But as you said it may be worth getting the Core switches to a newer version of code, as it will be the first thing that we are going to be asked.

Thanks again

scam