Posted on August 12, 2014 10:44 pm
 |  Asked by Clifford Trueman
 |  26985 views
RESOLVED
0
0
Print Friendly, PDF & Email

So, 2 – 3 times a week our desktop switch is going down which appears to be a spanning tree issue, and I’m not quite sure how to troubleshoot it.  So we get a hicup that’s approx 30 – 45 seconds. Which appears to be the average time of a re-converge process. I’ve gone back and verified that everything has the correct root bridge selected but i’m unsure what to check beyond that.

Below is the output of /var/log/messages from our core switch.

Aug 12 21:01:10 pcelab-7050q16-324chpc-rack-aj13-core Ebra: %LINEPROTO-5-UPDOWN: Line protocol on Interface Ethernet18 (Office 10GB trunk (SR OM3-MMF)), changed state to down
Aug 12 21:01:10 pcelab-7050q16-324chpc-rack-aj13-core Ebra: %LINEPROTO-5-UPDOWN: Line protocol on Interface Port-Channel200, changed state to down
Aug 12 21:01:10 pcelab-7050q16-324chpc-rack-aj13-core kernel: [20063663.403740] __dev_addr_discard: address leakage! da_users=1
Aug 12 21:01:10 pcelab-7050q16-324chpc-rack-aj13-core Stp: %SPANTREE-6-INTERFACE_DEL: Interface Port-Channel200 has been removed from instance MST0
Aug 12 21:04:41 pcelab-7050q16-324chpc-rack-aj13-core Ebra: %LINEPROTO-5-UPDOWN: Line protocol on Interface Ethernet18 (Office 10GB trunk (SR OM3-MMF)), changed state to up
Aug 12 21:04:46 pcelab-7050q16-324chpc-rack-aj13-core Ebra: %LINEPROTO-5-UPDOWN: Line protocol on Interface Ethernet17 (Office 10GB trunk (SR OM3-MMF)), changed state to up
Aug 12 21:04:47 pcelab-7050q16-324chpc-rack-aj13-core Ebra: %LINEPROTO-5-UPDOWN: Line protocol on Interface Port-Channel200, changed state to up
Aug 12 21:04:47 pcelab-7050q16-324chpc-rack-aj13-core Stp: %SPANTREE-6-INTERFACE_ADD: Interface Port-Channel200 has been added to instance MST0
Aug 12 21:04:47 pcelab-7050q16-324chpc-rack-aj13-core Stp: %SPANTREE-6-INTERFACE_STATE: Interface Port-Channel200 instance MST0 moving from discarding to learning
Aug 12 21:04:47 pcelab-7050q16-324chpc-rack-aj13-core Stp: %SPANTREE-6-INTERFACE_STATE: Interface Port-Channel200 instance MST0 moving from learning to forwarding
Marked as spam
0
Posted by Stuart
Answered on August 13, 2014 1:27 pm

As you’ve undoubtedly concluded the link flap on Et17/18 is probably the root cause of the STP reconvergence.  The question becomes: What’s connected to Et17/Et18 and why did the link drop?

It looks from your interface descriptions that these links are MAN or WAN links to another site.  Do they both pass through the same Data Communications Equipment?  If so did that unit reboot?  Do they go to the same switch at the remote site, and if so did anything occur on that switch?

0
Posted by Clifford Trueman
Answered on August 13, 2014 3:32 pm

So I’m looking at the output of ’show lldp neighbors’

Port Neighbor Device ID Neighbor Port ID TTL
Et9/1 ProCurve 6120XG Blade Switch 17 120
Et9/2 ProCurve 6120XG Blade Switch 18 120
Et9/3 ProCurve 6120XG Blade Switch 17 120

The neighbor port ID matches 17/18 which seems odd, but obviously they don’t match the physical port that po200 is on. I’m not sure if that’s a coincidence or not.

So if that’s the case LLDP is enabled for port et 17/18 but nothing shows.

0
Posted by Clifford Trueman
Answered on August 13, 2014 5:48 pm

I enabled LLDP on the other switch got a bit more meaningful output. From that, I was able to determine exactly what was connected for that port channel.  Still unsure on what’s causing the re-converge though.

Port Neighbor Device ID Neighbor Port ID TTL
Et1/1 pcelab-7050t64-324chpc-rack-aj Ethernet49/1 120
Et2/1 pcelab-7050t64-324chpc-rack-aj Ethernet51/1 120
Et3/1 pcelab-7050t64-324chpc-rack-aj Ethernet49/1 120
Et4/1 pcelab-7050t64-324chpc-rack-aj Ethernet51/1 120
Et5/1 pcelab-7050t64-324chpc-rack-e Ethernet49/1 120
Et7/1 00c0.dd27.3e20 00c0.dd27.3e20 120
Et7/2 00c0.dd27.3e22 00c0.dd27.3e22 120
Et7/3 00c0.dd27.3d78 00c0.dd27.3d78 120
Et7/4 00c0.dd27.3d7a 00c0.dd27.3d7a 120
Et8/1 pcelab-7050-64t-324chpc-rack-f Ethernet49/1 120
Et9/1 ProCurve 6120XG Blade Switch 17 120
Et9/2 ProCurve 6120XG Blade Switch 18 120
Et9/3 ProCurve 6120XG Blade Switch 17 120
Et10/1 f8b1.5609.7719 fortyGigE 0/33 120
Et12/1 pcelab-7050t64-324chpc-rack-g Ethernet51/1 120
Et17 pcelab-3750e-306chpc.mgmt.pce. Te1/0/2 180
Et18 pcelab-3750e-306chpc.mgmt.pce. Te1/0/1 180
Et23 innarista01 Ethernet48 120

Post your Answer

You must be logged in to post an answer.