Root port changes every one hour

dannylai1 · ‎02-15-2016

Hi,

I have a HP procurve 2510 acting as the core and 5 x Catalyst 2960S working as edge. MSTP is used but I keep having problem with the Cisco switches that the root port in all 5 cisco switches are changing every 1 hour. Below is the log from one of the Cisco Switch. Port GI1/0/27 and GI1/0/28 are the uplink connected to the HP procurve. I had tried using root guard on the 2 ports but the root port changes still happens every 1 hour. Hope that someone can shed some lights on what is the problem.

025475: Feb 15 2016 18:32:10.765 JST:     pm_vp 1/27(1): during state forwarding, got event 4(remove)
025476: Feb 15 2016 18:32:10.765 JST: @@@ pm_vp 1/27(1): forwarding -> notforwarding
025477: Feb 15 2016 18:32:10.765 JST: *** vp_fwdchange: single: notfwd: 1/27(1)
025478: Feb 15 2016 18:32:10.770 JST: @@@ pm_vp 1/27(1): notforwarding -> present
025479: Feb 15 2016 18:32:10.770 JST: STP SW: 1 virtual port link down single: Gi1/0/27.1
025480: Feb 15 2016 18:32:10.770 JST: MST[0]: updt roles, root port Gi1/0/27 going down
025481: Feb 15 2016 18:32:10.770 JST: MST[0]: Gi1/0/28 is now root port
025482: Feb 15 2016 18:32:10.770 JST: %SPANTREE-5-ROOTCHANGE: Root Changed for instance 0: New Root Port is GigabitEthernet1/0/28. New Root Mac Address is e839.3575.0b00
025483: Feb 15 2016 18:32:10.770 JST: STP SW: deleted vlan 1, ext id 4BDBA68
025484: Feb 15 2016 18:32:10.770 JST: *** vp_linkchange: single: down: 1/27(1)
025485: Feb 15 2016 18:32:10.770 JST: @@@ pm_vp 1/27(1): present -> not_present
025486: Feb 15 2016 18:32:10.770 JST: STP SW: 1 virtual port being removed single: Gi1/0/27.1
025487: Feb 15 2016 18:32:10.770 JST: *** vp_statechange: single: remove: 1/27(1)
025488: Feb 15 2016 18:32:10.775 JST:     pm_vp 1/28(1): during state notforwarding, got event 4(remove)
025489: Feb 15 2016 18:32:10.775 JST: @@@ pm_vp 1/28(1): notforwarding -> present
025490: Feb 15 2016 18:32:10.775 JST: STP SW: 1 virtual port link down single: Gi1/0/28.1
025491: Feb 15 2016 18:32:10.775 JST: MST[0]: updt roles, root port Gi1/0/28 going down
025492: Feb 15 2016 18:32:10.775 JST: STP SW: deleted vlan 1, ext id 4BDD970
025493: Feb 15 2016 18:32:10.775 JST: *** vp_linkchange: single: down: 1/28(1)
025494: Feb 15 2016 18:32:10.775 JST: @@@ pm_vp 1/28(1): present -> not_present
025495: Feb 15 2016 18:32:10.775 JST: STP SW: 1 virtual port being removed single: Gi1/0/28.1
025496: Feb 15 2016 18:32:10.775 JST: *** vp_statechange: single: remove: 1/28(1)
025497: Feb 15 2016 18:32:11.766 JST: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/27, changed state to down
025498: Feb 15 2016 18:32:11.777 JST: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/28, changed state to down
025499: Feb 15 2016 18:32:12.857 JST: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/27, changed state to down
025500: Feb 15 2016 18:32:12.857 JST: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/28, changed state to down
025501: Feb 15 2016 18:32:22.200 JST:     pm_vp 1/28(1): during state not_present, got event 0(add)
025502: Feb 15 2016 18:32:22.200 JST: @@@ pm_vp 1/28(1): not_present -> present
025503: Feb 15 2016 18:32:22.200 JST: STP SW: 1 virtual port being added single: Gi1/0/28.1
025504: Feb 15 2016 18:32:22.200 JST: *** vp_statechange: single: added: 1/28(1)
025505: Feb 15 2016 18:32:22.200 JST:     pm_vp 1/28(1): during state present, got event 8(linkup)
025506: Feb 15 2016 18:32:22.205 JST: @@@ pm_vp 1/28(1): present -> authentication
025507: Feb 15 2016 18:32:22.205 JST:     pm_vp 1/28(1): idle during state authentication
025508: Feb 15 2016 18:32:22.205 JST: @@@ pm_vp 1/28(1): authentication -> authentication
025509: Feb 15 2016 18:32:22.205 JST:     pm_vp 1/28(1): during state authentication, got event 18(authen_disable)
025510: Feb 15 2016 18:32:22.205 JST: @@@ pm_vp 1/28(1): authentication -> notforwarding
025511: Feb 15 2016 18:32:22.205 JST: STP SW: 1 virtual port link up single: Gi1/0/28.1
025512: Feb 15 2016 18:32:22.205 JST: MST[0]: Gi1/0/28 is now designated port
025513: Feb 15 2016 18:32:22.205 JST: *** vp_linkchange: single: up: 1/28(1)
025514: Feb 15 2016 18:32:22.405 JST:     pm_vp 1/27(1): during state not_present, got event 0(add)
025515: Feb 15 2016 18:32:22.405 JST: @@@ pm_vp 1/27(1): not_present -> present
025516: Feb 15 2016 18:32:22.405 JST: STP SW: 1 virtual port being added single: Gi1/0/27.1
025517: Feb 15 2016 18:32:22.405 JST: *** vp_statechange: single: added: 1/27(1)
025518: Feb 15 2016 18:32:22.405 JST:     pm_vp 1/27(1): during state present, got event 8(linkup)
025519: Feb 15 2016 18:32:22.405 JST: @@@ pm_vp 1/27(1): present -> authentication
025520: Feb 15 2016 18:32:22.405 JST:     pm_vp 1/27(1): idle during state authentication
025521: Feb 15 2016 18:32:22.405 JST: @@@ pm_vp 1/27(1): authentication -> authentication
025522: Feb 15 2016 18:32:22.405 JST:     pm_vp 1/27(1): during state authentication, got event 18(authen_disable)
025523: Feb 15 2016 18:32:22.405 JST: @@@ pm_vp 1/27(1): authentication -> notforwarding
025524: Feb 15 2016 18:32:22.405 JST: STP SW: 1 virtual port link up single: Gi1/0/27.1
025525: Feb 15 2016 18:32:22.405 JST: MST[0]: Gi1/0/27 is now designated port
025526: Feb 15 2016 18:32:22.405 JST: *** vp_linkchange: single: up: 1/27(1)
025527: Feb 15 2016 18:32:22.939 JST: MST[0]: updt roles, received superior bpdu on Gi1/0/27
025528: Feb 15 2016 18:32:22.939 JST: MST[0]: Gi1/0/27 is now root port
025529: Feb 15 2016 18:32:22.939 JST: %SPANTREE-5-ROOTCHANGE: Root Changed for instance 0: New Root Port is GigabitEthernet1/0/27. New Root Mac Address is e839.3575.0b00
025530: Feb 15 2016 18:32:22.939 JST: MST[0]: updt roles, received superior bpdu on Gi1/0/28
025531: Feb 15 2016 18:32:22.939 JST: MST[0]: Gi1/0/28 is now alternate port
025532: Feb 15 2016 18:32:22.950 JST:     pm_vp 1/27(1): during state notforwarding, got event 14(forward_notnotify)
025533: Feb 15 2016 18:32:22.950 JST: @@@ pm_vp 1/27(1): notforwarding -> forwarding
025534: Feb 15 2016 18:32:22.950 JST: *** vp_list_fwdchange: forward: 1/27(1)
025535: Feb 15 2016 18:32:22.955 JST: %SPANTREE-5-TOPOTRAP: Topology Change Trap for instance 0
025536: Feb 15 2016 18:32:24.203 JST: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/28, changed state to up
025537: Feb 15 2016 18:32:24.402 JST: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/27, changed state to up
025538: Feb 15 2016 18:32:25.204 JST: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/28, changed state to up
025539: Feb 15 2016 18:32:25.404 JST: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/27, changed state to up

Alexey Prilutskiy · ‎02-15-2016

Looks like both interfaces were restarted, maybe there is a L1 issue. And after restart gi1/0/28 got up earlier and became the root port. But why are you not using etherchannel to aggregate both links and exempt stp from usage?

Mark Malone · ‎02-15-2016

You could try an trace the stp change to the actual port may help that's causing it , this command will show you where the stp change occurred and what port its being seem from you may need to trace it

show spanning-tree detail | inc ieee|occurr|from|is exec

dannylai1 · ‎02-15-2016

Hi,

I have used the command that you mentioned. All the Cisco switch is pointing to those uplink ports connecting to the procurve switch. So this mean it is the HP procurve causing the TC but I am not able to see where is the problem. I had tried to change pathcost, priority and even use root guard but nothing seems to work. I have included the MST information below.

For Cisco Switch, its port 27 & 28(HP end is 43 & 44).

For HP Switch, Port 43 - 48 are the ports connecting to all the cisco switch.

Cisco Switch:

SW-TO04#sh span

MST0
Spanning tree enabled protocol mstp
Root ID    Priority    0
             Address     e839.3575.0b00
             Cost        0
             Port        27 (GigabitEthernet1/0/27)
             Hello Time   2 sec Max Age 20 sec Forward Delay 15 sec

Bridge ID Priority    32768 (priority 32768 sys-id-ext 0)
             Address     ccd5.3997.c080
             Hello Time   2 sec Max Age 10 sec Forward Delay 7 sec

Interface           Role Sts Cost      Prio.Nbr Type
------------------- ---- --- --------- -------- --------------------------------
Gi1/0/2             Desg FWD 200000    128.2    P2p Edge
Gi1/0/5             Desg FWD 200000    128.5    P2p Edge
Gi1/0/11            Desg FWD 20000     128.11   P2p Edge
Gi1/0/12            Desg FWD 20000     128.12   P2p Edge
Gi1/0/14            Desg FWD 200000    128.14   P2p Edge
Gi1/0/17            Desg FWD 20000     128.17   P2p Edge
Gi1/0/18            Desg FWD 20000     128.18   P2p Edge
Gi1/0/19            Desg FWD 20000     128.19   P2p Edge

Interface Role Sts Cost Prio.Nbr Type
------------------- ---- --- --------- -------- --------------------------------

Gi1/0/20            Desg FWD 20000     128.20   P2p Edge
Gi1/0/21            Desg FWD 20000     128.21   P2p Edge
Gi1/0/22            Desg FWD 200000    128.22   P2p Edge
Gi1/0/24            Desg FWD 20000     128.24   P2p Edge
Gi1/0/27            Root FWD 20000     128.27   P2p
Gi1/0/28            Altn BLK 200000    144.28   P2p

HP switch:

SW-TO05# sh span

Multiple Spanning Tree (MST) Information

STP Enabled : Yes
Force Version : MSTP-operation
IST Mapped VLANs : 1-4094
Switch MAC Address : e83935-750b00
Switch Priority : 0
Max Age : 20
Max Hops : 20
Forward Delay : 15

Topology Change Count : 9367
Time Since Last Change : 24 mins

CST Root MAC Address : e83935-750b00
CST Root Priority    : 0
CST Root Path Cost   : 0
CST Root Port        : This switch is root

IST Regional Root MAC Address : e83935-750b00
IST Regional Root Priority    : 0
IST Regional Root Path Cost   : 0
IST Remaining Hops            : 20

Root Guard Ports : 1,43-48
TCN Guard Ports :
Protected Ports : 1-42
Filtered Ports : 1-42

Mark Malone · ‎02-15-2016

Unfortunately I have seen these issues before when HP is integrated with Cisco in STP we actually ripped out the HP switches at the access layer it was not the only problem we came across with them , if this is only an occurrence since you upgraded the firmware I would try roll back for a quick fix if its locking up the network that often , I don't work with HP so im not sure if there's an equivalent command to trace STP like you can with Cisco , there is the option to turn off STP I believe on HP you could try that or I think there is a command you could try even though there both MST spanning-tree force-version stp-compatible or set Cisco switch as MST root take the control away from HP , sorry cant offer more on HP commands not overly familiar with them

http://www.hp.com/rnd/support/manuals/pdf/release_06628_07110/Bk2_Ch5_STP.pdf

dannylai1 · ‎02-15-2016

Hi,

Do you mean there is a problem with the physical cabling? I have taken over this remote site from another admin. It was actually working until I have upgraded the firmware for both Cisco and HP switch. So I could not at the moment reconfigure it the best practice way.