02-24-2008 06:47 PM - edited 03-05-2019 09:20 PM
Hi NetPro,
anyone who has experienced about this ?
i was wondering i have fixed the Root and Priority at Core-switch for all vlans. how come new root bridge will automatically changed ?
2008 Feb 24 16:06:01 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 2/1. New Root mac address is 00-0a-8b-04-02-00.
2008 Feb 24 16:06:17 GMT8 +08:00 %SPANTREE-5-MSGAGEEXPIRY: Msg Age timer expired on port 3/4 in vlan 71
2008 Feb 24 16:06:17 GMT8 +08:00 %SPANTREE-5-MSGAGEEXPIRY: Msg Age timer expired on port 1/1-2 (agPort 13/1) in vlan 71
2008 Feb 24 16:06:17 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 0/0. New Root mac address is 00-0a-8b-24-90-46.
2008 Feb 24 16:06:17 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 13/1. New Root mac address is 00-0a-8b-25-08-46.
2008 Feb 24 16:06:18 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 3/6. New Root mac address is 00-0a-8b-04-02-00.
2008 Feb 24 16:06:37 GMT8 +08:00 %SPANTREE-5-MSGAGEEXPIRY: Msg Age timer expired on port 3/6 in vlan 71
2008 Feb 24 16:06:37 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 0/0. New Root mac address is 00-0a-8b-24-90-46.
2008 Feb 24 16:06:37 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 2/2. New Root mac address is 00-0a-8b-25-08-46.
your reply will be very higly appreciated.
thanks ya.
Regards,
Jack
02-24-2008 08:20 PM
Hi Jack,
To understand the messages and these changes you probably need to post the topology and the connected switches..however..these messages indicate that there might be switches in the spanning tree network which are trying to override the current root bridge for VLAN 71..
To isolate the same try to put "spanning-tree root primary vlan71" command on the switch which you want to make primary root in the entire domain for VLAN 71..so that it dynamically changes priority values to become root in the spanning domain..
Also you can try putting "spanning-tree root guard" (globally or on interfaces on which you want to block BPDU from other switches for this switch to become ROOT..) that you want to block from command to guard the current root from being changed in the current spanning domain..
I also think that as Max Age Timer is expiring check if BPDU's are being transferred in the network..
Akhil
02-24-2008 08:46 PM
Hi Akhil,
your info very helpful, currently i am using PVST and this is 2 Tier network only. basically i have set the vlan 71 as priority 8192 in the core switch.
so, from your msg you advice to configure " spanning-tree root primary vlan71 " . would it be any impaction to entire network ?
since you have mentioned Max Age Timer . anything to do with this ?
your reply will be highly appreciated.
thanks
Regards,
Jack
02-24-2008 08:54 PM
Hi Jack,
I cannot suggest changes without a network diagram mentioning the current setup..but yes..if you would like to make the current switch as root for the VLAN 71 in spanning tree domain..it would only effect the VLAN 71 for root selection in spanning tree..so i would suggest to review your toplogy and select the best root bridge for VLAN 71 in your network..
spanning-tree vlan 71 root primary --this command when applied on a switch changes the bridges priority dynamically to select it as root bridge for VLAN 71..
Akhil
02-24-2008 09:48 PM
Hi Akhil,
again thanks for your helpful msg, i am curious about Spantree 71 took over by some other switch and it affected to some of my switch turn to Errordisabled .
" Feb 23 11:06:57.105: %UDLD-4-UDLD_PORT_DISABLED: UDLD disabled interface Gi0/1, unidirectional link detected
Feb 23 11:06:57.105: %PM-4-ERR_DISABLE: udld error detected on Gi0/1, putting Gi0/1 in err-disable state "
is this correct ?
regards,
Jack
02-24-2008 10:26 PM
Hi Jack,
The above messages suggest that UDLD was enabled on the links between switches and was put into error-disabled state when it recognised the unidirectional link..
Please refer to the following link :
-- Here you can check how the links are connected through copper or fiber cables..normally it is enabled on fiber links..also check for duplex mismatches or problems on the interfaces..
Akhil
02-24-2008 10:59 PM
Hi Akhil,
the reason i asked this is because the logs file from core show "
2008 Feb 24 16:11:26 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 2/6. New Root mac address is 00-0a-8b-04-02-00. " so, when i trace from 2/6 , it shown connected to the switch wich i got the UDLD logs.
so, i got the feeling it's something to do with this.
your reply will be higly appreciated.
regards,
Jack
02-24-2008 11:06 PM
Hi Jack,
Cannot comment on the same without knowing the current network topology for the connected switches..
Akhil
02-25-2008 12:54 PM
It seems you have some trace enabled (not a Catos expert), and what you see is the detail of the root election after the root information has aged out. The root information aging out seems to be consistent with your UDLD issue: a link was brought down, and, as a result, the root bridge was not accessible any more.
2008 Feb 24 16:06:01 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 2/1. New Root mac address is 00-0a-8b-04-02-00.
The bridge has aged out the root information on the root port and just elected a new root port (2/1 was probably alternate discarding before).
2008 Feb 24 16:06:17 GMT8 +08:00 %SPANTREE-5-MSGAGEEXPIRY: Msg Age timer expired on port 3/4 in vlan 71
2008 Feb 24 16:06:17 GMT8 +08:00 %SPANTREE-5-MSGAGEEXPIRY: Msg Age timer expired on port 1/1-2 (agPort 13/1) in vlan 71
Some other blocked ports are aging out the root information.
2008 Feb 24 16:06:17 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 0/0. New Root mac address is 00-0a-8b-24-90-46.
The root information has aged out on the root port and it has actually aged out on all the ports! The bridge thinks it is the root.
2008 Feb 24 16:06:17 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 13/1. New Root mac address is 00-0a-8b-25-08-46.
Some other bridge with a better priority than ours pretends to be the root (because it has also aged out the root information). We're considering it as the root right now, it's the best root we heard from so far.
2008 Feb 24 16:06:18 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 3/6. New Root mac address is 00-0a-8b-04-02-00.
A neighbor has not aged out the original root information, it just sent it to us, we are accepting this port as a root port
2008 Feb 24 16:06:37 GMT8 +08:00 %SPANTREE-5-MSGAGEEXPIRY: Msg Age timer expired on port 3/6 in vlan 71
The information we had received in the previous step now aged out...
2008 Feb 24 16:06:37 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 0/0. New Root mac address is 00-0a-8b-24-90-46.
Back to thinking we are the root...
2008 Feb 24 16:06:37 GMT8 +08:00 %SPANTREE-5-ROOTCHANGE:Root changed for Vlan/Instance 71: New root port 2/2. New Root mac address is 00-0a-8b-25-08-46.
And then we're receiving better information from someone else.
All those messages don't really matter. With regular STP, the root information is disappearing from the network in 20 seconds (max age). Some port might need to unblock if there is some change in the final topology. It will 30 seconds to do so. The intermediate steps are not important, they depend on who is aging the root information first. What is important is that the new root bridge is finally taking over and imposing its topology.
Regards,
Francois
02-25-2008 06:34 PM
Hi Francois,
your explanation really good and helpful, so is that any preventive guide ? because, this really affected my current network. i have few switches down caused by UDLD Port Disabled.
Feb 23 11:06:57.105: %UDLD-4-UDLD_PORT_DISABLED: UDLD disabled interface Gi0/1, unidirectional link detected
Feb 23 11:06:57.105: %PM-4-ERR_DISABLE: udld error detected on Gi0/1, putting Gi0/1 in err-disable state
regards,
Jack
02-26-2008 12:54 PM
Hi Jack,
The STP messages seem rather clear to me. I think they are a consequence of your problem, not the source. UDLD is supposed to detect unidirectional link failures (ie a link that is able to transmit in only a single direction while staying up). This is important because this kind of failure can cause loss of connectivity without reconvergence of the L2 control protocols, or even worse, introduce loops in your network. If you get this consistently on a link, I would suspect a hw issue.
Regards,
Francois
02-26-2008 05:33 PM
Hi Francois,
thanks for your information, yes i do agreed with you. would it be possible this is caused by Bug ? because i noticed most of the UDLD detected at Core to Distribution layer. and currently my Core-switch running
thanks a lot.
regards,
Jack
02-26-2008 05:37 PM
Hi Jack,
It could be a bug indeed. I cannot rule this out. I'm not very knowledgeable in either CatOS or UDLD, so if someone has any experience in those kind of trouble, please help;-)
Regards,
Francois
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide