08-08-2012 08:52 PM - edited 03-07-2019 08:14 AM
Hi all,
I have two 3550 switches running HSRP in between them.
Here is what i saw on logs on Active switch
Aug 8 02:15:02.352 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Active -> Speak
Aug 8 02:15:12.353 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Speak -> Standby
Aug 8 02:47:59.406 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Standby -> Active
Aug 8 05:00:02.629 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Active -> Speak
Aug 8 05:00:12.629 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Speak -> Standby
Aug 8 05:06:59.643 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Standby -> Active
Second switch has these logs
000487: Aug 8 02:15:02.351 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Standby
-> Active
000488: Aug 8 02:47:59.405 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Active
-> Speak
000489: Aug 8 02:48:09.405 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Speak -
> Standby
000490: Aug 8 05:00:02.628 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Standby
-> Active
000491: Aug 8 05:06:59.646 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Active
-> Speak
000492: Aug 8 05:07:09.646 MST: %HSRP-5-STATECHANGE: Vlan20 Grp 1 state Speak -
> Standby
3550SMIB#$
Right now they are working as expected.
Is there something i can check to know why the state changed for vlan 20 only?
Thanks
Mahesh
Solved! Go to Solution.
01-22-2013 11:04 PM
It happened once? Dont worry about it then, of course it happens...
We had some pretty aggressive timers and it kept happening until we set them to some reasonable values, but you´ve got the default ones (1,4s), and the preempt is configured with a 60 second delay... maybe you could tune those timers a bit, depends on the environment...
01-23-2013 11:45 AM
It depends on the link, and the distance. For example - if you´ve got a relatively close distribution switches with a GigabitEthernet link - you can go pretty aggressive, as 250ms and 800ms. If we´re talking about a few kilometers and a slower link - you´d have to adjust the values, but I guess 1s and 4s would be fine.
01-23-2013 11:51 AM
Hello,
In your configuration it looks like the HRSP timers haven't been change-
I can see tracking enabled, so when this feature is applied the timers are not applicable if that tracking interface becomes unavailable, HRSP will transition/preempt straight without using these timers -
On the other hand the default timers (default hello 3 secs holdtime 10 secs) are used so the hrsp peers can communicate with each other and if for some reason that communication is lost and the 3xhellos are not received from active to standby peers then then you will see a state change.
sh standby- this will show you detauil of the hrsp settings applied.
So i would start by investigating possible layer 1 & 2 issues - cabling / stp etc
res
Paul
.
Please don't forget to rate this post if it has been helpful.
08-09-2012 12:31 AM
seems to be there is a problem in HSRP timers (hello timer).
can you paste the configuration under HSRP interfaces?
08-09-2012 12:41 AM
Hi Mahesh,
Can you paste your config for hsrp int.
Then only we cabn see the error.
REgards
08-11-2012 07:48 AM
Hi Sandeep,
Here is config of vlan 20 interface from each switch
h run int vlan 20
Building configuration...
Current configuration : 214 bytes
!
interface Vlan20
ip address 192.168.20.1 255.255.255.0
ip ospf hello-interval 40
standby 1 ip 192.168.20.3
standby 1 priority 150
standby 1 preempt delay minimum 60
standby 1 track FastEthernet0/11 60
end
sh run int vlan 20
Building configuration...
Current configuration : 153 bytes
!
interface Vlan20
ip address 192.168.20.2 255.255.255.0
ip ospf hello-interval 40
standby 1 ip 192.168.20.3
standby 1 preempt delay minimum 60
end
Hello timers are same on both as per config.
It happened 1 time only so might be some weird thing.
Does not happen again.
Thanks
MAhesh
01-22-2013 11:04 PM
It happened once? Dont worry about it then, of course it happens...
We had some pretty aggressive timers and it kept happening until we set them to some reasonable values, but you´ve got the default ones (1,4s), and the preempt is configured with a 60 second delay... maybe you could tune those timers a bit, depends on the environment...
01-23-2013 09:54 AM
Hi Mateja,
Thanks for reply.
Any idea what value i should tune HSRP timers?
Thanks
Mahesh
01-23-2013 11:45 AM
It depends on the link, and the distance. For example - if you´ve got a relatively close distribution switches with a GigabitEthernet link - you can go pretty aggressive, as 250ms and 800ms. If we´re talking about a few kilometers and a slower link - you´d have to adjust the values, but I guess 1s and 4s would be fine.
01-23-2013 11:51 AM
Hello,
In your configuration it looks like the HRSP timers haven't been change-
I can see tracking enabled, so when this feature is applied the timers are not applicable if that tracking interface becomes unavailable, HRSP will transition/preempt straight without using these timers -
On the other hand the default timers (default hello 3 secs holdtime 10 secs) are used so the hrsp peers can communicate with each other and if for some reason that communication is lost and the 3xhellos are not received from active to standby peers then then you will see a state change.
sh standby- this will show you detauil of the hrsp settings applied.
So i would start by investigating possible layer 1 & 2 issues - cabling / stp etc
res
Paul
.
Please don't forget to rate this post if it has been helpful.
01-24-2013 08:06 AM
Hi Everyone,
I check the default HSRP timers it is 3 secs and holdtime is 10 secs.
This happened only once.
Also Both HSRP switches are close to each other just 2 feet away.
So i can live with these HSRP timers for now?
Thanks
MAhesh
01-24-2013 09:35 AM
Mahesh
Please don't forget to rate this post if it has been helpful.
01-24-2013 10:08 AM
Hi Everyone,
Many thanks to everyone for helping me out.
Thanks
Mahesh
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