Resiliency Failover - Sets do not return to Primary

[mitelmitel]

After Sys A fails over to Sys B, and Sys A comes back up the sets do not automatically rehome to their primary.

Any obvious places to look?

3300 R8.

 

[slapin]

system timers. It happens in about 5 to 10 minutes after multiple health checks of the primary controller after it comes on-line.

 

[mitelmitel]

I take it these timers are not configurable...?

I suppose the HST Cancel Handoff command will kick them back quicker?

 

[slapin]

Why you want them quicker? Most likely there are reasons behind this design. Just make sure that sets will reconnect after 10 minutes and leave them alone

 

[kwbMitel]

I was just looking for an answer to your MWI question and I came across a note that if a resilient phone has MWI appearances on keys the phone will not fail back until the messages are retrieved.

Worth noting.

*******************************************************
Occam's Razor - All things being equal, the simplest solution is the right one.

 

[mitelmitel]

Ah, that might be what the issue is!

Phones are not failing back.

I will check to see I'd that is the case with MWI Keys.

Thanks

 

[Lundah]

The MWI thing only happens if the MWI is activate while the set is in resilient mode. Also, there's some known bugs related to failback (particularly in larger implementations) that are fixed in 9.0UR3.

 

[mitelmitel]

All my systems are on R8 we currently have an upgrade plan to go to 9 UR3 so good news then.

 

[DomingoDolores]

What UR are you running?
Got the same failure.

See:

http://tek-tips.com/viewthread.cfm?qid=1569189&page=2

Imagination is more important than knowledge....

 

[UncleRalph]

This from the most recent KPI notes from Mitel.

MN00279788 - IP phones would not go to primary ICP, After primary ICP reboot phones registered OKl
REPORTED IN SW LOAD - 9.0.1.17
PLANNED SW LOAD - MCD4.1

SYMPTOMS
IP phones all went to resilent ICP.
They would not go back to primary.
Rebooting the phone did not solve the issue.

WORKAROUND
Reboot the primary ICP

Uncle Ralph

 

[veenstrahiem]

Yes you can change the timers for getting the phones back to primary quicker, look under system options>Controller registry configuration. There are 3 options available:
Resiliency: allow return to primary : yes
Resilience: Secondary ICP health check interval sec. : 60
Resliience: succesfull health checks prior to handoff: 5

So, after 5 succesfull checks with 60 sec between, the phone would go back to primary, you CAN change these.

We, however, have the same problem with 8.0 UR5 that the phone do not go back to primary, never. Only after rebooting every phone individualy the will return to their primary controller.

Also check out thread:

http://www.tek-tips.com/viewthread.cfm?qid=1569189&page=6

 

[Devolution]

I have just had this problem happen to a customer.

2xMxe's both Ver 8.0.5.12

Sys 1 -lost link to network,
Sys 2 -picked up all the extns succesfully,
Sys 1 -Came back online,
Sys 2 -Kept the extns past the programmed time limit,
Sys 1 -Rebooted and only about 60% of extns registered, left this overnight incase it took longer.
Sys 2 -Rebooted next morning and finally phones registered on Sys 1.

Problem is, this is a customer that cannot have any down time whatsoever. They're not impressed but we do have MCD 4.0 in the pipelines before xmas.

Is there any fix apart from upgrading for the mean time?