Failover Fails no resilency

[motivatedtek]

Help please
I have a mitel 3300 with 4.1 sp1. This is a new install.
2 controllers 2nd controller just for resiliency.
both controllers on the same subnet.
When I disconnect the network connection from the primary the resislent phones do not register to the secondary.

Cluster is built synched and sharing

Phones are in the remote directory

Users and devices are set to secondary ICP

When I register a phone I can see the mac address in both controllers.

If I set the call server IP address to the secondary in my dhcp scope:
Phones register and work fine
I can see the mac address in both controllers.
The phone gives the indication that it has failed over
(flashing black box)

I mirrored the port on my switch to capture frames going to my secondary.
The phones don't even try to connect to the secondary.

How does the phone recieve the IP address of the secondary ?
The real question of course is how do I fix this ?

I talked to tech support they said that the system seemed to be progammed correctly.Everything seems good except it doesn't work.

I know that one or all of you guys know the answer.
Thanks in advance.

 

[Supernova99]

Have you got both controllers IP addresses in the DHCP 125 option?

You need to put the primary controller IP address first separated by a comma and a space then the secondary controller
This needs to be both the call_srv and the srv_tftp
in this format 192,168.1.2, 192.168.1.20



Share what you know - Learn what you don't

 

[kwbMitel]

Supernova's info is not correct. You do not require both controller addresses to be delivered via DHCP.

The second controller address is populated in the phone when the phone downloads it's config info from the primary controller.

Phones are in the remote directory
On Both Controllers? Does the remote Directory show a Primary AND Secondary entry? Both controllers must match.

How does the phone recieve the IP address of the secondary ?
When downloading it's config. To check the settings in the phone, use the debug mode. THB has instructions on debug.

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

 

[Supernova99]

This taken from the Tech handbook is most certainly correct
It will not work without it as the phone does not what IP address to connect to

Call Server Address (call_srv)


The RTC IP Address for this controller. For resilient systems, enter up to four IP addresses of remote failover RTCs. Separate entries with a comma and a space: ", " .


RTC IP address of controller

Share what you know - Learn what you don't

 

[LoopyLou]

I was under the impression it worked as Super said. The DHCP 125 needs IP of secondary. The phones get it on boot and when the primary fails they should attach immediately or if on a call at the end of the call.

At least thats how I thought it worked.

Agree with kwb that you should put a phone into debug and check that it has the IP's of both systems.

There may be no i in team but there are three f's in fudge off.

 

[kwbMitel]

Supernova and Loopylou. Delivering the 2nd controller IP address with DHCP will help under certain conditions such as booting the phone when the Primary controller is down, but it is not otherwise necessary. The phone will be told the secondary address by the primary controller.

For the purpose of resolving this specific issue. It is not necessary at all.

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

 

[LoopyLou]

kwbMitel, for my own info how or when does the primary tell the phone the IP of the secondary? Is it once the phone has registered and is coming up? Just curious.

There may be no i in team but there are three f's in fudge off.

 

[kwbMitel]

Yes, when the phone is downloading it's config, and it is resilient, it will get data on the secondary according to what is programmed in the remote directory form. This is true regardless of how DHCP is programmed and which controller responds.

DHCP programming of a second controller has nothing to do with resiliency. All it does is designate a second controller (or more) as an alternate source for the boot files.

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

 

[Supernova99]

Does IP trunking work correctly between controllers?
Do you have the correct IP address of the secondary controller in the ICP/PBX Networking form?
The System uses the ICP/PBX Networking form to redirect the phone to the correct controller.

Share what you know - Learn what you don't

 

[Supernova99]

kwbMitel wrote Quote "DHCP programming of a second controller has nothing to do with resiliency. All it does is designate a second controller (or more) as an alternate source for the boot files."

Which is quite important if you were to power cycle your phone when the Primary controller is down.

A lot of sites do not have this programmed either on the 3300 DHCP server, or an external Windows DHCP server. This has caused problems as the phones are unable to get the boot code.

Agreed that this is probably not the reason why the phones are showing that they are resilient when they are not.



Share what you know - Learn what you don't

 

[motivatedtek]

Thank You for you responses
REoes IP trunking work correctly between controllers?
I have not built ip trunking between the controllers because no phones register to the secondary.
the only purpose for the secondary controller is resiliency and nothing will register to it unless the primay fails in which case all phones will register to the secondary.

REo you have the correct IP address of the secondary controller in the ICP/PBX Networking form?
I believe that I have programed that form correctly however I will verify monday afternoon which is the next time I have a chance to work on the system.

When the phone boots up I see the call server address on the display should I see two server addresses?

 

[kwbMitel]

Please check the Remote directory form on both servers.

Re: Cluster is built synched and sharing
I have not built ip trunking between the controllers

A cluster does not exist without IP trunking.


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

 

[Supernova99]

neither does resiliency

Share what you know - Learn what you don't

 

[motivatedtek]

Update
Re: Please check the Remote directory form on both servers
I checked the RDN on each controller the phones appear
in each.

REo you have the correct IP address of the secondary
controller in the ICP/PBX Networking form?
YES. Both have both.

I have built IP trunking still no go.
I did also try adding the 2nd server to my DHCP scope.
no go. It does say you should in the help applet.
I sent a copy of my databases to tech support.They were supposed to contact me by noon yesterday.Note the supposed to.
I will let you know. I am working on other Jobs so I won't be working on it again until Friday.
Thanks for your help

 

[MitelAU]

Tested it yesterday in Lab. Works perfectly.

Knowledge is power, thats why we share!

 

[motivatedtek]

It works
I knew it was something stupid.
You do have to configure IP trunking and ARS.
thanks for the help