danramirez
Programmer
Hi, I just opened a ticket with Tech Central, I just like to share it with you guys, I would appreciate your comments as well.
Be patient, it's a long story...
We recently networked two PBXs together as “Cluster” and then programmed some phones to be resilient on the mate controller.
Note that PBXs are in different subnets, different Vlans, but there’s one common DHCP server that provides Options to many scopes. (Scopes are: DataA Vlan, DataB Vlan, PBXA Vlan, and PBXB Vlan).
DHCP is an MSL Server with four SCOPES; note we are using Option 125 for Mitel Phones.
Also each PBX is being used as TFTP server for its phones.
Ports on switch are Tagged to correspondent Voice Vlan and UNTagged to Data Vlan, L3 Switch forwards DHCP/BootP broadcasts to IP Address of DHCP Server (DHCP Helper).
DHCP Scope for PBXA Vlan has got IP Address of “PBXA” Controller, then IP address of “PBXB” Controller separated by a “,” (Call server address: 172.16.10.10,172.16.20.10), and also these same addresses for TFTP (TFTP server address: 172.16.10.10,172.16.20.10).
DHCP Scope for PBXB Vlan has got IP Address of “PBXB” Controller, then IP address of “PBXA” Controller separated by a “,” as well. (Call server address 172.16.20.10,172.16.10.10) , and also these same addresses for TFTP (TFTP server address: 172.16.20.10,172.16.10.10).
Note that: When entering into debug mode on the phone, (Vol UP, Vol DWN, CFG…) I see: Current Controller, ICP1, ICP2, BUT ONLY ONE TFTP SERVER, Phone is 5312. I don’t see the second IP address of TFTP.
Resilient phones are programmed to have a secondary controller.
I have made an effort to program everything according to “the book”, Resiliency Guidelines, Help online, etc.
Up to this point almost everything works fine.
A phone that is registered to controller A nicely registers to secondary controller when PBXA is disconnected from the network, and then it rehomes back to controller A when it’s back connected to the network.
Resiliency also works fine when somebody on PBXB loses its home controller.
Now, here is where we have a problem:
If a phone which home element is PBXA, loses its controller (i.e PBXA is disconnected form the network) the phone will register on its secondary controller with no problem. But now, if that same phone is unplugged from the network while its home controller is out of service, and then the phone is plugged back in to the network, the phone will not boot up.
I think that the phone will receive options from DHCP server and will be asked to look for ICP1, or ICP2, but also to look for TFTP server on ICP1 (PBXA). At this point TFTP server on PBXA is not available as PBX is not connected to the network.
The phone doesn’t seem to get secondary TFTP server information, and that’s probably why it’s not booting up.
I have resolved this by setting only one external TFTP server (Only one TFTP server for both PBXs) and setting the DHCP Server to provide TFTP information. Option 125 for all DHCP scopes will point to TFTP server address: 172.16.20.4 (The External TFTP Server).
Questions:
1. Is this the way to do this? Do I need External TFTP servers when implementing resiliency?
2. If the answer to 1. is YES, where do I download TFTP files from? –I can’t find those on MOL. By now I had to get all the files from one of the 3300 controllers (/sysro/tftp) and load them on the external TFTP Server (solarwinds).
3. Any TFTP server that Mitel recommends for this purpose.
4. The customer surely will ask that they need TFTP redundancy, how is this achive? Should we leave this to the IT guys?
Please let me know if I’m doing something wrong or I’m missing some programming, maybe there’s a way to tell the phone to look for an alternative TFTP server in case its primary is not available.
Regards,
Be patient, it's a long story...
We recently networked two PBXs together as “Cluster” and then programmed some phones to be resilient on the mate controller.
Note that PBXs are in different subnets, different Vlans, but there’s one common DHCP server that provides Options to many scopes. (Scopes are: DataA Vlan, DataB Vlan, PBXA Vlan, and PBXB Vlan).
DHCP is an MSL Server with four SCOPES; note we are using Option 125 for Mitel Phones.
Also each PBX is being used as TFTP server for its phones.
Ports on switch are Tagged to correspondent Voice Vlan and UNTagged to Data Vlan, L3 Switch forwards DHCP/BootP broadcasts to IP Address of DHCP Server (DHCP Helper).
DHCP Scope for PBXA Vlan has got IP Address of “PBXA” Controller, then IP address of “PBXB” Controller separated by a “,” (Call server address: 172.16.10.10,172.16.20.10), and also these same addresses for TFTP (TFTP server address: 172.16.10.10,172.16.20.10).
DHCP Scope for PBXB Vlan has got IP Address of “PBXB” Controller, then IP address of “PBXA” Controller separated by a “,” as well. (Call server address 172.16.20.10,172.16.10.10) , and also these same addresses for TFTP (TFTP server address: 172.16.20.10,172.16.10.10).
Note that: When entering into debug mode on the phone, (Vol UP, Vol DWN, CFG…) I see: Current Controller, ICP1, ICP2, BUT ONLY ONE TFTP SERVER, Phone is 5312. I don’t see the second IP address of TFTP.
Resilient phones are programmed to have a secondary controller.
I have made an effort to program everything according to “the book”, Resiliency Guidelines, Help online, etc.
Up to this point almost everything works fine.
A phone that is registered to controller A nicely registers to secondary controller when PBXA is disconnected from the network, and then it rehomes back to controller A when it’s back connected to the network.
Resiliency also works fine when somebody on PBXB loses its home controller.
Now, here is where we have a problem:
If a phone which home element is PBXA, loses its controller (i.e PBXA is disconnected form the network) the phone will register on its secondary controller with no problem. But now, if that same phone is unplugged from the network while its home controller is out of service, and then the phone is plugged back in to the network, the phone will not boot up.
I think that the phone will receive options from DHCP server and will be asked to look for ICP1, or ICP2, but also to look for TFTP server on ICP1 (PBXA). At this point TFTP server on PBXA is not available as PBX is not connected to the network.
The phone doesn’t seem to get secondary TFTP server information, and that’s probably why it’s not booting up.
I have resolved this by setting only one external TFTP server (Only one TFTP server for both PBXs) and setting the DHCP Server to provide TFTP information. Option 125 for all DHCP scopes will point to TFTP server address: 172.16.20.4 (The External TFTP Server).
Questions:
1. Is this the way to do this? Do I need External TFTP servers when implementing resiliency?
2. If the answer to 1. is YES, where do I download TFTP files from? –I can’t find those on MOL. By now I had to get all the files from one of the 3300 controllers (/sysro/tftp) and load them on the external TFTP Server (solarwinds).
3. Any TFTP server that Mitel recommends for this purpose.
4. The customer surely will ask that they need TFTP redundancy, how is this achive? Should we leave this to the IT guys?
Please let me know if I’m doing something wrong or I’m missing some programming, maybe there’s a way to tell the phone to look for an alternative TFTP server in case its primary is not available.
Regards,