AES did not fail over to the CM/ESS as expected, when the CM core server went offline. AES switch link shows in state "TCP Down".
Question1: This configuration is needed if the Main server down and ESS can take the control on ALL G450s over. Right? Example 1,2.
Answer1: If you configure the MGC list in G450s(as you mentioned via phone), then it will go and register with ESS once the main CM is down.
Q2: This configuration is not needed for the local LSP to take over the control ONLY on local G450. Right? Example 3 should be removed!?
A2: If the ESS is not reachable, then if you have configured LSP in the backup servers in NR, then LSP would take over
Q3: If we would have 2 ESS server (1st in Krakow and 2nd in Kazan) then we can add the Kazan ESS to the 2nd place on Backup server list. Right?
A3: That is also right, so you can add it to the list.
Q4: What is the expected behavior of the AES when CM fails over to the ESS?
A4: AES would only use the ESS transport link connection when the main CM transport link goes down, refer to attached White Paper for further details.
1. Network region (Backup server) config needed only for Phone registration only?
- So, if Main server down then the Phones (only) will be registered to the NR config ESS/LSP but the G450 will not (or only if there is an MGC list configured).
2. G450 (MGC List) Server needed for G450 registration only?
- So, if Main server down then the G450 will register to MGC List ESS/LSP only but the phone will not (or only if NR configured).
Is there any overlap between these configurations? - No overlap.
For MGC to go to LSP you need to configure MGC list and put LSP/ESS and set transition point correctly.
For phones you will be getting the backup server from CM from network region form.
Note: For DMCC stations, the IP station range that AES would controlled in usually defined on the far end application.
Communication Manager 5.2.1 failover test to their ESS (Enterprise Survivable Server) and noted that the AES Switch and CTI Links did not reconnect to the CM ESS.
Avaya Application Enablement Services High Availability White Paper
Avaya Aura® Communication Manager (CM) provides survivable core servers (also known as ESS) and survivable remote servers (also known as LSP) for failover from the primary media server. This feature provides the ability for media gateways, endpoints, application servers like AE Services and its applications to continue their operations without a major outage. Survivable server configurations have been supported, with some limitations, since the AE Services 3.0 release. However, important enhancements were added to both the AE Services 6.1 release and the Communication Manager 6.0 release to better support these configurations.
Since the AE Services 6.1 and Communication Manager 6.0 release, switch connections on both the Control Local Area Network interface cards (CLANs) and Processor Ethernet (PE)
connections are fully supported in all survivable server configurations. Additionally, any Device, Media and Call Control (DMCC) endpoints registered to the primary switch using the
Time-To-Service feature (TTS) will automatically re-register to a survivable server. DMCC endpoint registrations not using the Time-To-Service feature will be unregistered when the Communication
Manager fails over to a survivable server.
AE Services 6.1 and later allows the Communication Manager survivable server nodes to be administered within a switch connection, in a priority order, along with the PE IP address of those nodes.
When used in conjunction with Communication Manager 6.0 or later, it provides the means to deterministically control the AE Services server connectivity in failover situations.