UCCX 7.01 with cold standby server

Unanswered Question
Apr 15th, 2009
User Badges:
  • Bronze, 100 points or more

Hello,


Im designing a solution of Disaster recovery with a pair of UCCX 7.01 and another pair of clustered CUCM7.


I have one UCCX server in one site and would like to have another one in other site, geographically distinct. The second server only works in case of failure of the first.(with manual actions if needed).


Im using this scenario because of the limitations of the HA UCCX solution.


My doubts are:


- what is the best way to keep the cold standby server updated? Can i have the server instaled and periodically update it. If so what do i need to update?


- what is the best way to point CUCM configuration to second UCCX server in case of failure of the first UCCX server?


thanks for your help in advance,

Met.

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 5 (2 ratings)
Loading.
ra_jeshkalra_2 Wed, 04/15/2009 - 19:30
User Badges:

1)If one of UCCX server is cold standby, then you can only do update through backup & restore , manually(refer UCCX SRND cold standby section).But you have to do this for the site where you want to make the UCCX server up, to keep the DBs syncronised.


2)The soultion should have one CCM cluster with two CCM servers.So in site A you will have one CCM server & site B you have one CCM server and 2nd UCCX server in cold standby mode.The site A & site B UCCX servers should have same IP address.


The UCXX server will point to cluster servers as CTI managers, and will automatically point to second CCM if one goes down.


Note>>You can't have more than two CCM servers in cluster with UCCX, since by design UCCX cluster can't point to more than two CCM servers(CTI Managers)


Thanks

metalium2007 Thu, 04/16/2009 - 03:17
User Badges:
  • Bronze, 100 points or more

Thanks for your response! (+5)

The CCM questions are all ok, no problem.


I would like to know if with the backup, restore all the databases are in synch (including historical reports), or if i can do daily replication of the database and it will still works (configuration and historic), all updated!


Thanks,

Met.

mwibowo Mon, 04/27/2009 - 20:21
User Badges:

1) What happen if Site A have 3 CUCM servers (1 Publisher, 1 TFTP, 1 Subscriber) and Site B have 2 CUCM servers (1 TFTP and 1 Subscriber)?


2) Will the design supported by Cisco if we put the primary UCCX at Site A and the UCCX HA server at Site B? FYI, Site A and Site B is connected over MPLS with QoS. The WAN bandwidth is 45 Mbps (DS3) with round-trip delay of 2 milliseconds.


3) Will there be any issue with database replication between primary UCCX server with its HA server in above scenario?


4) Will there be any impact on the existing call when the primary UCCX server fail on above scenario?


Thank you.

Jonathan Schulenberg Tue, 04/28/2009 - 17:27
User Badges:
  • Super Bronze, 10000 points or more
  • Cisco Designated VIP,

    2017 IP Telephony

1) CCX is capable of communicating with only two CTI Manager nodes. CCX will not communicate with the other UCM nodes regardless of how many you have.

1a) Understand that CCX can also be configured for only one UCM node for AXL integration. This is typically the first node.


2) The SRND states "Cisco Unified CCX high availability requires that the Cisco Unified CCX Engine and Database

components and the CTI Managers with which the Cisco Unified CCX servers communicate be located

in the same campus LAN and that the maximum round-trip delay between these servers be less than 2 ms."

Even if the RTT is under 2ms, you are not supported to have the CCX HA server OR ANY of the UCM CTI Managers at another site.


3) Not supported at this time. Keep an eye out for the release notes and SRND of CCX 8 when it is released (2010).


4) If the call is not connected to an agent on their IP phone, it will be dropped.

mwibowo Tue, 04/28/2009 - 18:13
User Badges:

Hmm...if both HA server and primary server must be in the same physical location, then I think it defeats the purpose of having HA server for Disaster Recovery (DR) scenario. In the early stage I thought of putting the HA server at DR site so when the primary server fail, the call center won't be affected. Hope Cisco R&D team can introduce this new features in their new release 2010?


Thank you for your reply. Highly appreciate it!


Actions

This Discussion