issues faced when publisher down

Unanswered Question
Jun 28th, 2010
User Badges:

Publisher goes down in my cluster of One Publisher and One subscriber (ver 4.3)....


Following are the issues that i face :


- Extension mobility stops working

- Call Center agents are unable to login to CAD

-Fast busy tone is heard on calling the UAN (pri) terminated on MGCP gateway


Although subscriber was up during this time. What are the possible reasons and how to resolve the issue.

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 4.8 (3 ratings)
Loading.
SJ SJ Mon, 06/28/2010 - 05:33
User Badges:

dont find any replication issues.

Md. Mijanur Rahman Mon, 06/28/2010 - 04:50
User Badges:
  • Silver, 250 points or more

Hi, seems you have dbreplication issue, your SUB might not relicating with the PUB.

What are the status of the phones? are the registered correctly when PUB is down? Can you use CFwdALL button to forward all calls to another phone?

Please follow the post just mentioned here to check the dbreplication issue:

https://supportforums.cisco.com/message/1139615?tstart=0


Also you can follow this link for the procedured to check dbreplication on CCM 4.x:

http://www.cisco.com/en/US/products/sw/voicesw/ps556/products_tech_note09186a00801e7ddf.shtml


HTH.

Rob Huffman Mon, 06/28/2010 - 05:59
User Badges:
  • Super Blue, 32500 points or more
  • Hall of Fame,

    Founding Member

  • Cisco Designated VIP,

    2017 IP Telephony, Unified Communications

Hi SJ,


For #1


You would need to go to CCM 6.x to have redundancy for EM. Here is the background (there is no way to do this on any CCM version until 6.x)


The configuration database is stored on a publisher server, and a read-only copy is replicated to the subscriber members of the cluster. Changes made on the publisher are communicated to the subscriber databases, ensuring that the configuration is consistent across the members of the cluster, as well as facilitating spatial redundancy of the database.


The publisher server is the only server that has read and write access to the configuration database. When configuration changes are made, other members of the cluster have a read-only copy of the database.


From this CCM SRND;


http://www.cisco.com/en/US/docs/voice_ip_comm/cucm/srnd/4x/42clproc.html






This changes with the release of CCM 6.x;




In prior versions of Cisco Unified Communications Manager, subscriber servers in the cluster use the publisher database for READ/WRITE access, and only use the local database for READ access when the publisher database cannot be reached. With Cisco Unified Communications Manager Release 6.0, subscriber servers in the cluster READ the local database. DB WRITES happens in both the local database as well as the publisher database, depending on the type of data. DBMS (IDS) replication is used to synchronize the databases on the nodes of the cluster. When recovering from a failover conditions such as loss of WAN connectivity for extended period of time, the Cisco Unified Communications Manager databases need to be synchronized with any changes that may have been made during the outage. This process happens automatically when database connectivity gets restored. This process may take longer over low bandwidth and/or higher delay links.


Database modifications for CallProcessing




User Facing features can be made on subscribers. These include updates for:




Call Forward All (CFA)


Message Waiting Indication (MWI)


Privacy Enable/Disable


Do Not Disturb Enable/Disable (DND)


Extension Mobility Login (EM)


Monitor (for future use, currently no updates at the user level)


Hunt Group Logout


Device Mobility


CTI CAPF status for end users and application users


Credential hacking and authentication




From this 6.x SRND;




http://www.cisco.com/en/US/docs/voice_ip_comm/cucm/srnd/6x/callpros.pdf





For #3


Cisco IOS gateways can maintain links to up to **two backup Cisco Unified CallManager servers in addition to a **primary Cisco Unified CallManager. This redundancy enables a voice gateway to switchover to a backup if the gateway loses communication with the primary. The backup server takes control of the devices that are registered with the primary Cisco Unified CallManager. The second backup takes control of the registered devices if both the primary and first backup Cisco Unified CallManager fail. The gateway preserves existing connections during a switchover to a backup Cisco Unified CallManager.




When the primary Cisco Unified CallManager server becomes available again, control reverts to that server. Reverting to the primary server can occur immediately, after a configurable amount of time, or only when all connected sessions are released.




Switchback




Switchback is the process a voice gateway uses to reestablish communication with the primary Cisco Unified CallManager server when the server becomes available again. Switchback can occur immediately, at a specified time after the last active call ends, or after a specified length of time.




MGCP Gateway Fallback




The MGCP gateway maintains a remote connection to a centralized Cisco Unified CallManager cluster by sending MGCP keepalive messages to the Cisco Unified CallManager server at 15-second intervals. If the active Cisco Unified CallManager server fails to acknowledge receipt of the keepalive message within 30 seconds, the gateway attempts to switch over to the next available Cisco Unified CallManager server.








SUMMARY STEPS




1. enable




2. configure terminal




3. ccm-manager redundant-host {ip-address | DNS-name} {ip-address | DNS-name}




4. ccm-manager switchback {graceful | immediate | schedule-time hh:mm | uptime-delay minutes}




5. ccm-manager fallback-mgcp




6. call application alternate




7. exit




8. ccm-manager switchover-to-backup


Cheers!
Rob




Please support CSC Helps Haiti


https://supportforums.cisco.com/docs/DOC-8895


https://supportforums.cisco.com/docs/DOC-8727

SJ SJ Mon, 06/28/2010 - 21:57
User Badges:

Thnx for such a detailed answer.


This means i am using call manager 4.X and if the publisher goes down there is no possiblity of fail over for extension mobility.

Actions

This Discussion