cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
5107
Views
5
Helpful
5
Replies

CUCM - replication status 3 and 0 after upgrade

Hi,

I have a CUCM three node setup.

1 node : Had 5.1.3-8 + we upgrade to 7.0.2a
2 node : Had 5.1.3-8 + we upgrade to 7.0.2a
3 node : We install 7.0.2a from scratch and add it as third node.

At this point everything was fine Database had a 2 state for all node, I did install the locale_FR_FR for 7.0.2 and then restart.

And then I saw that...

THE PROBLEM IS :

NODE 1 : replication status of 3
NODE 2 : replication status of 3
NODE 3 : replication status of 0

I allready restart the 3 server ( starting with publisher )  and still the problem,

any idea ?

5 Replies 5

David Hailey
VIP Alumni
VIP Alumni

Try this first:

In serial fashion (one at a time) - go to the each subscriber and login to the CLI.  After you login, run "utils dbreplication stop".

Once the subscribers are done, go to the publisher and login via CLI.  Run the same command.  Once the command has executed, run "utils dbreplication reset ".

Then repeat on the subscribers one by one.  You'll need to wait a few minutes and then use RTMT or the reporting tool to verify if your issue is resolved (i.e., replication status is 2).


Go here for specifics: http://www.cisco.com/en/US/docs/voice_ip_comm/cucm/trouble/7_0_1/tbsystem.html#wp1172876

You have to clean up the following 4 files.

/etc/hosts,

/etc/services,

sqlhosts and rhosts files.

You need to open a TAC case in order to get root access.

After cleaning those files you need to use "UTILS DBREPLICATION STOP" on each server starting with the 2 subs and the pub.   Then use "UTILS DBREPLICATION CLUSTERREST".

Reboot the pub, wait 15 minutes. reboot sub1, wait 15 minutes and reboot sub2.

MK

To check the Host files enter the commend "run sql select name,nodeid from ProcessNode" on each server to confirm that they know about each other. If not carry out the Utils dbreplication ....

.

kent.roberts
Level 1
Level 1

Had the same problem.  Ours turned out to be NTP setting in CCM was no longer pointing to a valid NTP server.

Once the NTP servers were cleaned up in ccm, waited a few minutes then did a database resync from CLI, and replication status returned to a 2.

I was gone for christmas time, and right now everything is 2 so that's fine.

I took a while to sync, because after 5 hours after upgrade it was still showing a bad status.

Thanks a lot everybody for all the info, I think every info here will be usefull when we will have a database replication problem.

thanks all !

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: