DB Replication Issue with CUCM 8.5

[mohamed helmy]

Hello,

I have a cluster with two CUCM 8.5. the DB replication went badly Pub status "0" and Sub status "3". I tried almost all the commands to fix the replication ( utils dbclusterrest, utils dbrepair, utils dbdropmain, etc....) also i tried to restart both of them in squence Pub first then the Sub.

After all these troubleshooting, the replication went ok and good between the two CUCM as any change done at the Pub, replicats to the Sub, but it was the same status (Pub 0, Sub 3).

After I restarted the service "utils service restart cisco database layer monitor", the status went to "0" in both Pub and Sub.

The replication is ok till now, but it is weird why not to be status "2"

So, any ideas??

Regards,
Mohamed Helmy       

[Kolesnik Sergey]

Hi. 0 = Not started, the value on each node MUST equal 2. Can you show the output of the Debug Replication report?

Regards, Sergey.

[mohamed helmy]

Hello Sergey,

Thanks for your reply. please find attached the Debug Replication report

for Pub and Sub.

Regards,
Mohamed Helmy       

[mohamed helmy]

Hello,

I solved the problem.

reinstall the sub and do the following commands in order:

•   Utlis dbreplicastion dropmainforce
•   Utlis dbreplication stop “SUB”
•   Utlis dbreplication clusterrest
•   Utlis dbreplication rest all

it fixed the problem and now the replication is "2"

Regards,
Mohamed Helmy

[Harmit Singh · Accepted solution]

Hi Mohamed,

Thank you for sharing the solution you followed (+5)! I am curious to know why you went with a rebuild of the Subscriber though. It seems quite an extreme step to take, unless it was at a point of no return. Also some of the commands you mentioned needn't have been run, not to mention the order they were run in. To troubleshoot dbreplication, it's very important to understand where replication is actually hosed up and run only the commands needed to sort that bit out. When you do get a chance, please review my blog, video and also the troubleshooting doc for dbreplication:

https://supportforums.cisco.com/community/netpro/collaboration-voice-video/ip-telephony/blog/2012/10/26/understanding-cucm-dbreplication-runtimestate

https://supportforums.cisco.com/docs/DOC-13672

If you have any questions, do let me know and I'll do what I can to help clarify your doubts.

--
Regards,
Harmit.

[mohamed helmy]

Hello Harmit,

Actually those two links were my guide through this case, and they are really helpful.

as I mentioned up here, the replication seemed to be ok in the cluster "any change at pub, reflected immediately to the Sub", however the status were (Pub 0, Sub 3).

After many troubleshooting steps, no effective result happened. I removed the server from the cluster then added it again, restarted the cluster, but no luck

so I aimed to rebuild the database replication from the beginning by reinstall the Sub.

finally it worked, but after alot of pain

Regards,
Mohamed Helmy

[Harmit Singh]

Hi Mohamed,

Ah I see, sounds like it was quite a task to get things working again, but glad to know it finally did. All's well that end's well :-) If you do run into something similar again, let me know and we can discuss in more detail to see if we can make the process easier to troubleshoot. See you around!

--
Regards,
Harmit.