Replication in Unity Connection 9 is failing

[ivan aguilar]

Hi

I have a problem with the replication between my CUC Pub and CUC subs, is failing.

In Cluster management from CUC serviceability shows in both server

All services are started and running in both server, from CLI i tried to make "utils dbreplication repaircluster" and utils dbreplication clusterrepair unitysubs9 but said that is not accesible.

With utils dbreplication status says is not accesible.

What can be the cause of the problem, any idea?

thanks!

[Hythim Ali El Hadad]

Hi,

Is it new environment ?

If so, are you sure your security passsword is the same for pub and sub

Regards

Haitham

[Aman Soi]

Hi Ivan,

can u share the output of show cuc cluster status from both servers?

are these servers placed across WAN?

regds,

aman

[ivan aguilar]

Hi

I have put in the pub "utils dbreplication rebulid all" and i am waiting for the result.

There are 3 processes with very high of cpu consume:CiscoDRFMaster, CuMixer and the most RisDC, i am looking for information was exactly do each one, and why make the cpu to be so high.

The two server are phisically together now, only connected by a sw.

The output of subs cuc cluster is

pubs is busy now.

[ivan aguilar]

Hi

Is all new, with a vmware 5 in a UCS520, preparing the installation for the end customer.

Just only need to be licensed but i dont think this could be the cause.

May be could be a performance issue? i have seen performance in both server are very very high, as shows below:

thanks.

[Aman Soi]

Hi ivan,

from the snapshot,I can see Split Brain Recovery when both servers take the role of Primary due to reachability issues.

Let us wait for th eoutput of command , u have executed.

regds,

aman

[ivan aguilar]

hi,

there are 3 services in pub which i activate and start and after a few minutes came back again to desactivate, these are:

Connection Branch Sync Service

Connection File Syncer

Connection System Agent

After the command executed "utils dbreplication rebulid all" nothing has happened, DB is not fine,

Pubs shows

-------------------- utils dbreplication status --------------------
Replication status is not currently available

Use this command to see the output - file view activelog cm/trace/dbl/sdi/ReplicationStatus.2013_10_28_16_13_14.out

thanks

[Aman Soi · Accepted solution]

Hi Ivan,

try the following steps:

1.          utils dbreplication stop all : from publisher

2.          utils dbreplication dropadmindb : on all servers successfully

3.          utils dbreplication reset all

wait for an hour and run the command utils dbreplication runtimestate to check status.

regds,

aman

[ivan aguilar]

Ok, i am going to do it right now and i will show you the results.

thanks!

[ivan aguilar]

Hi

The two first command was executed correctly, but the third one is failing

i tried to increase replication timeout more than 5 minutes but is not allowed.

thanks

[Aman Soi]

Hi Ivan,

run utils dbreplication resetall only on Pub.

regds,

aman

[ivan aguilar]

Hi

After execute commands sayd by almanso, the output of utils dbreplication runtimestate shows

And still in cluster managemente say is unrecheable the other server 

Thanks.

[Aman Soi]

Hi Ivan,

Suggest opening a TAC case.

regds,

aman

[ivan aguilar]

Hi amanso, one more thing.

Just seen in cluster managemt when trying to see from CUC subs the services started in cuc Pubs show this message, this could be the reason of the problem

how can i correct this problem?

thanks.

[ivan aguilar]

Hi

I tried to do these steps https://supportforums.cisco.com/thread/2109854 but still fails, so, i am reinstalling both server again because performance was always so high and i dont know why.

Lets see if with the reinstallation is all ok.

thanks.