CUCM 12.5 with PUB+SUB and add another SUB

[Jose Carlos Silva]

I have a CUCM 12.5 with PUB+SUB coresident and add another SUB on a new server.

Everything if running fine - sets registering and making calls, gateways, et,etc, on second SUB. I would like that, in case of fail of the first server where the PUB+SUB reside and TFTP that the second SUB gave the TFTP service. My issue is that I cant upload files to this SUB when I go to the OS administration and try to Install/Upgrade to install de device packs, background logos, device language, etc. I'm using PCD as the SFTP server deposit and the main server sees all the files but when I do the same on the second SUB he gives an error:

"Connectivity Issue to First Node. Verify that the first node is powered on, the network connection is up and the securityCUCM 12.5, Unified Communications password on this node and the first node are the same. You might be also seeing this message due to the installation of the free common space COP file on the publisher. If the intent was to set the "use publisher setting" to "yes/no" in order to facilitate a cluster wide upgrade, then you may ignore this message. ".

TFTP service is active and running on this second SUB. Is there any step missing ot its not suported and I need to deploy a machine, only, with the TFTP service?

Can you help me or point me to the right direction?

Best Regards,

[Nithin Eluvathingal]

its not clear what you are trying to achieve ?

Are you installing a new subscriber node ? And during the installation, are you facing any issues ?

or are you facing any issues when installing the COP files on second subscriber node ?

[Jose Carlos Silva]

Sorry.

I already installed the second subscriber node; he is working and with registered sets and gateways.

Just want to activate the TFTP service on this sub to be a backup of the main one installed on the first node.

My issue is when I try to install/upgrade the cop files with the device firmware and languages on this second TFTP server I receive the error I mentioned.

Hope I'm clear, now

[Elliot Dierksen]

Are the COP files you are trying to install already installed on the publisher? Check a "show version active" on the publisher and existing subscriber. I don't know that it blocks the install if it isn't on the publisher, but it is definitely best practice to install those sorts of thing on the publisher first and subscriber(s) afterwards.

[Jose Carlos Silva]

Thanks for your answer.

Yes, the publisher have all the files and firmware needed for the phone and gateways to work. Just need to have a backup and I read that each cluster can have until 3 tftp servers; but can't upload to this second one.

Best Regards

[Elliot Dierksen]

There has to be something else going on. I have multiple clusters with 4+ nodes in them without any issue upload devpack or firmware files.

[Jose Carlos Silva]

I believe you, but this is only one cluster with one PUB+SUB and another SUB; I already saw configurations to add alternative TFTP server on another cluster (remote cluster), but on the same cluster I didnt saw any configuration I need to do.

Thanks for your answer.

[Elliot Dierksen]

At the CLI on the publisher, do a "utils dbreplication runtimestate" and see if replication is working properly. You might need to do a "utils dbreplication status" to force a check. You then monitor the status of the check with "utils dbreplication runtimestate". I would also do a "utils diagnose test" on each node to make sure that DNS is correct and there aren't other issues.

[Jose Carlos Silva]

Hi!!

Made the commands you asked but still cant upload files to install the device packs and language packs because give the same error "Connectivity Issue to First Node. Verify that the first node is powered on, the network connection is up and the security password on this node and the first node are the same. You might be also seeing this message due to the installation of the free common space COP file on the publisher. If the intent was to set the "use publisher setting" to "yes/no" in order to facilitate a cluster wide upgrade, then you may ignore this message.". 

Conectivity is ok, passwords are ok.

utils dbreplication runtimestate

Server Time: Tue May 17 12:20:26 WEST 2022

Cluster Replication State: BROADCAST SYNC ended at: 2022-04-11-14-56
Sync Result: SYNC COMPLETED on 732 tables out of 732
Sync Status: All Tables are in sync
Use CLI to see detail: 'file view activelog cm/trace/dbl/20220411_145359_db l_repl_output_Broadcast.log'

DB Version: ccm12_5_1_10000_22

Repltimeout set to: 300s
PROCESS option set to: 1

Cluster Detailed View from hla-bilbo-01.hla.local (2 Servers):

PING DB/RPC/ REPL. Replication REPLICATION SETUP
SERVER-NAME IP ADDRESS (msec) DbMon? QUEUE Group ID (RTMT) & Details
----------- ---------- ------ ------- ----- ----------- ------------------
hla-bilbo-01 172.17.xx.yyy 0.016 Y/Y/Y 0 (g_2) (2) Setup Completed
HLA-Subscriber-02 172.17.xx.zzz 0.347 Y/Y/Y 0 (g_3) (2) Setup Completed

utils dbreplication status

Replication status check is now running in background.
Use command 'utils dbreplication runtimestate' to check its progress

The final output will be in file cm/trace/dbl/sdi/ReplicationStatus.2022_05_17_12_22_04.out

Please use "file view activelog cm/trace/dbl/sdi/ReplicationStatus.2022_05_17_12_22_04.out " command to see the output
admin:utils dbreplication runtimestate

Server Time: Tue May 17 12:22:36 WEST 2022

Cluster Replication State: Replication status command started at: 2022-05-17-12-22
Replication status command in PROGRESS. Checked 150 tables out of 732
Last Completed Table: functionrole
No Errors or Mismatches found.

Use 'file view activelog cm/trace/dbl/sdi/ReplicationStatus.2022_05_17_12_22_04.out' to see the details

DB Version: ccm12_5_1_10000_22

Repltimeout set to: 300s
PROCESS option set to: 1

Cluster Detailed View from hla-bilbo-01.hla.local (2 Servers):

PING DB/RPC/ REPL. Replication REPLICATION SETUP
SERVER-NAME IP ADDRESS (msec) DbMon? QUEUE Group ID (RTMT) & Details
----------- ---------- ------ ------- ----- ----------- ------------------
hla-bilbo-01 172.17.xx.xxx 0.016 Y/Y/Y 0 (g_2) (2) Setup Completed
HLA-Subscriber-02 172.17.xx.zzz 0.341 Y/Y/Y 0 (g_3) (2) Setup Completed

[Roger Kallberg]

There are AFAIKT only 2 servers in your replication list and in your posts you state that you have 3.

So your cluster configuration is not in an okay state as you should see three servers in this list. How did you go about when you added the second subscriber node? Can you please outline the process you followed, in as much details as you can share?

[Jose Carlos Silva]

I wrote when I stared this post:

"I have a CUCM 12.5 with PUB+SUB coresident and add another SUB on a new server."

So my PUB and SUB1 are on the same server and now I have a second SUB for redundancy; PUB+SUB have only one IP address and another IP to the second SUB2 on another phisical server.

Hope is clear now. 

[Elliot Dierksen]

---

@Jose Carlos Silva wrote:

PUB+SUB have only one IP address and another IP to the second SUB2 on another phisical server.

They might be on the same ESXi host, but each CUCM virtual machine must have a unique IP. I don't think it is possible for things to be the way you describe that.

[Jose Carlos Silva]

So what youre saying is that PUB must be one VM, SUB1 another VM and SUB2 another VM, right?

Having PUB and SUB running on the same VM does not suport have a seconf TFTP server on the second SUB!!!

Is that right?

[Elliot Dierksen]

A VM is a publisher (in the end there can be only one) or a subscriber. Totally binary situation.

[Jose Carlos Silva]

Ok, thanks.

Didn't read on the documentation that this is a problem, but ok. This customer started with a simple solution and now wants to add some redundancy.

Thanks for you help; in the last all the phones work because all the config are in the device and dont need to upgrade or load nothing to work, only switch to the available subscriber.

Best Regards