CUCM backups not being completed - stuck at final step "Copying backup TAR file to archive media"

[Liv_Liv]

Hi, I have the following setup

CUCM PUB + SUB System version: 11.5.1.15900-18

Everytime a backup is attempted (no matter if its scheduled or manual), when all components are backed up and each one of them show "100% SUCCESS", the process get stuck in CUCM PUB with this progress message: "Copying backup TAR file to archive media"

DRF Master & Local services traces gathered via RTMT didn´t show any clear error message that can be potentially related. 

The backup process keeps getting stuck at its final step with this progress message: "Copying backup TAR file to archive media" endlessly, until we manually restart DRF services. 

Things that I´ve done so far with no positive results at all:

- restarted Tomcat service on both nodes

- restarted DRF Master service + DRF Local service on CUCM PUB

- restarted DRF Local service on CUCM SUB

- checked no expired ipsec & tomcat / trust certs on both CUCM PUB & SUB

- deleted old backup files from SFTP server

- tried another different destination server and backup location - same results

- tried two different SFTP software applications: freeFTPd & WinSCP  - same results (no connection timeout configured on none of them)

- checked dbreplication between CUCM PUB & SUB - all good

Does anybody have any clue on how to solve this?

thank you, 

[Nithin Eluvathingal]

Can you try different SFTP application and see the behavior ?

[Liv_Liv]

Hi Nithin, thanks for your reply, 

yes I already tried freeFTPd & WinSCP with same result. 

Any other advise?

thank you, 

[Nithin Eluvathingal]

Freeftpd is not recommends and it has many know issues.

Try  another SFTP server.

Is there any firewall between the backup target and CUCM?
 

[Liv_Liv]

Hi Nithin, 

Freeftpd was only used for testing purposes and behaviour was exactly the same than earlier with main WinSCP server already in place. 

There is no firewall between backup servers tried so far and CUCM servers. No connection errors spotted on the logs.

IMHO DRF Master or Local agent service in the PUB hungs just when the backup is finished and I can´t find out why. 

There is a bug for the exact same issue we are facing, but we tried several times the workaround suggested with no results: https://bst.cloudapps.cisco.com/bugsearch/bug/CSCsq50512

thanks,

[Nithin Eluvathingal]

That bug ID is related to older version.

I suggest to contact  TAC.

[Sadav Ansari]

Change the path directory and check.

Pls rate if its “Helpful”. If this answered your question please click “Accept as Solution”.

[Liv_Liv]

Did that already as I tried with two different target servers and locations
with different paths, no results

[Elliot Dierksen]

DRS is rather picky about the ciphers and key exchange algorithms. Anything with an sshd_config can usually be made to work by adding these commands.

KexAlgorithms +diffie-hellman-group1-sha1
KexAlgorithms +diffie-hellman-group-exchange-sha1

Ciphers +aes128-cbc
Ciphers +3des-cbc

Some newer servers don't like the 3des cipher, so you might need to drop that if your ssh server groks on that. A number of newer windows versions allow you to add "OpenSSH server" as an optional component, and I have had success using that as a DRS target. I have also had success using the OpenSSH server component of Cygwin.

[Liv_Liv · Accepted solution]

Hi all, ended up opening TAC case. 

Problem was I was hitting bug CSCuu46430 related to DRF process creating logs without purging them, ence not finishing properly. 

https://bst.cloudapps.cisco.com/bugsearch/bug/CSCuu46430/?rfs=iqvred

Thing is, the only way to manually delete those logs is having root access to the CUCM OS, so TAC needs to be involved - no chance to do this on your own.

Hope this info will be valuable to somebody someday.

Thanks all for your help !!!