Problem with backup of CDR_CAR

[Roger Kallberg]

Hi,

Since FreeFTPd no longer works reliably for backups of CUCM I've searched for a free replacement and found SilverSHielD. That seems to work good for the UCM backup, but not for the CDR_CAR backup.

For that I get the below error in the DRA backup log.

Creating Tar ball

/common/drf/db_drf_backup/drf_ontape_backup.gz

/bin/tar: -: Wrote only 6144 of 10240 bytes

/bin/tar: Error is not recoverable: exiting now

Tar ball creation failed, unable to tar data to master agent

In the log file for SilverSHeilD I see this error for the failed CDR_CAR backup

SSH_ ERROR _PACKET_SIZE: 11

Anyone with an idea what might cause this and what can be done to solve it?

Please remember to rate helpful responses and identify helpful or correct answers.       

[Roger Kallberg]

No one has any thoughts or ideas about this?


Please remember to rate helpful responses and identify helpful or correct answers.

[V.Naveen Kumar · Accepted solution]

First check ccm free face available not

By using the command show status in CLI mode of the call manager

check free disk of the active cucm.

If it have free space in disk and try to use titan as your sftp server as Cisco recommended.

If it is not have free space delete it.

Complete these steps in order to resolve this issue:

1. Complete these steps in order to clean up the CDR files temporarily so that DRS can proceed:

Stop the CDR Agent service on all servers in the cluster, so that no new CDR files are pushed

to the publisher.

a.

Run this command in order to verify that all files have been pushed to the billing server(s):

file list activelog /cm/cdr_repository/destination/*

b.

Complete these steps in order to verify that there are no symbolic links in any of the subfolders:

a. Stop CDR Repository Manager, CAR Scheduler, and CAR Web Service on the publisher.

Use this command in order to remove all the files under

/var/log/active/cm/cdr_repository/preserve/ that have been accumulated:

file delete activelog /cm/cdr_repository/preserve/* noconfirm

b.

Use this command in order to remove all the symbolic links under

/var/log/active/cm/cdr_repository/car/:

file delete activelog /cm/cdr_repository/car/* noconfirm

c.

d. Restart CDR Repository Manager, CAR Scheduler, and CAR Web Services on the publisher.

2.

Complete these steps in order to stop further accumulation of CDR files:

Note: In order to stop further accumulation of CDR files, you must start the CAR Scheduler service,

set the loader to schedule continuous loading, and load CDR only.

If it is not yet created, create a ccmadmin account in user group management on the

ccmadmin page.

a.

b. Log in to CAR, and go to System > Scheduler > CDR Load.

Check the Continuous Loading 24/7 and the Load CDR only check boxes, and click

Update.

c.

d. Choose System > Database > Configure Automatic Database Purge.

Enter 1 for Min Age of Call Detail Records and Max Age of Call Detail Records, and click

Update.

e.

f. Choose Report Config > Automatic Generation/Alert.

g. For each report, choose Disabled, and click Update.

3.

4. Restart the CDR Agent service on all the servers.

[Roger Kallberg]

Thank you for the reply.
I've checked the used space on the servers and there are only a maximum 61% used across all partitions. So I don't think that is the cause of the problem.

I might add that I didn't have this problem with FreeFTPd, but then I faced the problem that FreeFTPd would crash quite frequent. I know there is a known limitation in this software that makes it crash if there are files in the backup that are larger or equal to 1GB, but none of the files are bigger than 371MB at the moment. So that shouldn't be the reason for the crash.

I'll try Titan, but that will add a cost to the installation, since it's not a free solution as I was searching for in the first place.

Sent from Cisco Technical Support iPhone App

[allan.thomas]

Hi Roger,

Have you looked at the other SFTP daemon Cygwin that Cisco uses for internal use? I have used this and recommended this as an alternative to FreeFTP, although it could be a bit of pain to setup for the correct destination path syntax, rather using the simple "/". Unfortunately I've never come across the error you mentioned before, but I have not specifically used FreeFTP in production and to back up CDRs if enabled. It could be attributed to FreeFTP as you suspect, trying another application will conclusively prove your theory either way, let us know how you get on:

http://www.cisco.com/en/US/docs/voice_ip_comm/cucm/cucos/8_6_1/cucos/iptpch7.html#wp1118964

Regards
Allan

Sent from Cisco Technical Support iPad App

[Jon Clark1]

Roger or anyone,

Did anyone find a resolution?  I'm having backups fail when the CDR/CAR is part of the backup.  Backups run fine when I remove CDR/CAR from the backup.

I'm using solarwinds.  Below is the free space:

Total Free Used
Disk/active 14643056K 1518208K 12976084K (90%)
Disk/inactive 14643120K 1826128K 12668228K (88%)
Disk/logging 50179664K 8326272K 39263216K (83%)

where do I start?