Scheduled bakcup - empty files

[BigWeb.EU]

Hi,

Kloxo 7. (updated in August) .

Manual bakcup would work just fine, however scheduled backup would create a gz file with two file sinside :
kloxo.file and
kloxo.metadata

No other files inside .

/usr/local/lxlab/kloxo/log shows no obvious error (Also rotated logs in gz format are also emty files ! )

Any hint on what could it be ?

[MRatWork]

inform here the content of backup inside /usr/local/lxlab/kloxo/log

[BigWeb.EU]

Oh, actually manual backup fails as well, with

 program_interrupted

backup file has these records (I have shortened it, but it lists my domains , email, ftp, and then ends):

Code: [Select]

21:15 Oct/05/2015: *** Backup from 'admin' user - BEGIN ***
21:15 Oct/05/2015: Setting parent of 'mysqldb-urlshort' to 'client-admin'
21:15 Oct/05/2015: Setting parent of 'domain-uat.mysite.com' to 'client-admin'
21:15 Oct/05/2015: Setting parent of 'mailaccount-postmaster@uat.mysite.com' to 'mmail-uat.mysite.com'
21:15 Oct/05/2015: Setting parent of 'dnstemplate-mysite.com.dnst' to 'client-admin'
21:15 Oct/05/2015: Setting parent of 'ftpuser-admin' to 'client-admin'
21:15 Oct/05/2015: Setting parent of 'sslcert-mysite' to 'client-admin'
21:15 Oct/05/2015: Setting parent of 'ticket-1' to 'client-admin'
21:15 Oct/05/2015: Setting parent of 'tickethistory-1___0' to 'ticket-1'
21:15 Oct/05/2015: Taking backup of 'client:admin'


[BigWeb.EU]

Ok, scheduled backups generates more line in logs :

Code: [Select]

21:15 Oct/05/2015: Taking backup of 'client:admin'
21:48 Oct/05/2015: - Alert: Failed backup of 'client:admin'
21:48 Oct/05/2015: * Kloxo Backup on 2015-Oct-05 at 21 Hours - Kloxo Backup Succeeded for 'admin' to 'BACKUP_SERVER_IP'
21:48 Oct/05/2015: *** Backup from 'admin' user - END ***


[BigWeb.EU]

More info :

I can see backup file is being generated in /usr/local/lxlabs/kloxo/serverfile/tmp (it generates a proper big backup of , and it looks like it is all fine with the archive , but it never gets to be compressed to final gzip .

also in shell exec theres records:

Code: [Select]

21:09 Oct/06/2015: 1: tar: ONEOFMYSITES/var/log/system.log: file changed as we read it [(__system__) nice -n +10 ionice -c3 tar -czf /usr/local/lxlabs/kloxo/serverfile/tmp/adminR4ZKhr.tmp '.bash_logout' '.bash_profile' '.bashrc' '.pki'
21:09 Oct/06/2015: Closed Exec /usr/local/lxlabs/kloxo/cexe/closeallinput 'sh /script/load-wrapper >/dev/null 2>&1 &' >/dev/null 2>&1 &
21:09 Oct/06/2015: 0: (No message) [(__system__) nice -n +10 ionice -c3 tar -czf /home/kloxo/client/admin/__backup/test33-7.0-admin-2015-Oct-06-1444159440.tgz 'kloxo.file' 'kloxo.metadata' --ignore-failed-read]

Any hints where i should look into ? Debug ?

PS - i created a new client - and backup for a client looks like doing correctly , but for an admin - it still failing (Data of backup - 45GB , time to do ~45-60minutes) Maybe any kind of timeouts, etc  ?

[BigWeb.EU]

And another update - , i tried running backup from SSH , and got this message at the end :

Code: [Select]

Taking backup of 'client:admin'
Backup failed due to: could_not_zip_dir: : /home/admin/

Still need advise ...

[MRatWork]

Maybe missing something missing in admin. Possible delete/erase domain directly.

Inform here 'admin > all list > all domains'.

[BigWeb.EU]

Hi - found the problem :

all happens because of
"21:09 Oct/06/2015: 1: tar: ONEOFMYSITES/var/log/system.log: file changed as we read it..."

the system log is outputing all notices errors , and as traffic is VERY big the file is written constantly . Once i stop logging to that file - backup works fine .

Is it something that can be fixed somehow ?

[medo.mine1]

I've the same problem for all clients
how to fix that?
Thanks