Faster backup?

[cwscribner]

Hi all.

Is there a way to speed up the backup process? I started the backup script on a CentOS 5 32-bit server with ~2500 hosts and ~2000 services and its been running for an hour and a half and still hasn't gotten past backing up Nagios Core. I'm not sure if I should be concerned or what. Also, I've tried a manual backup and with an update script which handles a backup; all hang at the backup portion for an inordinate amount of time. What can I do?

[mguthrie]

That stage of the script backs up everything under /usr/local/nagios. It's essentially just creating a tarball of that entire directory. If you've got a lot of performance data that file size could be pretty large. I suppose we should get the easy ones out of the way, any chance you're running out of disk space?

[cwscribner]

Free space is plentiful. I just moved onto a brand new enterprise grade server.

[agriffin]

You could modify the scripts to use a less intensive compression method. I've attached examples to this post which use LZO compression, which is often even faster than no compression at all (because it reduces I/O while using very little CPU time). I haven't tested them, however.

[cwscribner]

Thanks, I'll have to test them internally as this server in question is production. Are there any alternative backup options on the horizon for development? Especially in the enterprise arena, storage space isn't really a problem, so a script that did a 1:1 copy of /usr/local/nagios/ would work just as well. Also, a backup management module in XI would be a big help. Maybe an algorithm to do auto-backups and use a logarithmic function to rotate them. I'm just spit balling at this point so we can consider this closed. I'll explore some backup options and report my findings...if I find anything.