Backup/restore indexes to offsite cluster?

[gsl_ops_practice]

Hello,

I would like to find out if it is possible to back up indexes from a running prod NLS cluster and restore on a different cluster? The prod cluster we are looking to deploy has a disk space limitation so we have to set the data retention to 7 days max. However, we need this data for far longer, so the idea is to back up indexes to another offlne location and restore, and keep for as long as we need.

Can you please advise if this is doable, and if so, what should the process be?

Thank you,
Alex

[cdienger]

Hi Alex,

1. On the offline machine, make a new directory to move the backups to. If this is a cluster, make the repo on all machines in the cluster.
2. On the offline machine, make a new repo under Administration > System > Backup & Maintenance > Repositories, pointing to the directory created in the first step.
3. Copy the repo directory from the prod machine over to the offline machine's repo directory.
4. Set the ownership with:

Code: Select all

chown -R nagios:users *repo*

You should then see the backups available on the offline machine under Administration > System > Backup & Maintenance > Repositories with the option to then restore them.

[gsl_ops_practice]

That makes sense, thank you.

It sounds like we would be copying the live index as well, is that correct? If we are doing this on a daily basis, is there any other way to do it? I am thinking rsync, it will just pick up the changes and not copy what's already there...

What do you think?

[cdienger]

That should work. You probably want to script some logic to make sure the curator process isn't running before the sync(ps aux | grep curator). You can control the time and frequency of the backup under Administration > System > Command Subsystem.

[gsl_ops_practice]

Hello,

Can you please help me understand what the curator does? We would like to have this automated so that the NagiosLog getting live data doesn't have to undergo any kind of scheduled maintance, as this is for a 24x7 transactional system.

Thank you.

[cdienger]

It's used to execute all the settings on the Backup & Maintenance page - removing indexes, making snapshots, optimizing, etc...