During our exhibition at FOSDEM, the journalist Ken Fallon of Hacker Public Radio interviewed Ruben to learn more about DRLM Project.
We leave the interview so that you may read:
Can you tell me what this project is and who are you?
Ok, we are BrainUpdaters a Team from Barcelona (Spain).
We started this project because one of our big customers, a global pharma industry called Grifols asked us for a solution to backup their Linux Operating systems, like they was doing with their proprietary Unix systems.
DRLM is a manager for Relax-and-Recover (ReaR) deployments, in order to centrally manage Disaster Recovery for GNU/Linux.
ReaR is a standalone disaster recovery software made by Gratien D´haese among others, he is over here right now.
DRLM will be installed in a server and all ReaR backups will be stored there. The point here is that you can recover any Linux system fully from network.
You have all OS backups on DRLM server, these backups can be scheduled from DRLM. You have a list of clients, networks and backups in a DRLM database. And then If you have to recover those OS backups in the same or other machine, doesn’t matter if is Virtual or Physical, just boot from network and recover.
So, Is this for client machines or we are talking about servers only?
It’s more focused for servers but can be used for client machines also.
Ok, so you run a scheduled job on the server? How are these jobs, full system backups? incremental?
At this moment is a full OS backup, we strongly recommend to separate OS backups and Data backups. I mean, first you will recover the OS at same state since last backup with ReaR/DRLM, and then if you have a database,… you can use an standard backup tool like Bacula o BareOS in order to restore the application data.
We are focused on OS recovery and must have a size as small as possible, 2-5Gb to recover OS as fast as possible. Usually in 5-10 minutes you will have your OS up & running again.
The new server had to have same hardware than old server?
You can recover in other server and is not required exactly the same hardware, but must be same platform.
Now only works on Intel platform, but at this time we are working to support more platforms like PowerPC,…
So, how this works exactly?
Ok, assuming you have a DRLM server, you have to install relax-and-recover in any system you want to take DR backups, and then from DRLM you can run the backups and manage them.
And you can schedule via your system?
Yes, you can do this, also configure how many backups will be stored on DRLM for each client and manage those backups in order to select which stored backups you want to recover.
And what do you hope to get from FOSDEM? Get more Users?
Yes, but not only more users. We wish that the community around the project start growing, we want more people contributing to the project, and this is our main interest coming here.
Listen to the Hacker Public Radio interview here.