On 11/07/16 18:18, Jessica Otey wrote:
All,
Soon I'm going to reinstall my production instance of robinhood on a
dedicated machine. Currently, I use robinhood to do usage reporting on
our two production lustre systems. Going forward, I would also like to
use robinhood to manage what gets copied from our production lustre to
our backup system (itself another lustre file system).
I am aware that the backup features that robinhood provides are in a
separate rpm.
Hi,
This is no longer the case with v3. All "legacy" robinhood purposes can
now implemented as a single robinhood v3 instance.
I am wondering what the best practice would be in terms of
robinhood to have it do all of the things I want it to do, i.e., monitor
two (maybe all three) lustre filesystems as well as be the policy engine
for the backups. Is one instance of robinhood be the best way to do all
this? Or should I dedicate one to monitoring and one to backups?
The rule is 1 robinhood instance per filesystem. They can run on the
same host, but each will have its own configuration file and DB.
Then your rbh commands will switch between filesystems by specifying a
"-f" option.
Examples:
rbh-report -f yourfs1 ... (will use /etc/robinhood.s/yourfs1.conf)
robinhood -f backup ... (will use /etc/robinhood.s/backup.conf)
...
On RHEL7, there is particular manner of naming systemctl service for
managing multiple instances:
systemctl robinhood@/fsname/ start|stop|status
Regards
Thomas
Thanks,
Jessica
------------------------------------------------------------------------------
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today. http://sdm.link/xeonphi
_______________________________________________
robinhood-support mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/robinhood-support
