On Thu, 30 Jul 2009, Anton VG wrote: > Hello friends! Just watching for a while for the CEPH progress, and see 0.10 > have quite promising description. Do you think I already can try 0.10 > for non-critical > file storage (though would be not good to lose that storage ;) or anyway you > do > expect files corruption, etc with current code? > In short could be current CEPH be used somehow or not yet?
If you don't want to lose data, I would wait for another release or two. I'm currently filling up a ~100TB cluster for some larger-scale testing and am still running into occasional problems. If you're in a situation where you can tolerate some data loss (e.g., computation cluster, temporary storage, logs, tertiary backups, testing, etc.), I would very much welcome the testing and feedback. For critical data, it will still be a while. Objects are stored in btrfs, which is coming along but still isn't completely solid either. My general plan is to harden incrementally: first the object store, then single-mds, then clustered mds. The object store can be used independently as an S3-like object storage cloud via an object based api (librados). sage ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july _______________________________________________ Ceph-devel mailing list Ceph-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ceph-devel
