Hi all, I don't understand how the cluster's storage can be expandable on the fly (hot swap?) without affecting current data and usage. We have a file server with 32 swapable trays, and we can configure the disks with any RAID we want. My understand is that for this storage to be available on the cluster we have to do the following steps:
1. put disks in enclosures (swap if any disk fails etc...) 2. configure raid to have a disk volume on the file server 3. format disk volume to be usable in the file server 4. mount the disk volume 5. share it to the cluster I know that we can hot-swap any failed disk in the first step, and the 2nd step can be done automatically by the RAID software, but the admin still has to finish the next three steps to make the new storage available. This process then can affect the whole disk volume created before. So, let say if we have 30TB on a file server shared to the cluster, and the cluster uses them all, and we want to add 10TB more into that, how would we do? Please advise us. Thanks, D. _______________________________________________ Beowulf mailing list, [email protected] sponsored by Penguin Computing To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
