October 7, 20187 yr Hi, on the pricing page, it says the unlimited (??) license only supports 28 data and 2 parity drives. I want to know if this info is outdated or it is true. What if I need more storage. Could you raise the limit to a greater limit or make it truly unlimited in a future update if it is true?
October 7, 20187 yr It is unlimited. There is no limit on the amount of data. The practical matter of needing more than 336TB of storage (which is the practical limit of the system using retail drives at this time), is a completely different story. If you think you need more storage than that, perhaps you should re-examine your planned usage and the appropriateness of the tool in question?
October 7, 20187 yr Our md/unraid driver (parity protected array) supports a single array of up to 28 'data' devices and 0, 1, or 2 'parity' devices. We support a single btrfs-formatted 'cache pool' of up to 24 devices. If you have more than 54 storage devices, or don't want to assign them to either the parity-protected array or the cache pool, you can make use of them using the Unassigned Devices plugin, or other ad-hoc method, e.g., assigning to virtual machines. It remains on the "future featues" list to increase the number of parity-protected arrays and cache pools. Not likely we would increase the width of either however.
October 8, 20187 yr 15 hours ago, limetech said: Our md/unraid driver (parity protected array) supports a single array of up to 28 'data' devices and 0, 1, or 2 'parity' devices. We support a single btrfs-formatted 'cache pool' of up to 24 devices. If you have more than 54 storage devices, or don't want to assign them to either the parity-protected array or the cache pool, you can make use of them using the Unassigned Devices plugin, or other ad-hoc method, e.g., assigning to virtual machines. It remains on the "future featues" list to increase the number of parity-protected arrays and cache pools. Not likely we would increase the width of either however. I'm nowhere close to hitting any limits but is there a reason you have limited the number of drives on a single parity protected array to a specific number, how did you come up with that number?
October 8, 20187 yr Community Expert 19 hours ago, limetech said: It remains on the "future featues" list to increase the number of parity-protected arrays For me this would be even better. Probably too soon to ask, but in that case would the multiple arrays be accessible under /mnt/user , e.g., using share rules to say which share goes where or would they be under separate mount points? I would vote for all on /mnt/user if possible.
October 8, 20187 yr Community Expert 52 minutes ago, johnnie.black said: I would vote for all on /mnt/user if possible. Ooh, but with that said, I'd still want a way to access each array individually, e.g. /mnt/array1, and disks could be under /mnt/arrayX/diskY
October 8, 20187 yr 6 hours ago, Koshy said: I'm nowhere close to hitting any limits but is there a reason you have limited the number of drives on a single parity protected array to a specific number, how did you come up with that number? Bigger coding effort to go beyond 30.
October 8, 20187 yr 3 hours ago, limetech said: Bigger coding effort to go beyond 30. No time to explain it, that's ok. I was just curious.
October 9, 20187 yr Although I've read it's not supported, and those who have done it I think faced some complexity, you could create multiple unraid vm's on the same hardware as your bare metal unraid install. I guess you'd only be limited by your hardware capacity at that point. Am I wrong? Edited October 9, 20187 yr by TSM
October 9, 20187 yr 7 hours ago, TSM said: Although I've read it's not supported, and those who have done it I think faced some complexity, you could create multiple unraid vm's on the same hardware as your bare metal unraid install. I guess you'd only be limited by your hardware capacity at that point. Am I wrong? Not wrong, but not the point. It would be the equivalent of running multiple bare metal installs, so you would need a license USB key for each VM, and merging the different VM's file systems into single share points would be either painful or impossible depending on exactly what you are trying to accomplish. Management would be of each individual VM, so basic config changes you made in one would not automatically propagate. You would be dealing with either a 3rd party VM manager (esxi) or running unraid as a VM inside unraid, either way it's a support nightmare, and not something limetech would consider endorsing. Yes, you could do it, but your headache factor would increase exponentially. "faced some complexity" is not exactly how I would describe it.
Archived
This topic is now archived and is closed to further replies.