Hi all, we have been a part of the Drobo family since the Generation 1 systems and one of our group companies has even been a Drobo channel partner in the Caribbean market over a decade ago.
We still have a generation 2 DAS in full operation (albeit quite slow operation by today’s standards) here in our Dutch office and have recently acquired a Drobo 8D as an upgrade.
While we are aware that this unit was built for MacOS we have seen the compatibility issues over the years with MacOS and in addition we are Linux users in-house.
We reached out to Drobo support before making the purchase to find out if the unit could be unofficially supported under Linux, if we simply formatted it with the ext3 fs. We were informed that the hardware, like other Drobos could support ext3 and that this could work but would not be supported by Drobo Inc.
We purchased a unit (after waiting months to find the first one back in stock), popped 3 18TB drives into it and set it up via the dashboard with dual disk redundancy under MacOS Monterey. Then we cut 16TB volumes into it (the largest that can be supported by ext3 with a 4K blocksize) and formatted them in ext3.
Under Linux we can only see:
XX:XX.X PCI bridge: Intel Corporation JHL6540 Thunderbolt 3 Bridge (C step) [Alpine Ridge 4C 2016] (rev 02) XX:XX.X PCI bridge: Intel Corporation JHL6540 Thunderbolt 3 Bridge (C step) [Alpine Ridge 4C 2016] (rev 02) XX:XX.X PCI bridge: Intel Corporation JHL6540 Thunderbolt 3 Bridge (C step) [Alpine Ridge 4C 2016] (rev 02) XX:XX.X SCSI storage controller: Device 4452:3844 (rev 08)
but no volumes. So no, despite what we were told, it won’t work under Linux as a mass storage device out of the box.
We would like to see this unit supported directly via Linux and have already reached out to the maintainer of the drobo-utils tools for Linux, in a first attempt to get it working.
So far it looks like we will need the help from Drobo Inc for documentation and if possible source code for the drivers or we will have to go the path of reverse engineering which will take considerably more time than the team can spare at present.
Is there anyone here who has already looked into this and would be interested in collaboration on such a project?
Would the Drobo engineering team be willing to provide a bit of informal assistance to us to get support added to the drobo-utils suite?
We have always thought that Drobo devices are brilliant and would love to continue to work with them. However, being forced to use the fastest and most modern DAS model on MacOS is not ideal for our use-case and without other options available in the Drobo family, this may leave us with no other choice but to look for alternatives.
Open for collaboration and seeking help on this.