Any chance the Drobo 5N gets updated to Netatalk 3.1 to gain support for Spotlight?
I’ve enabled it with the procedure described here:
http://krypted.com/mac-os-x/enabling-spotlight-for-network-volumes/
But I don’t know what the consequences of this could be.
Man, don’t I wish. Sadly, past history would indicate that Drobo Inc will keep the various packages right where they are version-wise. While adding new functionality mid-stream like this is possible, there’s also a not-insignificant chance of unexpected bugs or integration problems when you do that, and they’re understandably paranoid about maintaining your data integrity.
Not to mention that the most interesting feature of netatalk 3.x won’t be possible on the 5N – the integration of Apple metadata with the Linux extended attributes. In other words, no more ._* files polluting folders. Unfortunately the 5N’s kernel doesn’t have support for extended attributes.
Mind you, it is perfectly possible to replace the embedded netatalk with a self-compiled one. I did something like that when Apple released a new version of OSX that was incompatible with the embedded netatalk. It is not a user-friendly solution, but it is something that can be done. But as of today there seems to be little reason to do it.
The page you linked doesn’t seem to indicate that netatalk 3.x is required to enable spotlight indexing. Are you sure that it is required?
Thank you both for answering.
When I’ve enabled Spotlight indexing on the Drobo, things gotten worse: it took like forever to index, and I can’t tell for sure if it was due to indexation, but it was silently disconnecting and reconnecting itself! I saw that at first by slowdown, then some transfers failed, and this logged to the console (it’s a bit long, sorry):
04/03/2014 21:12:36,000 kernel[0]: ASP_TCP Disconnect: triggering reconnect by bumping reconnTrigger from curr value 777 on so 0xffffff80560f48e8
04/03/2014 21:12:36,000 kernel[0]: AFP_VFS afpfs_DoReconnect: doing reconnect on /Volumes/Drobo5N/drb134701a00168/1/Public
04/03/2014 21:12:36,000 kernel[0]: AFP_VFS afpfs_DoReconnect: posting to KEA EINPROGRESS for /Volumes/Drobo5N/drb134701a00168/1/Public
04/03/2014 21:12:36,000 kernel[0]: AFP_VFS afpfs_DoReconnect: Max reconnect time: 30 secs, Connect timeout: 15 secs for /Volumes/Drobo5N/drb134701a00168/1/Public
04/03/2014 21:12:36,000 kernel[0]: AFP_VFS afpfs_DoReconnect: connect to the server /Volumes/Drobo5N/drb134701a00168/1/Public
04/03/2014 21:12:36,235 KernelEventAgent[76]: tid 54485244 received event(s) VQ_NOTRESP (1)
04/03/2014 21:12:36,235 KernelEventAgent[76]: tid 54485244 type 'afpfs', mounted on '/Volumes/Drobo5N/drb134701a00168/1/Odysseus', from '//GeJe@Drobo-G.local/Odysseus', not responding
04/03/2014 21:12:36,235 KernelEventAgent[76]: tid 54485244 type 'afpfs', mounted on '/Volumes/Drobo5N/drb134701a00168/1/Public', from '//GeJe@Drobo-G.local/Public', not responding
04/03/2014 21:12:36,236 KernelEventAgent[76]: tid 54485244 found 2 filesystem(s) with problem(s)
04/03/2014 21:12:36,000 kernel[0]: AFP_VFS afpfs_DoReconnect: Logging in with uam 8 /Volumes/Drobo5N/drb134701a00168/1/Public
04/03/2014 21:12:41,000 kernel[0]: AFP_VFS afpfs_DoReconnect: Restoring session /Volumes/Drobo5N/drb134701a00168/1/Public
04/03/2014 21:12:42,000 kernel[0]: AFP_VFS afpfs_DoReconnect: posting to KEA EINPROGRESS for /Volumes/Drobo5N/drb134701a00168/1/Odysseus
04/03/2014 21:12:42,241 KernelEventAgent[76]: tid 54485244 received event(s) VQ_NOTRESP (1)
04/03/2014 21:12:42,241 KernelEventAgent[76]: tid 54485244 found 2 filesystem(s) with problem(s)
04/03/2014 21:12:46,000 kernel[0]: AFP_VFS afpfs_DoReconnect: Primary Reconnect failed 22 on /Volumes/Drobo5N/drb134701a00168/1/Public
04/03/2014 21:12:46,000 kernel[0]: AFP_VFS afpfs_DoReconnect: trying Secondary Reconnect on /Volumes/Drobo5N/drb134701a00168/1/Public
04/03/2014 21:12:46,000 kernel[0]: ASP_TCP CancelOneRequest: cancelling slot 122 error 35 reqID 7325 flags 0x29 afpCmd 0x48 so 0xffffff80560f48e8
04/03/2014 21:12:46,765 KernelEventAgent[76]: tid 54485244 received event(s) VQ_NOTRESP (1)
04/03/2014 21:12:46,000 kernel[0]: AFP_VFS afpfs_DoReconnect: get the reconnect token
04/03/2014 21:12:47,000 kernel[0]: AFP_VFS afpfs_DoReconnect: already reconnected socketID 778 on /Volumes/Drobo5N/drb134701a00168/1/Odysseus
04/03/2014 21:12:47,000 kernel[0]: AFP_VFS afpfs_DoReconnect: trying secondary reconnect /Volumes/Drobo5N/drb134701a00168/1/Odysseus
04/03/2014 21:12:47,000 kernel[0]: AFP_VFS afpfs_DoReconnect: get the reconnect token
04/03/2014 21:12:47,300 KernelEventAgent[76]: tid 54485244 received event(s) VQ_NOTRESP (1)
Furthermore, the search wasn’t fast at all, like a normal search I’d get with software like EasyFind.
I’ve since disabled it, hoping that was the cause for the disconnections.
Ricardo, do you think, after reading the Netatalk page about Spotlight, even with a kernel unaware of extended attributes, there is a chance to have Spotlight working? And furthermore, won’t this fix the flooding log issues in the console I mentioned in my other post?
I don’t know. Maybe it would, maybe it won’t. In any case, it is not just a matter of having netatalk 3.1. You also need a good chunk of the Gnome backend installed. From the page you linked:
Right there you can see that netatalk doesn’t really do anything, it just delegates the indexing and searching to the Gnome Tracker component, and communicates with it using D-Bus.
In other words, even if you managed to get a recent version of netatalk on the 5N, you’d still have a long way to go. I haven’t looked at it in detail, but I’m pretty sure that compiling both D-Bus and Tracker for the 5N will be a non-trivial thing.
Thank you for your answer.
Well Spotlight isn’t the biggest issue with this product, but having a more recent version of Netatalk OR SMB wouldn’t hurt to make it so much better…