For the second time in the last 7 or so years i've done a restore from a time machine backup stored on Drobo FS to a new macbook. Here are my very foggy notes for posterity in case i need to do it again.
Both times it has been painful, i've nearly given up - but perseverance and a multitude of internet breadcrumbs kept me poking and prodding.
I actually had started to pull my important data off of backblaze (yes, this is a #humblebrag about my multiple backup redundancy prowess, not to mention all source is also in a private github!), but the annoyance of setting up apps and thigns "just so" kept me hopeful and was going to give it a go for a few more days.
How this started - water spillage when using a computer in a not so computer friendly restaurant. Shame on me. Genius folks said much of the innards are fried. A nice guy on Kijiji that does something called reflowing (??) is going to give me $225 and hopefully sell it for 6-800$ i imagine. Good for him. I've since got a new graphite macbook pro and its ready to be prolific!
Anywhew, my Drobo FS was in bad shape when i went to restore from it. It had been a bit neglegted and i couldn't find it to connect to it. Installing Drobo Dashboard on a different system showed I had a drive failure... how long had THAT been there for!? Doh.
But Drobo's are amazing right? My data should still be accessible even though it strongly warned about not being able to properly protect data...
In a nutshell, this article here gives the high level primer on how it SHOULD work to restore from a Drobo FS using migration assistant:
https://supportportal.drobo.com/retrieve/s3/knowledge/AA/AA-01431.html
A few things that are tricky about this:
- you need to know where your Drobo is by IP. A good reason to give it a static IP.
- you need to know the credentials to access it.
- you need to know the name of the share. Thats one key reason i needed to install the dashboard on another computer.
Hence, my connection string that i used:
afp://tmuser:tmuserpassword@192.167.1.XXX/tm_mbp13_feb2018/
I had all that info, but when i went through the steps i got the dreaded "some backups could not be opened" message, as well described here:
That article didn't contain the info i needed to solve the issue, but i felt i was on a decent track. Something is likely corrupt, and if could fix it, it might mount.
Anyways, given the fact one of the drives was out, i figured part of the reason for the not mounting could have been that. So a trip to the local store for a new 6TB WD Red drive, popped it in, and waited... 2+ DAYS for the yellow lights to stop blinking on my drobo, a few reboots, and finally it was back and connetable (it went totally offline 30 minutes after plopping the new drive in).
I went back and did all the steps... same thing :(.
Anyways, i now had at least rules out the drives or the drobo in and of itself being bad, and set to try to repair the sparse bundle. Pretty much this:
I hacked and slashed through many of these commands, never getting a "repaired successfully" message. However, some of the messages did seem to indicate some kind of success.
After this (which may or may not have had much an impact) i realized i was able to do the proper "hdiutil attach" for the volume from the command line - restarting migration assistant, waiting a good few minutes - the mounted TM volume showed up, and restore was possible!!!
Over wifi - about 26 hours. But its allllllll good now. Perseverance.