Sanity check before restoring from backup

[cdiddy]

So is it safe to assume that if i have replicated a pool to secondary physical volume, I can destroy my original volume, rebuild it, replicate the data back to the new primary from the secondary, and i will be back up and running? I have read through the docs, and read a few dozen posts about this topic, and this is my first time needing to restore from backup so I am quite nervous to pull the trigger. In case it matters, my .system is on a separate physical volume from either of the two i've been describing here.

 

[dlavigne]

Was your assumption correct?

 

[cdiddy]

I have not tried yet, but I will this week. I'll be sure to post here and let you know. Everything I have read seems to imply that yes, this is exactly what replication is for, and there is seemingly little info or documentation on it, because "of course it works and we shouldn't have to be more specific"... or something...

 

[adrianwi]

I've done this, and remember the same state of worry before clicking the button to destroy the pool on my primary pool! It all worked as I had expected, and even the jails 'just worked' after updating the configuration path.

 

[cdiddy]

So, after months of procrastinating I have decided to go ahead and do this... but FreeNAS wont let me destroy the pool because devices are busy. I've tried deleting jails, replication tasks, etc, to no avail. Any advice?

 

[dlavigne]

How did you try to destroy the pool?

Also, which build version (System -> Information)?

 

[wblock]

Make a full backup of everything before doing this.

Then see Detach Volume.

Don't overwrite data until the new volume is imported and verified.

 

[cdiddy]

How did you try to destroy the pool?

Also, which build version (System -> Information)?

The build is FreeNAS-11.1-U2.

below is what I THINK I learned NOT to do...

In the GUI, I tried "Detach Volume" on the storage tab, but the device was busy, so I tried destroying the datasets in that volume (per something I read somewhere in the forums from someone else who looked like they had a similar problem... I didn't bookmark it so i cant link it here). That still left me with device busy errors, and I continued searching forums... I tried to delete jails and jail storage, etc [and my config backup ran somewhere in the process and overwrote my good config... another lesson learned! (now I have the date saved in the config backup file name)] and finally ended up in CLI using zfs destroy commands, deleting one partition at a time and dd'ing zeros over the last MB of every drive........... pretty sure I did the whole detach thing wrong, but not sure how/why.

 

[cdiddy]

Make a full backup of everything before doing this.

I did have everything, jails and all, replicated to another system, so i believe that qualifies as a full backup, correct?

Then see Detach Volume

I feel like I followed those instructions (see my previous reply), but it seems like it just didnt work as described in the Docs.

Don't overwrite data until the new volume is imported and verified.

Was I supposed to Import Volume? I was under the impression that after I wiped the drives, I could simply use the drives to manually build a new zpool/volume, replicate from the remote system, jails and all, back to my main system, and all would be well? And what is meant by "verified"? Is that just browsing my data and making sure it looks like it worked?

 

[wblock]

Was I supposed to Import Volume? I was under the impression that after I wiped the drives, I could simply use the drives to manually build a new zpool/volume, replicate from the remote system, jails and all, back to my main system, and all would be well?

Either way works.

And what is meant by "verified"? Is that just browsing my data and making sure it looks like it worked?

Verifying that all the data is there and is correct. The way to do that can vary. Mostly, it's just checking to be sure that some original data was not forgotten during the copy.