David Dyer-Bennet
Patron
- Joined
- Jul 13, 2013
- Messages
- 286
My original ZFS fileserver is still running (it's still my main home fileserver). It's an AMD box with an add-on controller, 8 hot-swap bays (well, 12; 4 2.5" and 8 3.5").
Will FreeNAS understand the ZFS pool on my existing disk drives?
Given how FreeNAS can boot and run from a thumb drive, it kind of looks like I can try this with little risk.
If I export the pool under Solaris, shut down the system, and boot from a FreeNAS thumb drive (assuming FreeNAS can tolerate the hardware), should I be able to import my existing pool? What are the risks of trying this?
Under what circumstances would FreeNAS upgrade an old ZFS on-disk version? Does it require an explicit command from me? (I'm sure the on-disk ZFS version I'm running is older than what FreeNAS defaults to, but it should read old versions, and so long as it doesn't upgrade without my permission, it won't do any damage.)
Is this a simple easy safe thing to try? Or is it an insane idea nearly certain to destroy my data? Or roughly where in between?
(I've got three backup sets at any given time, I would of course make sure they were all up-to-date before trying anything like this. But having to restore from backup is still somewhat of a pain.)
(The Solaris box identifies its version as:
Sun Microsystems Inc. SunOS 5.11 snv_134 February 2010
The pool is:
-bash-4.0$ zpool get all zp1
NAME PROPERTY VALUE SOURCE
zp1 size 3.63T -
zp1 capacity 50% -
zp1 altroot - default
zp1 health ONLINE -
zp1 guid 123924226373954006 default
zp1 version 22 default
zp1 bootfs - default
zp1 delegation on default
zp1 autoreplace off default
zp1 cachefile - default
zp1 failmode wait default
zp1 listsnapshots off default
zp1 autoexpand on local
zp1 dedupditto 0 default
zp1 dedupratio 1.00x -
zp1 free 1.81T -
zp1 allocated 1.83T -
And a representative dataset in the pool is:
-bash-4.0$ zfs get all zp1/ddb
NAME PROPERTY VALUE SOURCE
zp1/ddb type filesystem -
zp1/ddb creation Sat Feb 7 12:05 2009 -
zp1/ddb used 1.56T -
zp1/ddb available 1.75T -
zp1/ddb referenced 1.35T -
zp1/ddb compressratio 1.00x -
zp1/ddb mounted yes -
zp1/ddb quota none default
zp1/ddb reservation none default
zp1/ddb recordsize 128K default
zp1/ddb mountpoint /home/ddb inherited from zp1
zp1/ddb sharenfs off default
zp1/ddb checksum on default
zp1/ddb compression off default
zp1/ddb atime on default
zp1/ddb devices on default
zp1/ddb exec on local
zp1/ddb setuid off inherited from zp1
zp1/ddb readonly off default
zp1/ddb zoned off default
zp1/ddb snapdir visible inherited from zp1
zp1/ddb aclmode groupmask default
zp1/ddb aclinherit restricted default
zp1/ddb canmount on default
zp1/ddb shareiscsi off default
zp1/ddb xattr on default
zp1/ddb copies 1 default
zp1/ddb version 3 -
zp1/ddb utf8only off -
zp1/ddb normalization none -
zp1/ddb casesensitivity mixed -
zp1/ddb vscan off default
zp1/ddb nbmand on inherited from zp1
zp1/ddb sharesmb name=ddb local
zp1/ddb refquota none default
zp1/ddb refreservation none default
zp1/ddb primarycache all default
zp1/ddb secondarycache all default
zp1/ddb usedbysnapshots 209G -
zp1/ddb usedbydataset 1.35T -
zp1/ddb usedbychildren 0 -
zp1/ddb usedbyrefreservation 0 -
zp1/ddb logbias latency default
zp1/ddb dedup off default
zp1/ddb mlslabel none default
zp1/ddb net.dd-b:bup:backup yes inherited from zp1
)
Will FreeNAS understand the ZFS pool on my existing disk drives?
Given how FreeNAS can boot and run from a thumb drive, it kind of looks like I can try this with little risk.
If I export the pool under Solaris, shut down the system, and boot from a FreeNAS thumb drive (assuming FreeNAS can tolerate the hardware), should I be able to import my existing pool? What are the risks of trying this?
Under what circumstances would FreeNAS upgrade an old ZFS on-disk version? Does it require an explicit command from me? (I'm sure the on-disk ZFS version I'm running is older than what FreeNAS defaults to, but it should read old versions, and so long as it doesn't upgrade without my permission, it won't do any damage.)
Is this a simple easy safe thing to try? Or is it an insane idea nearly certain to destroy my data? Or roughly where in between?
(I've got three backup sets at any given time, I would of course make sure they were all up-to-date before trying anything like this. But having to restore from backup is still somewhat of a pain.)
(The Solaris box identifies its version as:
Sun Microsystems Inc. SunOS 5.11 snv_134 February 2010
The pool is:
-bash-4.0$ zpool get all zp1
NAME PROPERTY VALUE SOURCE
zp1 size 3.63T -
zp1 capacity 50% -
zp1 altroot - default
zp1 health ONLINE -
zp1 guid 123924226373954006 default
zp1 version 22 default
zp1 bootfs - default
zp1 delegation on default
zp1 autoreplace off default
zp1 cachefile - default
zp1 failmode wait default
zp1 listsnapshots off default
zp1 autoexpand on local
zp1 dedupditto 0 default
zp1 dedupratio 1.00x -
zp1 free 1.81T -
zp1 allocated 1.83T -
And a representative dataset in the pool is:
-bash-4.0$ zfs get all zp1/ddb
NAME PROPERTY VALUE SOURCE
zp1/ddb type filesystem -
zp1/ddb creation Sat Feb 7 12:05 2009 -
zp1/ddb used 1.56T -
zp1/ddb available 1.75T -
zp1/ddb referenced 1.35T -
zp1/ddb compressratio 1.00x -
zp1/ddb mounted yes -
zp1/ddb quota none default
zp1/ddb reservation none default
zp1/ddb recordsize 128K default
zp1/ddb mountpoint /home/ddb inherited from zp1
zp1/ddb sharenfs off default
zp1/ddb checksum on default
zp1/ddb compression off default
zp1/ddb atime on default
zp1/ddb devices on default
zp1/ddb exec on local
zp1/ddb setuid off inherited from zp1
zp1/ddb readonly off default
zp1/ddb zoned off default
zp1/ddb snapdir visible inherited from zp1
zp1/ddb aclmode groupmask default
zp1/ddb aclinherit restricted default
zp1/ddb canmount on default
zp1/ddb shareiscsi off default
zp1/ddb xattr on default
zp1/ddb copies 1 default
zp1/ddb version 3 -
zp1/ddb utf8only off -
zp1/ddb normalization none -
zp1/ddb casesensitivity mixed -
zp1/ddb vscan off default
zp1/ddb nbmand on inherited from zp1
zp1/ddb sharesmb name=ddb local
zp1/ddb refquota none default
zp1/ddb refreservation none default
zp1/ddb primarycache all default
zp1/ddb secondarycache all default
zp1/ddb usedbysnapshots 209G -
zp1/ddb usedbydataset 1.35T -
zp1/ddb usedbychildren 0 -
zp1/ddb usedbyrefreservation 0 -
zp1/ddb logbias latency default
zp1/ddb dedup off default
zp1/ddb mlslabel none default
zp1/ddb net.dd-b:bup:backup yes inherited from zp1
)