Creating a ZFS Snapshot Archive for Recovery

Before you create the ZFS root pool snapshot, consider saving the following information:

Capture the root pool properties.
```
sysA# zpool get all rpool
```

Identify the size and current capacity of the root pool disk.

sysA# zpool list
NAME   SIZE  ALLOC   FREE  CAP  DEDUP  HEALTH  ALTROOT
rpool   74G  5.42G  68.6G   7%  1.00x  ONLINE  -

Identify the root pool components.

sysA# zfs list -r rpool
NAME                      USED  AVAIL  REFER  MOUNTPOINT
rpool                    5.48G  67.4G  75.5K  /rpool
rpool/ROOT               3.44G  67.4G    31K  legacy
rpool/ROOT/solaris       3.44G  67.4G  3.14G  /
rpool/ROOT/solaris/var    303M  67.4G   214M  /var
rpool/dump               1.01G  67.4G  1000M  -
rpool/export             97.5K  67.4G    32K  /rpool/export
rpool/export/home        65.5K  67.4G    32K  /rpool/export/home
rpool/export/home/admin  33.5K  67.4G  33.5K  /rpool/export/home/admin
rpool/swap               1.03G  67.4G  1.00G  -

How to Create a ZFS Snapshot Archive

The following steps describe how to create a recursive snapshot of the root pool that will include all file systems in the root pool. Other non-root pools can be archived in this same way.

Consider the following points:

For complete system recovery, send the snapshots to a pool on a remote system.
Create an NFS share from the remote system and also configure ssh to allow privileged access, if necessary.
The recursive root pool snapshot is sent as one large snapshot file to a remote system, but you could send the recursive snapshots to be stored as individual snapshots on a remote system.

In the steps that follow, the recursive snapshot is named rpool@snap1. The local system to be recovered is sysA and the remote system is sysB. Note that rpool is the default root pool name and might be different on your system.

Become an administrator.
Create a recursive snapshot of the root pool.
```
sysA# zfs snapshot -r rpool@rpool.snap1
```
Reduce the snapshot archive by removing the swap and dump snapshots, if desired.
```
sysA# zfs destroy rpool/dump@rpool.snap1
sysA# zfs destroy rpool/swap@rpool.snap1
```
The swap volume does not contain data that is relevant to a system migration or recovery. Do not remove the dump volume snapshot if you wish to preserve any crash dumps.

Send the recursive root pool snapshot to another pool on another system.

Share a file system on a remote system for receiving the snapshot or snapshots:
In the following steps, the /tank/snaps file system is shared for storing the recursive root snapshot.
```
sysB# zfs set share=name=snapf,path=/tank/snaps,prot=nfs,root=sysA tank/snaps
sysB# zfs set sharenfs=on tank/snaps
```

Send the recursive root pool snapshot to a remote system.

Send the recursive snapshot to the remote file system that was shared in the previous step.

sysA# zfs send -Rv rpool@rpool.snap1 | gzip > /net/sysB/tank/snaps/
rpool.snap1.gz
sending from @ to rpool@rpool.snap1
sending from @ to rpool/export@rpool.snap1
sending from @ to rpool/export/home@rpool.snap1
sending from @ to rpool/export/home/admin@rpool.snap1
sending from @ to rpool/ROOT@rpool.snap1
sending from @ to rpool/ROOT/solaris@install
sending from @ to rpool/ROOT/solaris@install
sending from @install to rpool/ROOT/solaris@rpool.snap1
sending from @ to rpool/ROOT/solaris/var@install
sending from @install to rpool/ROOT/solaris/var@rpool.snap1

Skip Navigation Links
Exit Print View
	Oracle Solaris Administration: ZFS File Systems Oracle Solaris 11 Information Library