ZFS is a good file system, it can manage lots of disks.
I have three 1G harddisks when I use ZFS beginning. In accordance with my experience, raid 5 is my first choice. Raid 5 is called raidz in ZFS. Suppose my pool is named nas, I created this pool by the following command:
zpool create nas raidz c9d0t0 c9d1t0 c9t2d0 |
zpool create nas raidz c9d0t0 c9d1t0 c9t2d0
After create pool, check the pool statu:
$ zpool status -v nas
pool: nas
state: ONLINE
scrub: none requested
config:
NAME STATE READ WRITE CKSUM
nas ONLINE 0 0 0
raidz1-0 ONLINE 0 0 0
c9t0d0 ONLINE 0 0 0
c9t1d0 ONLINE 0 0 0
c9t2d0 ONLINE 0 0 0
errors: No known data errors |
$ zpool status -v nas
pool: nas
state: ONLINE
scrub: none requested
config:
NAME STATE READ WRITE CKSUM
nas ONLINE 0 0 0
raidz1-0 ONLINE 0 0 0
c9t0d0 ONLINE 0 0 0
c9t1d0 ONLINE 0 0 0
c9t2d0 ONLINE 0 0 0
errors: No known data errors
It’s time to consider to expand the capacity when this pool is full.
After I get a 2G harddisk, I have two choice:
1.replace a old disk in the pool;
2.Add the new disk into the pool.
I wouldn’t waste a disk, so I choose the second.
I checked some document, raidz don’t support add a new disk to raidz. If I use the following command:
the new disk will add to the pool by raid 0. It isn’t what I want.
So, the raid 1 + 0 is recommended in the future.
Create raid1 pool first. Create another raid1 and add the new mirror to the old pool by raid0, when it need to add new disks.