Volume expansion question

[scottmaverick]

My original volume consisted of 12 3tb drives in raidz3. I had read that performance is degraded if the capacity is over 80%. So I bought another 12 3tb drives and used them to extend the volume. I did this using the GUI and extended the existing volume by adding the new 12 drives in a raidz3 configuration. I now have a volume that consists of 2 vdevs each of which is raidz3.

I started copying more files to the volume and noticed that it was still only writing to my first vdev which was 80% full. The idea was to keep the vdev under 80% to avoid the performance degradation. Is there a way to have new files be only written to the new vdev? Or is my understanding of how everything works completely wrong?

 

[Arwen]

Please show us the following output in CODE tags;

Code:

zpool status
zpool list -v

 

[scottmaverick]

Code:

 zpool status
  pool: Media
 state: ONLINE
  scan: scrub repaired 0 in 17h9m with 0 errors on Thu Dec 15 21:09:49 2016
config:

		NAME											STATE	 READ WRITE CKSUM
		Media										   ONLINE	   0	 0	 0
		  raidz3-0									  ONLINE	   0	 0	 0
			gptid/04c115ba-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/05c1e9ee-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/06c70250-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/07c5a98d-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/08c1d3f1-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/09bf94c1-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/0abdd3ba-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/0bbe0b65-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/0cbd894c-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/0dbaaae6-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/0eb7691c-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/0fbbce51-6a25-11e6-a63b-e0cb4e0ffe38  ONLINE	   0	 0	 0
		  raidz3-1									  ONLINE	   0	 0	 0
			gptid/72193012-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/733d960f-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/7467fddf-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/75845a00-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/76a5dfeb-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/77c36a78-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/78e5066a-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/7a0ffc76-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/7b3745ee-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/7c5d623b-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/7d7dcd79-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0
			gptid/7e9c399a-c792-11e6-a5e5-e0cb4e0ffe38  ONLINE	   0	 0	 0

errors: No known data errors

  pool: Plex
 state: ONLINE
  scan: scrub repaired 0 in 0h7m with 0 errors on Thu Dec 15 04:07:43 2016
config:

		NAME										  STATE	 READ WRITE CKSUM
		Plex										  ONLINE	   0	 0	 0
		  gptid/2d25dedb-696f-11e6-8378-e0cb4e0ffe38  ONLINE	   0	 0	 0

errors: No known data errors

  pool: freenas-boot
 state: ONLINE
  scan: scrub repaired 0 in 0h0m with 0 errors on Wed Dec  7 03:45:38 2016
config:

		NAME										  STATE	 READ WRITE CKSUM
		freenas-boot								  ONLINE	   0	 0	 0
		  gptid/92ff4e72-68c0-11e6-bc4a-e0cb4e0ffe38  ONLINE	   0	 0	 0

errors: No known data errors

Code:

zpool list -v
NAME									 SIZE  ALLOC   FREE  EXPANDSZ   FRAG	CAP  DEDUP  HEALTH  ALTROOT
Media									 65T  27.7T  37.3T		 -	18%	42%  1.00x  ONLINE  /mnt
  raidz3								32.5T  27.4T  5.10T		 -	36%	84%
	gptid/04c115ba-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/05c1e9ee-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/06c70250-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/07c5a98d-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/08c1d3f1-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/09bf94c1-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/0abdd3ba-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/0bbe0b65-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/0cbd894c-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/0dbaaae6-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/0eb7691c-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/0fbbce51-6a25-11e6-a63b-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
  raidz3								32.5T   275G  32.2T		 -	 0%	 0%
	gptid/72193012-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/733d960f-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/7467fddf-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/75845a00-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/76a5dfeb-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/77c36a78-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/78e5066a-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/7a0ffc76-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/7b3745ee-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/7c5d623b-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/7d7dcd79-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
	gptid/7e9c399a-c792-11e6-a5e5-e0cb4e0ffe38	  -	  -	  -		 -	  -	  -
Plex									81.5G  20.9G  60.6G		 -	71%	25%  1.00x  ONLINE  /mnt
  gptid/2d25dedb-696f-11e6-8378-e0cb4e0ffe38  81.5G  20.9G  60.6G		 -	71%	25%
freenas-boot							14.2G  1.28G  13.0G		 -	  -	 8%  1.00x  ONLINE  -
  gptid/92ff4e72-68c0-11e6-bc4a-e0cb4e0ffe38  14.2G  1.28G  13.0G		 -	  -	 8%

 

[Arwen]

It would appear your second vDev, "raidz3-0", is currently using 275GB.

How much new data have you written to the pool overall?

ZFS should favor the new RAID-Z3 vDev over the original one. If you
have old static data in the pool that you can copy to a new dataset, that
should make use of the new vDev.

But otherwise, you appear to have done the correct thing in adding your
new RAID-Z3 vDev. (Occasionally we see mis-matched pools...)