First NAS. First NAS Build.

[Stux]

I hope it wont be a problem for a 12 drive pool with 8 Seagates and 4 WDs?

It’ll be fine. How are you structuring the pool?

 

[rajy3k]

It’ll be fine. How are you structuring the pool?

Earlier I was thinking about ZFS1. However after looking at the badblocks testing I am convinced about ZFS2.

Sent from my Pixel using Tapatalk

 

[Ericloewe]

ZFS1

RAIDZ1.

ZFS2

RAIDZ2.

 

[Johnnie Black]

RADIZ1.

RAIDZ1 :p

 

[Ericloewe]

RAIDZ1 :p

Fixed. Now, to intensely curse Firefox's spellchecker for making me look like a fool.

 

[Stux]

12 way RaidZ1. Definitely a bad idea.

12 way RaidZ2, a little bit too wide.

 

[rajy3k]

RAIDZ2.

My bad! I meant RAIDZ2.

12 way RaidZ2, a little bit too wide.

What do you think will be the right thing then? I am not too sure that I need more than 2 disks of redundancy, however, I am a complete novice here and suggestions are welcome.

Also, on some of the disks badblocks testing is done. There is nothing other than 0 in 3 of the critical parameterers. However I am continuously getting "Failed SMART usage Attribute: 5 Reallocated_Sector_Ct" message. on it. Also I am getting huge values under RAW Value field for the two categories, namely "Raw_Read_Error_Rate" and "Seek_Error_Rate". Did I do something wrong?

 

[Stux]

What do you think will be the right thing then? I am not too sure that I need more than 2 disks of redundancy, however, I am a complete novice here and suggestions are welcome.

I wouldn't use more than 10 disks in a RaidZ2 vdev.

So, you could go with 2x 6-way RaidZ2, but that will mean 4 disks of parity, which you might not like.

Also, on some of the disks badblocks testing is done. There is nothing other than 0 in 3 of the critical parameterers. However I am continuously getting "Failed SMART usage Attribute: 5 Reallocated_Sector_Ct" message.

Reallocated sector count. This should not happen on a new disk, thus the disk is broken/damaged and you should RMA it.

Also I am getting huge values under RAW Value field for the two categories, namely "Raw_Read_Error_Rate" and "Seek_Error_Rate". Did I do something wrong?

Normal for Seagate drives. Seagate drives encode the total number of reads and seeks into these values.

 

[rajy3k]

Two of the HDDs (Both Seagate IronWolf 4Tb drives) are yet to finish testing for bad blocks after 115 hours. And because of the error messages, I am not able to understand what is the progress level.

Is it normal to take so long for testing?

Sent from my Pixel using Tapatalk

 

[dezufnoC]

Sooooooo.... getting back to the OP! I just purchased the HPE ProLiant ML10 and am now waiting for the rest of the parts to show up including the memory and drives. I read in another post about peeps adding fans to this box. Is it really necessary? I plan to have (4) 4TB WD Red drives.

 

[fr33n0zeer]

Sooooooo.... getting back to the OP! I just purchased the HPE ProLiant ML10 and am now waiting for the rest of the parts to show up including the memory and drives. I read in another post about peeps adding fans to this box. Is it really necessary? I plan to have (4) 4TB WD Red drives.

I have read that the HPE tower has poor airflow. I have five WD Red 4TB 5400 RPM drives and an SSD for the boot drive. This is the max number of drives this case can handle. And to do this I would have needed to purchase a hard drive enablement kit which cost about $70. I decided to get a Fractal Design Define R5 Blackout Silent tower for $90 with free shipping and a new power supply. The Fractal Design case comes equipped with eight 3.5" drive bays, two SSD drive bays, and two 5 1/4" drive bays. And the airflow is very good with this case. It comes with two 140 mm fans, with the capacity to add a bunch more. So, that's my solution.