Jump to content
garycase

Seagate 8TB Shingled Drives in UnRAID

476 posts in this topic Last Reply

Recommended Posts

Initial parity build != parity check. Totally different operation, you need to do a check after the build completes if you want full confidence in the array. Many people see much higher speeds on the build than they do on the check.

 

Good point, and thanks for the correction. Once my build is complete and my data is on the array, I'll run a check to see the times.

Share this post


Link to post

These 3 drives are plugged into my AOC-SASLP-MV8 card.

 

I don’t know if you have plans to add more Seagates to the SASLP, but so you’re not surprised, expect parity check to slow down considerably with each additional disk you add past the 3 you have now.

Share this post


Link to post

Just to throw my results in.  I just did a pre-clear on a new 8T

 

~61 hours to preclear.  I'm currently running a parity sync right now.  It seems to be taking longer than my older 4T drive.  Maybe I'll do a re-sync with the 4T drive for comparison??..

 

Call it 13.5 hours to get to my old 4T mark where it would have been done.  Now it's just parity drive left as I have no other disks bigger then 4T.

So I went back and put my 4T back in as parity and rebuilt parity just so I could have a comparison.

13h  37m  to build parity. 

Since I didn't have an exact point where I crossed the 4T mark with the 8T parity build, I estimated 13h 30m..  I'll say +/- 15 minuted on that number. 

So now I can conclude that the 8T Shingled did not take noticeable time longer or shorter.  The preceived slowness WAS all in my head!

And now back to the 8T parity drive! :)

 

 

Share this post


Link to post

I don’t know if you have plans to add more Seagates to the SASLP, but so you’re not surprised, expect parity check to slow down considerably with each additional disk you add past the 3 you have now.

 

Honestly, I'm actually considering removing my SASLP cards. My UnRAID system was built back in 2011 with 2 of those cards and 15 HDD trays (I had 10 of the slots filled), and I just recently upgraded everything to the latest version of UnRAID and replaced all the drives. My mobo has 6 SATA on-board connections, and that's more than enough for my system today (I only need 4 of them with these new high capacity drives).

Share this post


Link to post

... So I went back and put my 4T back in as parity and rebuilt parity just so I could have a comparison.

13h  37m  to build parity. 

Since I didn't have an exact point where I crossed the 4T mark with the 8T parity build, I estimated 13h 30m..  I'll say +/- 15 minuted on that number. 

So now I can conclude that the 8T Shingled did not take noticeable time longer or shorter.  The preceived slowness WAS all in my head!

And now back to the 8T parity drive! :)

 

:) :)

 

As I had already told you, your limits aren't the speed of the parity drives anyway => with the 2TB and 3TB drives, and the other 4TB drives, your parity check speeds are limited by THOSE drives ... NOT by the faster drives you're using for parity [although your 4TB parity may be the same make/model as your 4TB data drives ... in which case they're all running at the same speeds between the 3TB and 4TB point (after the slower drives are out of the picture)].

 

But I suspect you feel better having physically confirmed it  :)

Share this post


Link to post

... My parity check is way faster though.

 

Of course.    You don't have the slower 2TB, 3TB, and 4TB drives in your array.    You'll always get the best possible parity check speeds if ALL of your drives are the same.

 

Share this post


Link to post

But I suspect you feel better having physically confirmed it  :)

  I do!  It was bugging me! :D

 

As I noted before, it couldn't have been different, because of the slower drives that limit the speed.  [unless you had forgotten to do the nr_requests change to fix the slowdown that v6 has in parity checks with some controllers]

 

If you really want good parity check speeds, you simply need to eliminate your "bottleneck drives".  Especially if you have any older drives with densities below 1TB/platter (I suspect some of your 2TB drives may be in that category).

 

Share this post


Link to post

If you really want good parity check speeds, you simply need to eliminate your "bottleneck drives".  Especially if you have any older drives with densities below 1TB/platter (I suspect some of your 2TB drives may be in that category).

I"m not too concerned with the parity check speeds...  I just didn't want it to be worse with the shingled drive...

But..  Can I tell from the smart data what my densities are?  or do I have to look it up online by model number?

 

Jim

Share this post


Link to post

That's the best source I've found => they don't have ALL drives listed, but they have a large number of them, and it's updated fairly regularly as they get information about additional units.

 

Did you determine what the densities are for the older drives in your system?

 

Share this post


Link to post

two of my 2T drives are 500G platters and the 3rd is 3x 667G

My 4T drives are 5x 800G platters.  I didn't bother to look up my 3T drives..

The 2T's will be next to be replaced when I fill up my array again..  Which shouldn't be for a while since I will soon add a 2nd 8T drive

to go with my 8T parity! :D

Share this post


Link to post

How is everyone's 8tb drives going? Mine are fine apart from the click sounds.

 

As far as I know I don't have the any click sounds. Its far enough away that I may just not be able to hear them.

 

As far as performance goes, I am quite satisfied with it.

Share this post


Link to post

Well....  I'm 50% with the 8T Drives... I got 2 of them and one doesn't post!  crap.  Jet won't send me a new one..  Just will refund my money! :-(

So I'll have to try and buy it again and try to work around the one per household deal! :(

 

*pout*

 

Now I have an 8T parity system with a 4T max drive! :(

 

Jim

Share this post


Link to post

I’ve just bought two of these drives that are currently on their first out of three rounds of preclearing together with a 3tb WD red in a brand new HP Gen8 Microserver (G1610t, 6gb of ram, unraid 6.1.3) using Bjp’s faster preclear: /preclear_bjp.sh -f -A -c 3 /dev/sdX.

 

The pre read time was around 20 hours which seems normal but it’s currently writing zeros on both drives at 28-29MB/s while the WD red is on its second preclear writing zeroes at closer to 150MB/s. Will the Seagate’s pick up speed or do you think something is wrong? It’s still only 5-6 hours in to the second step of the preclear so I will probably give it some more time…

 

Also; Thank you Danioj for your contributions and tests that made me confident to buy the Seagates!  :)

Share this post


Link to post

 

Also; Thank you Danioj for your contributions and tests that made me confident to buy the Seagates!  :)

 

No problemo. :)

 

As an update I've had no issues whatsoever with the drives. I have been away working most of this year (came back but then got sent away again) but am back for holidays and foreseeable future now. But the point is they have been operating 24x7 in my study humming away perfectly.

 

I am at 98% capacity on my Main Server again so will be adding more this week. Will buy without hesitation.

Share this post


Link to post

... I am at 98% capacity on my Main Server again so will be adding more this week. Will buy without hesitation.

 

Time to replace those 5 3TB Reds with 8TB SMR's  :)

 

Share this post


Link to post

Still writing zeroes at only 26.5MB/s. Will probably let it finish since its 66% but it's very frustrating that it's running so slow. My WD red 3tb finished it's 3 rounds of preclears yesterday...

 

Edit: If anyone is wondering the speed issue was solved by enabling drive cache in the HP Microserver bios!

Share this post


Link to post

One of my 8tb has smart errors on Command Timeout. What the hell does this mean?

 

It means 6.1.14 has issues with Seagate drives.  I have two Seagates throwing command timeouts, and the 5 WDs on the same controller are fine.

Share this post


Link to post

There's no issue with 6.1.4 and Seagate drives. 

 

Its just monitoring an attribute that is meaningless to anyone other than Seagate

Share this post


Link to post

There's no issue with 6.1.4 and Seagate drives. 

 

Its just monitoring an attribute that is meaningless to anyone other than Seagate

 

There is an issue.  UnRAID is suddenly reporting errors.  If the error is meaningless, then UnRAID should not be reporting it, plain and simple.

 

UnRAID is spreading FUD with this sudden change of reporting.

 

Share this post


Link to post

I have 6 x seagate 8tb in my server and only one has the command timeout issues.

 

Give it time.  Took two days for one of mine to throw the errors.  The other did it immediately.

Share this post


Link to post

You can tell unraid to ignore that SMART attribute.

 

The only issue is the default setting of monitoring that attribute.

Share this post


Link to post

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.