Very slow parity copy on parity swap


Recommended Posts

I am trying to use the parity swap procedure to upgrade my parity disk from a 2TB WD Green (5400RPM) to a 4TB WD Blue (WD40EZRZ, 5400RPM, CMR). Using the old 2TB drive to replace a data disk with SMART errors.

 

This is on an N54L microserver with the AMD Turion II Neo Dual Core @2.2Ghz. Unraid was just upgraded to 6.8.3 from 6.1 prior to starting the swap.

 

I started the array copy yesterday but it has only done 17% in 26 Hours.  CPU usage bar sitting around 60% with one core sitting near 90%.

 

Is this expected? I know this isn't the fastest hardware, but surely it shouldn't take a week to copy 2TB?

 

Is there any way to do this faster? If it is a CPU issue can I take the drives out and do the copy (e.g. with dd) on my desktop?

 

Thanks in advance!

Edited by jacksonliam
Link to comment
4 minutes ago, trurl said:

Nothing obvious there and nothing in syslog really after parity copy begins.

 

Can you tell us more about the hardware?

Thanks for looking!

 

Not much more to add than its:

N54L HP Microserver

AMD Turion II Neo Dual Core @2.2Ghz

8GB RAM (RAM usage around 13%).

 

Link to comment
1 hour ago, johnnie.black said:

Write cache is disabled for all disks, HP Microservers have a write cache setting in the BIOS that is disabled by default, enable it and try again.

Thanks, that appears to have done the trick! Now going at about 1% every 12 minutes (so should complete in 20 hours!)

 

Do you know if there is a risk to the whole array if I leave write cache enabled? Or would it just be a case of the files being written (e.g. currently being moved by mover) becoming corrupted in the event of power failure?

Link to comment
  • 1 month later...

Hey everyone - same problem here, very slow speeds (~700 KB/s read and write) while performing the parity swap procedure.  One of my disks died (6TB) and since the replacement drive (12TB) I bought is bigger than my current parity (10TB), I made the relevant swaps / reassignments.  Everything appears to be going smoothly, except it's been almost 24 hours and it's only array copying is only 25% complete.

 

My diagnostics zip is attached, and I'm running all of this off a ASRock B75 Pro3-M motherboard FWIW.  Six drives (including parity) are all shucked WD 8-12TB Easy Stores, with a 500GB Samsung 850 EVO SSD cache drive (also with sub-1 MB read/write speeds currently).

 

Any help much appreciated!!

 

 

unraid-diagnostics-20200927-1840.zip

Link to comment

I was having high CRC errors and the SMART disk warning on the "old" disk2 before it failed, which is why I just removed / replaced it.

 

Parity2 has been my (only) parity drive for a long time now, since the last time I upgraded my parity to a bigger size.

 

So yes, I'm currently doing a parity swap, following the procedure here to the letter, essentially a physical swap of the 6TB (disk2) that had failed with a new 12TB.  I unassigned the failed 6TB disk2, assigned the prior 10TB Parity2 to disk2, and assigned the new 12TB as Parity2.  I stopped the arrays, powered down, etc when needed per the procedure steps.  

 

I used a new SATA cable when swapping out disk2 so that shouldn't be the issue.  No errors in Main, screenshots attached.  Any thoughts?

 

EDIT:  and for the record, I only bothered with the 12TB / parity swap route instead of just replacing the failed drive with a 10TB was because Best Buy couldn't locate the three 10TB's they supposedly had in stock, and gave me a 12TB for the same price.  I didn't overcomplicate this intentionally.  

 

Speeds Main Screen Shot 2020-09-27 at 10.29.26 PM.png

Main Read:Write Screen Shot 2020-09-27 at 10.28.32 PM.png

Edited by brklynmark
Link to comment
  • 2 weeks later...
10 minutes ago, trurl said:

Connection problems on new parity

What would you recommend doing? Should i leave it for another day and see how it goes?

 

It is a brand new drive, but i haven't run a pre-clear on it. Should i cancel the parity swap and do the pre-clear then try again?

Edited by xxbigfootxx
Link to comment

Pre-clear will not help except as a confidence check that the drive is not failing.   As war raid it is most likely a connection (i.e. cabling) issue although it could also be disk controller related.

 

1 hour ago, xxbigfootxx said:

Should i cancel the parity swap and do the pre-clear then try again?

I would do this after checking that all the cabling (both power and SATA) looks OK.

Link to comment

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.