Fast Repair much slower than Slow Repair!

Currently reading
Fast Repair much slower than Slow Repair!

6
0
NAS
DS920+
Operating system
  1. Linux
Mobile operating system
  1. Android
I have a DS920+ with expansion unit. (9 drives) Raid = SHR.

Currently I am expanding from 36TB to about 47TB using a 12TB HDD. I have 25TB of data. As part of the installation I am running Fast Repair to repair the pool. It is running painfully slow, probably three times slower than Slow Repair when I upgraded from 29TB a few weeks ago. After 28 hours the process is only 44% complete! There is no indication of time scale this time, but the system suggests that it is on step one out of two.

Why is the process so slow and why is it so much slower than slow repair even though the amount of data is virtually the same? OK. I am not using any drives as a hot spare. The system is automatically using one 12TB HDD for repair purposes.
 

Attachments

  • Synology DS920+ Pool repair.png
    Synology DS920+ Pool repair.png
    189 KB · Views: 15
I get that, but why has the replacement of one 4TB HDD with one 12TB HDD made all the difference, given that the data quantity is identical to what it it was before and that Fast Repair is meant to ignore free space?
 
The larger the array, the longer it takes for repairs/expansion. Be patient.
Finally, after four days and seven hours, repair completed. There's just a day of data scrubbing left to complete. My puzzlement remains.

Why, when I expanded from 27TB to 35Tb, on standard repair, did the process take about 36 hours, but when I expanded from 35TB to just over 43TB it took so long on Fast Repair - 103hours? Did that extra 12TB HDD take me across some critical total volume line that made all that difference?

I'm probably fussing about nothing, but I do want to understand.
 
My guess is that with a smaller drive it's not going to have much SHR data written to it, but with a 12TB drive (I guess a bigger drive than the previous expansion), more space will be taken up with SHR redundancy data...? More data written = more time taken...?
 
My guess is that with a smaller drive it's not going to have much SHR data written to it, but with a 12TB drive (I guess a bigger drive than the previous expansion), more space will be taken up with SHR redundancy data...? More data written = more time taken...?
That makes sense up to a point, but this was my third 12TB HDD. (See the screenshot in my OP) Surely in SHR only the first of the larger discs gets used for redundancy data etc.
 

Create an account or login to comment

You must be a member in order to leave a comment

Create account

Create an account on our community. It's easy!

Log in

Already have an account? Log in here.

Old thread notice: There have been no replies in this thread for quite some time. The last reply was on .
The content in this thread may no longer be relevant. It might be better to open a new thread instead.

Similar threads

So you ruled out the cables by swapping them and now it’s down to the provided port. I guess you’ll need...
Replies
8
Views
1,013
There are tools in the synology app store that you can setup to backup to an AWS S3. Back it up to there...
Replies
6
Views
888
Thx @Tedster for the link from YT. A short analysis of the issue from the link: According to Intel ATX...
Replies
11
Views
8,648
My system is used primarily for media. We prefer classic stuff so I've ripped a lot of box sets, first to...
Replies
3
Views
955
  • Question
Hi, After ditching the WD Red 3TB disks due to 3 out 4 died, I decided to install new drives with clean...
Replies
0
Views
1,025

Welcome to SynoForum.com!

SynoForum.com is an unofficial Synology forum for NAS owners and enthusiasts.

Registration is free, easy and fast!

Back
Top