Hello All,

I have started the process of upgrading my parity drive from 18 TB to 22 TB.

I have used the following steps on swapping the drive Replacing disks in array | Unraid Docs

Which was simple 1. stop array 2. remove parity 3. start array 4. stop array 5. replace drive 6. start array 7. start parity check

So far, 5 days have passed and only 11.1% from parity completed. My average speed looks around 1-2 MB/sec

I was using my server as usual "downloading/reading/playing media" and while trying to troubleshoot the issue I read in the forum that read/write while doing the parity check will affect performance.

I have currently disabled docker and VMs and the speed is still 1-2 MB/sec.

Is there anything else I should look at?

htop

system load

disk performance

j-tower-diagnostics-20251228-1026.zip

There are reads going on from multiple disks, stop all other array activity and see if it improves.

P.S. there appears to be a container constaly restarting, see if you can find which one by looking at their uptimes.

I saw that a single container was restarting from the uptime like you mentioned. I stopped it for now.

I have a scheduled restart for a multiple of my containers on daily basis at midnight but I don't think this is affecting it.

NOTE: I have tried to stop docker engine and VM Manager to see if by stopping everything will it help or not.

Unfortunately, it did not.

This is what htop I/O tab looked like

Post a new screenshot from main showing the disk activity.

Sorry for the late reply.

These are the disks while Docker is enabled.

And these are the disks while Docker is disabled.

Not really sure what is running the disks for 255 MB Read

Do you have the Dynamix File Integrity plugin installed? I had this issue at the beginning of the month. The next question is do you have a cache drive? I noticed an issue today on mine Mover keeps running while the check is running.

I saw you are running ZFS you could try upping the memory allocated to the ZFS Cache as well

Something else i have seen to try is boot into safe mode so all plugins are removed as well.

1 minute ago, DeepDiver said:

Do you have the Dynamix File Integrity plugin installed? I had this issue at the beginning of the month. The next question is do you have a cache drive? I noticed an issue today on mine Mover keeps running while the check is running.

I saw you are running ZFS you could try upping the memory allocated to the ZFS Cache as well

Something else i have seen to try is boot into safe mode so all plugins are removed as well.

Hello,

No, I don't have Dynamix File Integrity plugin installed.

I am not configuring any cache drives. 5 Data disks and one parity.

Yes, I am configuring my drives to use ZFS.

I will try to increase the ZFS memory allocation and see if it helps or not.

If not, I will try the safe mode method to test

Had one other thought as i was rereading the forms, just to test try your old parity drive - i know its the wrong direction but you could have a bad drive out of the box.

Increasing the zfs memory allocations did not help unfortunately.

I have stopped the parity check and started the array in maintenance mode and now the parity sync speed is +- 100MB

While it is unfortunate that my services will be down for three days but at least the parity check will not take half a year.

Not really sure what was the issue here

5 hours ago, jeezany said:

Sorry for the late reply.

These are the disks while Docker is enabled.

And these are the disks while Docker is disabled.

Not really sure what is running the disks for 255 MB Read

When you took the screen shot was parity check running? the fact its faster now means its not a disk issue which is good news

Also glad to hear it is running faster in maintenance mode. I was thinking of this screen

(borrowed image from another post - thanks) - using the Unraid OS Safe Mode (no plugins, no gui)

the array should still be able to be run but no plugins will be loaded.

when you said you stopped docker did you just stop the containers or did you go in and disable docker in setting?

8 hours ago, DeepDiver said:

When you took the screen shot was parity check running? the fact its faster now means its not a disk issue which is good news

Also glad to hear it is running faster in maintenance mode. I was thinking of this screen

(borrowed image from another post - thanks) - using the Unraid OS Safe Mode (no plugins, no gui)

the array should still be able to be run but no plugins will be loaded.

when you said you stopped docker did you just stop the containers or did you go in and disable docker in setting?

All screenshots were taken while the parity check was running.

I disabled docker completely from Settings --> Docker --> Enable Docker "No"

I also disabled VMs just in case it will help or not "I am not running any VMs currently"

In that screenshot there was still something reading from disks 1 and 3, stop all other array activity for now.

Back again.

The parity has been completed successfully.

Now, I have swapped one of my data drives to increase the capacity "this was the final goal"

unfortunately, I am facing the same issue as the parity.

Speed is averaging 1.5-2 MB

There's something else reading from disk3

But I think disk 3 should be involved in the disk rebuild.

Correct me if I am wrong

All disks are involved, but they should all be showing the same read speed and be the same as the write speed.

I have placed the array in maintenance mode again and the speed went up to 100 MB +

shutting down the docker engine did not really help.

shutting down the VM manager also did not help.

maintenance mode starts the array while stopping all other services beside it if I am not mistaken.

I will try to troubleshoot more in the plugins or the shares if one of them is making this issue

11 minutes ago, jeezany said:

maintenance mode starts the array while stopping all other services beside it if I am not mistaken.

In maintenance mode, the disks are not mounted, so they cannot be accessed by anything else, including plugins or even shares, like over SMB.

So far, I have disabled NFS, FTP, SMB, all shared from global share settings.

Downgraded from 7.2.2 to 7.1.3 and still same slow speed.

Not really sure how to pinpoint the issue

And booting in safe mode?

Booting in safe mode and running in maintenance mode will return the speed to what should be 100~200 MB

What about disable all plugins?

Updates.

Looks like there was some scrub jobs in two of my drives. I stopped it for now and the speed rose to ~40MB

Just now, Veah said:

What about disable all plugins?

Is there a way to do that?

Finaly, I have an acceptable speed for the disk rebuild

So, I think my issue was that there was an ongoing scrub job for multiple of my disks at the same time that I wanted to do this disks swap activity.

not really sure what was my lesson learned from this. check everything? :')

I think this topic can be considered closed.