[Unraid OS version 6.8.0-rc5 available]

Thanks for your help. I'm on rc1 now. Streaming from another drive and running mover works.

Before I streamed from one drive and also did write on that same drive with the mover because of my data layout.

I have to test the same with sittingmongoose's settings on rc5 and I will report back.

 

 

@Squid I'm running it at Normal at the moment on rc1.

[Unraid OS version 6.8.0-rc5 available]

7 minutes ago, sittingmongoose said:

How are you testing array performance?

 

Did you ever try RC1?

No, I didn't test it. What's the best way to do that? Maybe with reinstalling the last official version and move on from that?

 

I start a videostream and open up my shares. After that I start mover.

About 10-15 seconds after starting the mover my stream stops completely. Shares get unresponsive immediately.

That's my test for the array. 

[Unraid OS version 6.8.0-rc5 available]

I did try your suggestion: disk scheduler set to none and disk IO priority set to low on 6.8 rc5.

Array performance is not better with these settings.

[Unraid OS version 6.8.0-rc5 available]

6 hours ago, sittingmongoose said:

For anyone interested about the array performance issues compared to RC1.  Setting the disk tuning scheduler from "mq-deadline" to "none" along with using the mover tuning plugin to set disk IO priority to "low" fixed my problems.  Now it performs like RC1 did. 

Tested it, but it didn't work for me. I guess I'm going to disable the cache for now.

[Unraid OS version 6.8.0-rc5 available]

I have the same problem with array performance on rc5 and the last versions. I didn't test rc1 because I started with rc3.

When the mover runs shares become completely inaccessible and streams completely stop. 

I have two Samsung 860 EVO 1TB in RAID 1 (btrfs) assigned to cache.

Changing the priority in CA Mover Tuning to Low and even very low didn't change a thing.

Iowait is really high. Up to 84 sometimes. Usually about 50-60 with really high cpu usage.

Turbo write is disabled, but didn't make a difference when enabled.

[[6.7.x] Very slow array concurrent performance]

I don't think that issue is fixed with v6.8. Atleast not for me and some other users.

I still have high iowait and a completely unresponsive array with rc5 when the mover is running.

Streams / shares are not accessible or stop completely. Im running two SSDs in raid 1 with btrfs.

Somebody said that it could be a btrfs issue but I don't know.