clay_statue: Data Rebuild taking over 100 days at 600 KB/s... what's going wrong?

[clay_statue]

This is sub-optimal.  Most of the data on the recently emulated 6TB disk is just accumulated media that I don't care about but I have about 2.5TB of data (mostly movies of my dogs as puppies) that isn't critical but I'd highly prefer to keep it.

I did a quick SMART test on the drive and it seems okay.  Originally tried unbalance to remove data off the drive to another on the array that is mostly empty but the speeds on that were also very slow (2.5 MB/s).  SO I am now trying to rebuild the disk but it's even slower!

Why is my data only moving at a trickle?  I would prefer to shorten the timeline on recovering this drive from six months down to a day or two if possible.

fractaltower-diagnostics-20240110-1843.zip

Edited January 11 by clay_statue

[trurl]

Connection problems with parity.

 

Check connections, all disks, both ends, SATA and power, including splitters.

[clay_statue]

16 minutes ago, trurl said:

Connection problems with parity.

 

Check connections, all disks, both ends, SATA and power, including splitters.

May I ask which part of the diagnostics revealed this information?  I am suffering incredibly slow data rebuild speeds and I'm looking for clues as to why.

Pls help....

[trurl]

Sorry, had you confused with another user.

 

Let me take another look.

[clay_statue]

Yes, sorry somehow I am confused too.  My topic got mixed up with another user with a similar issue and somehow they're both titled the same.  My diagnostics are above but I'll put them here again.

fractaltower-diagnostics-20240110-1843.zip

[trurl]

Moving from an emulated disk to other disks in the array is not really a good idea. All disks are read to emulate the disk, and you are modifying the contents of the emulated disk, parity, and other disks in the array when your array is already compromised.

 

If you are really concerned about saving the data before attempting rebuild, better to copy (not move) the data somewhere off the array.

[trurl]

2 minutes ago, clay_statue said:

My topic got mixed up with another user with a similar issue and somehow they're both titled the same.

That was totally on me. I thought they were the same user.

 

 

[trurl]

Rebuild attempts only seem to run a short time before being cancelled, and nothing about the rebuild logged during that time. Are you cancelling it?

[trurl]

4 minutes ago, trurl said:

That was totally on me. I thought they were the same user.

You confused me by linking to that other thread.

[clay_statue]

I stopped everything dead silent and watched the disk activity and it was nil.  So I tried restarting the data rebuild and it's chugging along at 20 MB/s which will get the job done 3 days which is an acceptable timeline.

I realize I might have borked it by running the unbalance to move the data off initially and was seeking wise council from the forum of elders.  I hope it'll be fine.  Thanks for your attention towards it.

[trurl]

9 hours ago, clay_statue said:

chugging along at 20 MB/s which will get the job done 3 days which is an acceptable timeline.

Seems very slow to me. Mine are usually 100+ 

 

Are you reading/writing to the server at the same time? Do you have port multipliers?

 

Post new diagnostics if you want another look at how it's working now.