Parity Disk "disk disabled contents emulated"

[solidus_ty]

I recently upgraded my parity and one of my other array drives (2 used 16 TB EXOS drives for serverpartdeals). Sadly after this upgrade, my parity drive is throwing a "disk disabled contents emulated" but has zero smart errors. I looked through the forums and found a few posts regarding this issue checked cables, deleted the disk, and retried the sync to see if it would fix it.

Sadly the disk worked overnight fine but while transferring some data this morning the drive is back into this error state. The log shows Disk Read Errors but the unraid says the drive is healthy. Is the drive bad or is it another issue?

I have attached Diagnotiscs and the most recent SMART report. Im not sure what to look for in these documents to determine what is wrong or if it is a disk issue.

I greatly appreciate the help.

syslog.txt ST16000NM001G-2KK103_WL20DQVN-20240528-0743.txt

[JorgeB]

It's not logged as a disk problem, I would try using that disk with the onboard SATA controller, that will rule out the miniSAS cable and some possible controller compatibility issue, also replace/swap the power cable to rule that out.

[solidus_ty]

3 hours ago, JorgeB said:

It's not logged as a disk problem, I would try using that disk with the onboard SATA controller, that will rule out the miniSAS cable and some possible controller compatibility issue, also replace/swap the power cable to rule that out.

I'll do the onboard SATA connector and try that. Then try the power connector. Would there be a good way to stress test it after? Seems odd it made it through both Parity Syncs when I installed it with no errors and then randomly threw it.

[JorgeB]

I would just user the server normally, especially since before it wasn't high i/o that caused issues.

[solidus_ty]

On 5/28/2024 at 2:13 PM, JorgeB said:

I would just user the server normally, especially since before it wasn't high i/o that caused issues.

Currently running a Read-Check after swapping cables and also adding the drive to a single data power cable from the PSU. I was using the 4 slot extensions but read they could cause issues.

After running for 11 hours it's only at 7.9% and showing just over 1 million errors on the array 16tb drive, not the parity one. Is this expected from the drive being down and its fixing the errors?

[itimpi]

8 minutes ago, solidus_ty said:

After running for 11 hours it's only at 7.9% and showing just over 1 million errors on the array 16tb drive,

You should be getting 0 errors on the array drives during a parity check/sync.    My suspicion would be that the drive in question had dropped offline - the diagnostics zip file would probably allow us to confirm this.

[solidus_ty]

3 hours ago, itimpi said:

You should be getting 0 errors on the array drives during a parity check/sync.    My suspicion would be that the drive in question had dropped offline - the diagnostics zip file would probably allow us to confirm this.

Here is the whole zip file that I pulled the diagnostics from. Let me know if that helps. I just went ahead and stopped the server checked all cables and everything seems fine. I only plugged in the Sata cables directly to the motherboard to eliminate the SAS to Sata cables as a problem. I am still using the Sata power splitters but I set them up here only three drives are per cable from the PSU (which is per rail I believe).

Let me know if anything else could help narrow this down.

tpmediaserver-diagnostics-20240528-0719.zip

Edited May 30 by solidus_ty
spelling

[itimpi]

According to the syslog in the diagnostics you are getting continual read and write errors on the sdg device which is the parity drive.   I thought you said it was a different drive showing the errors?    The parity drive shows no obvious issue in the SMART report.    The errors still look like they could be cabling related (SATA or power).

[solidus_ty]

6 hours ago, itimpi said:

According to the syslog in the diagnostics you are getting continual read and write errors on the sdg device which is the parity drive.   I thought you said it was a different drive showing the errors?    The parity drive shows no obvious issue in the SMART report.    The errors still look like they could be cabling related (SATA or power).

Apologies I should have been more clear. I didn't pull a new diagnostics. That was from the initial post. so it would show the errors from the other drive.

I am currently running a new parity sync while only having all the drives connected via Sata on the MB. 

It's sitting at 9 hours of run time and zero errors so far. Once this is done ill let the server resume as normal and keep an eye on the drives. If I see not errors for 24 hours after that ill assume it would be my SAS to SATA cables.

Is it fairly common to have those cables go bad? Are they fairly fragile? 

 

[itimpi]

33 minutes ago, solidus_ty said:

Is it fairly common to have those cables go bad? Are they fairly fragile? 

It is not the cable that typically goes bad (although it can).   The problem is that the SATA connector itself is a bad design as it can easily stop making perfect contact over time, especially if the drives are introducing any vibrations into the system.

[solidus_ty]

52 minutes ago, itimpi said:

It is not the cable that typically goes bad (although it can).   The problem is that the SATA connector itself is a bad design as it can easily stop making perfect contact over time, especially if the drives are introducing any vibrations into the system.

Good to know. I have a bit of a Frankenstein build (3d printed drive housing) currently but I have been working towards fixing and setting up the system a lot better. Just hoping this issue will be solved so I can move on to other things I have been working on.

I'll keep you guys updated on the parity sync. It was looking better today but we will see when it is completed.