Every check has lots of errors

[AlbertoGa]

Hi,

 

Every check I do has a lot of errors and I don't know what to do about this. I attach a screenshot of the check history.

 

Any help is appreciated.

 

Thanks

[itimpi]

You definitely want to get to the bottom of this!  It might be indicative of an underlying issue.  Are these checks correcting or non-correcting?  Have you had any unclean shutdowns?

 

You should post your system's diagnostics zip file in your next post in this thread to get more informed feedback.  It is always a good idea to post this if your question might involve us seeing how you have things set up or to look at recent logs.

 

BTW:  You might want to consider installing the Parity Check Tuning plugin.  Even if you do not make use of its other features the Parity History entries will start being enhanced to give more information about the check such as whether it was correcting or not.

[AlbertoGa]

5 minutes ago, itimpi said:

You definitely want to get to the bottom of this!  It might be indicative of an underlying issue.  Are these checks correcting or non-correcting?  Have you had any unclean shutdowns?

 

You should post your system's diagnostics zip file in your next post in this thread to get more informed feedback.  It is always a good idea to post this if your question might involve us seeing how you have things set up or to look at recent logs.

 

BTW:  You might want to consider installing the Parity Check Tuning plugin.  Even if you do not make use of its other features the Parity History entries will start being enhanced to give more information about the check such as whether it was correcting or not.

 

 

Thanks for the reply, here's the ZIP.

 

I've had a disk show up as unmountable a couple times but fixed it with this:

 

I've installed the plugin but I don't know how to distinguish between correcting and non-correcting errors.

nas-diagnostics-20240225-1822.zip

[itimpi]

4 minutes ago, AlbertoGa said:

but I don't know how to distinguish between correcting and non-correcting errors

Wait until you run another check - and then look at the Parity History - it will be obvious in the new entry.

[AlbertoGa]

That will take a couple of days, I'll do it ASAP. Is there anything else I should do beforehand/mean time?

[itimpi]

The diagnostics show you are getting continual resets on whatever device is ata3.   This will be badly slowing down performance.  The diagnostics do not go back far enough for me to see which device that is, but you should be able to easily find out by clicking on the icon at the beginning of the Identification column on each drive.   

 

You should carefully check the cabling (both power and SAT) to the drive as it looks like the sort of error we see if a SATA cable is not properly seated or there are issues getting sufficient power to the drive.

[AlbertoGa]

Could overheating cause this?

[itimpi]

12 minutes ago, AlbertoGa said:

I've had a disk show up as unmountable a couple times but fixed it with this

Were you doing this via the GUI or the command line?  If the command line what device name were you using?

[itimpi]

4 minutes ago, AlbertoGa said:

Could overheating cause this?

Not sure if it can do it directly unless the server was shutting itself off unexpectedly due to the CPU overheating.    However it could definitely cause cables to start working themselves loose.

[AlbertoGa]

1 minute ago, itimpi said:

Were you doing this via the GUI or the command line?  If the command line what device name were you using?

I used the GUI, and followed the instructions in that thread.

 

1 minute ago, itimpi said:

Not sure if it can do it directly.    However it could definitely cause cables to start working themselves loose.

I'll have to check the cables.

 

5 minutes ago, itimpi said:

The diagnostics show you are getting continual resets on whatever device is ata3.   This will be badly slowing down performance.  The diagnostics do not go back far enough for me to see which device that is, but you should be able to easily find out by clicking on the icon at the beginning of the Identification column on each drive.   

 

You should carefully check the cabling (both power and SAT) to the drive as it looks like the sort of error we see if a SATA cable is not properly seated or there are issues getting sufficient power to the drive.

The disk icon opens this log:

[itimpi]

1 minute ago, AlbertoGa said:

The disk icon opens this log:

That tells you which drive is generating those errors.

[itimpi]

2 minutes ago, AlbertoGa said:

I used the GUI, and followed the instructions in that thread.

OK.   I was just checking as if you do it from the command line depending on the device name used it is possible to start invalidating parity.  From the GUI this will not happen.

[AlbertoGa]

Thank you for the help and fast replies, I'll check the disk and report back.

[AlbertoGa]

 

39 minutes ago, itimpi said:

That tells you which drive is generating those errors.

I shut down the server, checked the wiring and booted again. I've started a new check but the disk still looks more or less the same. The weird thing is that the disk status (disk 2) shows as healthy:

[AlbertoGa]

Just now, AlbertoGa said:

 

I shut down the server, checked the wiring and booted again. I've started a new check but the disk still looks more or less the same. The weird thing is that the disk status (disk 2) shows as healthy:

Also, the check is VERY slow. It mayu take several days according to the estimate.

[itimpi]

Those errors cause retries, and you will only get an error reported if all retries fail.   

 

Just now, AlbertoGa said:

Also, the check is VERY slow. It mayu take several days according to the estimate.

 

That is expected as the continual retries are slowing everything down.   Personally I would not bother checking until you can get to the bottom of why you keep getting these retries.   

 

You are not by any chance using power splitters on the cabling to the drive?

 

[AlbertoGa]

Just now, itimpi said:

Those errors cause retries, and you will only get an error reported if all retries fail.   

 

 

That is expected as the continual retries are slowing everything down.   Personally I would not bother checking until you can get to the bottom of why you keep getting these retries.   

 

You are not by any chance using power splitters on the cabling to the drive?

 

Yes I am using splitters. Is this bad? Shouldn't it have been a problem from the start?

[AlbertoGa]

1 minute ago, itimpi said:

Those errors cause retries, and you will only get an error reported if all retries fail.   

 

 

That is expected as the continual retries are slowing everything down.   Personally I would not bother checking until you can get to the bottom of why you keep getting these retries.   

 

You are not by any chance using power splitters on the cabling to the drive?

 

I'm, quite honestly, a novice. How do I get to the bottom of this?

[itimpi]

2 minutes ago, AlbertoGa said:

Yes I am using splitters. Is this bad? Shouldn't it have been a problem from the start?

What type of splitters and how many way?

 

Voltage sag to a  drive can cause these symptoms.

[AlbertoGa]

1 minute ago, itimpi said:

What type of splitters and how many way?

 

Voltage sag to a  drive can cause these symptoms.

It's these: https://www.amazon.es/gp/product/B08HMCBHG5/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1

Molex to 2x SATA power.

[itimpi]

3 minutes ago, AlbertoGa said:

It's these: https://www.amazon.es/gp/product/B08HMCBHG5/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1

Molex to 2x SATA power.

In principle that should be fine - with Molex you can normally go to 4xSATA.

 

However beware of those splitters where the cable goes vertically into a moulded connector at the SATA end - I have seen them being reported as a potential fire risk as the connectors can accidentally touch if badly manufactured.

[AlbertoGa]

57 minutes ago, itimpi said:

In principle that should be fine - with Molex you can normally go to 4xSATA.

 

However beware of those splitters where the cable goes vertically into a moulded connector at the SATA end - I have seen them being reported as a potential fire risk as the connectors can accidentally touch if badly manufactured.

I've checked the smart attributes. Also I've done both a SMART short self-test and SMART extended self-test. First one was ok, second one doesn't complete. It says: Interrupted (host reset)

1 hour ago, AlbertoGa said:

I'm, quite honestly, a novice. How do I get to the bottom of this?

 

[itimpi]

You do need to see if the drive in question can complete an extended SMART test without error.   If necessary disable spindown on the drive before starting the test and make sure you do not try to access it while the test is running.   If it cannot complete the Extended SMART test without error then the drive needs replacing (the test is entirely internal to the drive and not affected by cabling issues).

[AlbertoGa]

I've done several attempts and it doesn't complete it. How come it doesn't show the disk as unhealthy or damaged? Can I claim warranty on this?

[itimpi]

1 minute ago, AlbertoGa said:

I've done several attempts and it doesn't complete it. How come it doesn't show the disk as unhealthy or damaged? Can I claim warranty on this?

Has it failed with an indication there was a read failure?   If so you can probably make a warranty claim.

 

note that the SATA overall health assessment takes no account of the effect of running tests.   It is based purely on whether any of the attributes have a “failing now” status so is frequently not a useful indication of failure.