switched to 4.3.3 yesterday noticed the following:

 

the system did a parity check and reported 11388 errors

the drives are identified as "ata" (prefix before the drive type)

it seems as though the disks were not spun down they were all active when I checked the array in the morning after a night of inactivity.

 

Any idea why?  Should I go back to my previous 4.2.4 system ?

 

Thanks

Although there may be a problem with your server, I suspect that you may just have had bad timing.  The parity check errors may have been from adding a drive during an unRAID version that had a bug causing a failure to clear a drive.  That results in a very large number of parity errors showing up on the next check.  Thankfully that parity check fixes them.  The 'ata' prefix is normal with the newest versions.  Once this parity check has finished, stop the array and reboot the server.  It may just need that one more reboot.  After it boots, check that everything is OK, including spin down.  If good, then you may want to try one more parity check, just as a confidence builder.  I don't expect it to find any more errors.

 

If it does not cleanly boot, or you find more parity errors, then we will need to see a syslog and go from there.

Thanks Rob.

 

I just rebooted, the array came back on line OK. I have to wait an hour to see if the disks spin down automatically and then proceed according to you recommendations.

 

I'll post the results.

The disks do not spin down even after the specified time delay had elapsed.

 

How to proceed?

 

Rob wrote that the "ata" prefix is OK, is this the case even if the disks are in fact "SATA"?

I have noticed that sometimes my disks don't spin down when expected, but they do eventually spin down (if left overnight).

 

There used to be an issue that you had to do a reboot after adding a drvie to the array to get that drive to spindown properly.  I think this has been fixed.

 

Bringing up the Web GUI does not cause drives to spin up, but I'm not sure if it resets the spin down timer for drives that are not spun down yet.

 

You can always press the spin down button on the Web GUI.

switched to 4.3.3 yesterday noticed the following:

 

the system did a parity check and reported 11388 errors

the drives are identified as "ata" (prefix before the drive type)

it seems as though the disks were not spun down they were all active when I checked the array in the morning after a night of inactivity.

 

Any idea why?  Should I go back to my previous 4.2.4 system ?

 

Thanks

 

tsotz, you are a lucky dog.  If you had had a drive failure, you'd have been screwed!

 

Anyone reading this message that hasn't done so in the last month, do a parity check NOW!  There are theories as to what is causing these large number of parity errors, but we are not absolutely sure.  If you run a check and get errors, please post your results.

I have noticed that sometimes my disks don't spin down when expected, but they do eventually spin down (if left overnight).

.....

You can always press the spin down button on the Web GUI.

 

Still, this used to work like a charm on release 4.2.4 so there must have been some modification in 4.3 that screwed it up.  Surely I can spin down all the drives from the GUI, but this is a setback from the original functionality that I don't like to give up, let alone that I'm the only one complaining so it must be working correctly for the rest of the world - why not me?

 

 

 

 

A simple reboot from the web utility didn't fix the spin down problem, however after a shutdown and restart the array resumed it's normal exepcted behaviour.

 

Thank you guys for helping out.

 

 

A simple reboot from the web utility didn't fix the spin down problem, however after a shutdown and restart the array resumed it's normal exepcted behaviour.

 

Thank you guys for helping out.

 

 

 

I've read that many hard disks do not act on "sleep" commands until after a subsequent power cycle.  Not sure of why, but probably a hardware thing.

 

Joe L.