After ungrade UnRaid says HDD is unformated

[SuperAl]

Using Version 5.0 beta 14 and working PERFECTLY. Doing some upgrades and though might as well upgrade to the latest version of UnRaid while at it. BAD news. Upon bootup and looking at TOWER thru webpage it told me that 2 HDD were unformatted. I powered down and rebooted. Now it told me that 3 HDD were unformatted. PANIC is not a nice thing. I put beta 14 back on and viola the HDD are back on line and all is ok.

 

What went wrong? I have to point out that from previous versions the HDD all except 1 have MBR:unaligned.

 

Suggestions?

[tyrindor]

As long as you don't press FORMAT your data will still be there. B14 to RC5 is a simple upgrade (2 files to overwrite). I'm not sure why this would happen, and I would of grabbed a syslog before downgrading. I don't think anyone could tell you why it happened without that.

 

I assume you didn't upgrade unRAID and the hardware at the exact same time, I would of upgraded the hardware.. made sure it's working.. then upgraded.

[SuperAl]

Thats what I thought when I read the upgrade procedure. Didn't do any hardware upgrades yet just over wrote the 2 files and this is what happened. So only changes was the version.

 

With a perfectly great system, I wanted to make sure the operating system worked first. As the warning says this is BETA software.

 

And crap this forum can't even handle the log because it's too big so I had to edit it.

 

Syslog attached.

syslog-20120627-170331-1.txt

[tyrindor]

I'm not the best at logs so hopefully someone can help there. However something is strange... You did smart checks on 4 disks, and 3 of those disks are failing according to SMART (possibly the same drives that are showing up as unformatted)

 

Jun 27 17:03:07  smartctl[2692]: Device Model:    ST2000DL003

Jun 27 17:03:07  smartctl[2692]: Serial Number:    5YD35T8L

 

Jun 27 17:03:08  smartctl[2692]: Device Model:    ST32000542AS

Jun 27 17:03:08  smartctl[2692]: Serial Number:    6XW1S19J

 

Jun 27 17:03:08  smartctl[2692]: Device Model:    ST3500630AS

Jun 27 17:03:08  smartctl[2692]: Serial Number:    6QG3N3X4

 

Hopefully this is just an error, but they all have a large number of raw read errors, seek errors, and/or reallocated sectors. If you have more than 4 drives in your system, I would check those too. It would be rare for all these drives to be failing at the same time unless your PSU was faulty, or your server got a power surge. I assume you have it connected to a UPS or a quality surge bar. I hope I am wrong, and hopefully someone can chime in, but I feel the file system could be corrupted on these drives if those SMART errors are correct (which explains why they would show up as unformatted). Not sure why an older version of unRAID would not detect this though.

[RobJ]

Quick summary, your array of drives all look fine to UnRAID, mounted correctly but the system is not booting correctly, and Disk 1 needs to be replaced ASAP.

 

You always want to ignore raw read and raw seek error numbers, especially on Seagate drives, and look at the VALUE column only, which in these cases are fine, quite normal for Seagates.  However, he is very right about the second drive listed, Disk 1 with serial ending in 19J.  It is already failing SMART, with almost all of its available spare sectors used up, 3949 sectors remapped so far.  All other parameters look fine, so its not an immediate stop-the-world emergency, but you should try to replace Disk 1 as soon as possible!

 

Your syslog is rather strange, missing logging dates until the UnRAID module kicks in.  What exactly did you edit?  As we say constantly, it's best to zip the entire raw syslog and attach it, zipped but unmodified.

 

Something is wrong with the way the UnRAID part of the boot started, because it looks like something is starting the array twice, at the same time!  The first mounts are succeeding, but the second ones are failing, with errors indicating it cannot mount because already mounted!  The stats at the bottom of the syslog indicate all drives were fine, none disabled or missing or new.  I suspect that the web page showing your array is looking only at the second error response, and assuming wrongly about the drive status.

 

Because you clearly have some extra things running, I would suggest trying a clean boot with the clean 3 line go file, like the one included in each UnRAID distro.  Just rename yours to save it.  Then, if and when you are running OK, try adding some of your extras one at a time.