Hey all,

 

Had the unRAID setup going a few days. Decided to add another drive and had a problem on startup. System was hung. So I forced a reboot, and now when I try to start unmenu (or anything in /boot) I get the following:

 

root@spock:~# /boot/unmenu/uu

-bash: /boot/unmenu/uu: cannot execute binary file

 

The USB drive is brand new from Lime. <1 week old. I did a checkdisk in Windows and had no issues, but the problem persists.

 

Any ideas?

 

syslog attached.

 

Thanks,

 

-Dave

syslog.txt

Hey all,

 

Had the unRAID setup going a few days. Decided to add another drive and had a problem on startup. System was hung. So I forced a reboot, and now when I try to start unmenu (or anything in /boot) I get the following:

 

root@spock:~# /boot/unmenu/uu

-bash: /boot/unmenu/uu: cannot execute binary file

 

The USB drive is brand new from Lime. <1 week old. I did a checkdisk in Windows and had no issues, but the problem persists.

 

Any ideas?

 

syslog attached.

 

Thanks,

 

-Dave

 

First, one of your data disk file-system has some corruption.  It  was detected by the OS and to prevent you from causing more damage when you write to it, it was made read-only.

Apr  8 04:21:03 spock kernel: REISERFS error (device md13): reiserfs-2025 reiserfs_cache_bitmap_metadata: bitmap block 39059456 is corrupted: first bit must be 1

Apr  8 04:21:03 spock kernel: REISERFS (device md13): Remounting filesystem read-only

This is /dev/sdh and you need to go through the procedure described here on it http://lime-technology.com/wiki/index.php?title=Check_Disk_Filesystems

 

Second, it sounds as if you might have viewed/edited the "uu" command in the unmenu folder and it now has carriage return/linefeeds instead of just carriage returns at the ends of the lines.  (or some other unexpected characters)

 

You will probably also want to perform a full memory check, at least several passes, preferably overnight.  Some motherboards set the timing, clock speed, and voltage on the memory correctly, some do not.  You do not want to leave it to chance.  Those parameters must be set for your specific brand/model memory strips.  (poorly configured memory can lead to all sorts of strange problems)

 

Lastly, you have a lot of disks in that server, is your power supply up to the task?  Adding one additional drive might have pushed it near its limits, especially if it is not a single 12volt rail style.

 

Joe L

Joe,

 

1. Thanks. Fixed the filesystem corruption on that disk. I hadn't noticed that. Odd since that's the sort of thing I was looking for in the first place.

 

2. I had not edited uu. It (and everything in the unmenu folder) somehow got blown out when I had that bad reboot. I discovered (via cat -v) that every file in there was full of "^@".  The filesystem on the USB drive seems intact. I'm hoping it was just that folder, but I don't have the fuzziest feeling. Regardless, removing the folder and recreating/reinstalling unmenu solved my issue. Altho, I imaged the drive before I mucked with it.. I'm curious to see if anything else is trashed.

 

3. Running memtest now, just in case. 

 

4. Hmm.. I have a 750W PSU with a single 12v rail. So I hope so.  Altho, I only have 7 drives in there now.

 

So.. in the end, I seemed to have solved my apparent problem.. I'm hoping nothing else got trashed, and I'm not really sure why it happened in the first place.

 

Thanks again.. and any more feedback is totally appreciated.

 

-Dave

Interestingly enough.. I've been running memtest like crazy with no errors, however I just noticed it was reporting my my DDR3 PC3-10600 RAM as:

 

Settings: RAM: 398MHz (DDR797)

 

Which seems really odd, but I'm not familiar with memtest86. Although it's pulling the right info from SPD. If I set the BIOS from auto to either 1333 or 1066 I can't boot and I have to clear CMOS..  come to think of it, it (memtest86) is reporting my FSB as 199MHz.

 

Opinion on this?