100 GB limit on root partition within Intrepid

di11rod · 11-21-2009, 05:12 PM

I've got a weird (to me) problem with my Ubuntu install. I've formatted the 1 TB hard drive with three partitions--

2 gb swap
2 gb something else
900+ gb root

The problem is that when I store more than 100 GB of files, the system won't boot and it acts as if the root partition is filled. If I boot off a live CD and delete some files, then I can boot the system.

Following threads on these forums, I thought maybe I had under-sized the root partition. Nope. I used gparted while booting off a live CD, and the partition shows hundreds of GB of unused space on the same partition that has /root and everything else.

Could this be a problem with the motherboard bios not understanding file systems greater than 100 GB?

Any suggestions will be appreciated.

michaelk · 11-21-2009, 05:53 PM

There is something missing because what you posted does not match the pictures.
Are you copying the files to /mnt/sdb1 or /. The attachments only indicate two data partitions i.e swap (sdb5) and sdb1 (/mnt/sdb1). So where is / located? sda? Is the rest of the drive just unallocated space?

di11rod · 11-22-2009, 01:53 PM

Quote:

Originally Posted by michaelk

There is something missing because what you posted does not match the pictures.
Are you copying the files to /mnt/sdb1 or /. The attachments only indicate two data partitions i.e swap (sdb5) and sdb1 (/mnt/sdb1). So where is / located? sda? Is the rest of the drive just unallocated space?

Michael,

The (previously)attached images were taken while the partitions were mounted on a different computer that was booted off a live CD. I've now run gParted on the computer that's manifesting this problem. It's booted off the root partition. I interpret gparted's display to indicate that the shaded area represents used space, while the boxes indicate the partitions.

Does this make any more sense? See the new photo I shot of the gparted display while booted off the system in question.

Thanks for any advice you can offer!

michaelk · 11-22-2009, 04:06 PM

Could be that you have run out of inodes. What is the total number of files stored on this filesystem and post the output of the command:
df -i

di11rod · 11-22-2009, 09:27 PM

Michael,

wow. That might be the culprit. Will check tonight when I get home.

On a tangent, I used to do tech support for a big web publishing software company and the temp files generated by the
main product could sometimes consume all available inodes. One afternoon the admin at playboy dot com
called in with a site down situation because of inodes. He was stressed because he had to pick up his dogs
at the vet and didn't have time to flush out the millions of temp files. He gave me root credentials on
playboy dot com to fix the issue and get the site running again while he got his dogs.

Back on topic, any suggestions on what the culprit could be burning up my inodes? I have a borked
mythtv install that I can't seem to uninstall and mythbackend will thrash the drive until I kill the process after each startup. I'm going to have to end that pest once and for all.

Will post result of the df -i later tonight.

Thanks!!!

michaelk · 11-22-2009, 09:56 PM

Impossible to say since we have no idea what processes are running on your system except that something could be creating lots and lots of small files.

di11rod · 11-23-2009, 04:41 AM

Quote:

Originally Posted by michaelk

Could be that you have run out of inodes. What is the total number of files stored on this filesystem and post the output of the command:
df -i

Here's the result of the df -i command:

Quote:

df -i
Filesystem Inodes IUsed IFree IUse% Mounted on
/dev/sda1 60858368 248846 60609522 1% /
tmpfs 129189 3 129186 1% /lib/init/rw
varrun 129189 82 129107 1% /var/run
varlock 129189 3 129186 1% /var/lock
udev 129189 5074 124115 4% /dev
tmpfs 129189 2 129187 1% /dev/shm
lrm 129189 17 129172 1% /lib/modules/2.6.27-15-generic/volatile

Doesn't look like I'm running out of inodes... Any other ideas?

di11rod · 11-23-2009, 03:12 PM

Is there a log that I can check to see the exact startup error message when this boot failure occurs? I can easily copy a few more gigs to the hard drive, reboot, and check the log. But I don't know much about linux logfiles to know where startup is logged.

Appreciatively,

emotional friend

michaelk · 11-23-2009, 06:16 PM

By any chance is quota enabled?

di11rod · 11-23-2009, 07:40 PM

Quote:

Originally Posted by michaelk

By any chance is quota enabled?

Not seeing anything in /etc/fstab--

Code:

To access official Ubuntu documentation, please visit:
# /etc/fstab: static file system information.
#
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
proc            /proc           proc    defaults        0       0
# /dev/sda1
UUID=9e876903-0838-44cf-9318-ab3d9b296335 /               ext3    relatime,errors=remount-ro 0       1
# /dev/sda5
UUID=a44d828f-dcb5-4cd4-b8c1-879000a0e0c8 none            swap    sw              0       0
/dev/scd0       /media/cdrom0   udf,iso9660 user,noauto,exec,utf8 0       0

None of the quota packages seem to be installed, either. Could ubuntu have a quota setting somewhere else? Where to check?

thanks!

emotional friend

di11rod · 11-24-2009, 10:36 PM

Is there a log I can check to see startup problems?

Appreciatively,

Emotional Friend

syg00 · 11-24-2009, 10:47 PM

Have a look in /var/log/syslog* - use less or zless to look at them.
Won't show messages out of the init scripts tho'. I generally like to come up with the recovery boot option in these situations so I can see the messages fly by, and where it stops.

di11rod · 11-26-2009, 05:28 PM

Quote:

Originally Posted by syg00

Have a look in /var/log/syslog* - use less or zless to look at them.
Won't show messages out of the init scripts tho'. I generally like to come up with the recovery boot option in these situations so I can see the messages fly by, and where it stops.

I will check both resources. Thank you for this advice.

Emotional friend