Ask Your Question
0

"No space left on device" error but I have at least 1.5TB available overall

asked 2016-03-02 22:03:38 -0600

koesherbacon gravatar image

Hey there, I hope some one can answer this question for me.

Whenever I reboot I receive the following error, which is multiple lines of the same or similar output:

# cat /var/log/messages | grep space
....
Mar  2 16:05:07 fedora systemd-udevd: inotify_add_watch(9, /dev/sda, 10) failed: No space left on device
Mar  2 16:05:07 fedora systemd-udevd: inotify_add_watch(9, /dev/sda, 10) failed: No space left on device
Mar  2 18:53:39 fedora systemd[1]: Failed to add watch on /run/systemd/ask-password: No space left on device
Mar  2 18:53:39 fedora systemd[1]: Failed to add watch on /run/systemd/ask-password: No space left on device
Mar  2 18:53:39 fedora systemd[1]: Failed to add watch on /run/systemd/ask-password: No space left on device
Mar  2 18:53:39 fedora systemd[1]: Failed to add watch on /run/systemd/ask-password: No space left on device
Mar  2 18:53:39 fedora systemd: Failed to add watch on /run/systemd/ask-password: No space left on device
Mar  2 18:53:39 fedora systemd: Failed to add watch on /run/systemd/ask-password: No space left on device
Mar  2 18:53:39 fedora systemd: Failed to add watch on /run/systemd/ask-password: No space left on device
Mar  2 18:53:39 fedora systemd: Failed to add watch on /run/systemd/ask-password: No space left on device

When I go to check what's going on with my limited knowledge of linux, I can see that I have more than enough space on my 2TB HDD, which is formatted with one ext2 /boot/ partition, one btrfs / partition (with /home, /usr, /var, & /tmp subvolumes) and one swap partition.

$ df -T
Filesystem     Type      1K-blocks      Used  Available Use% Mounted on
devtmpfs       devtmpfs    8168344         0    8168344   0% /dev
tmpfs          tmpfs       8181592     37984    8143608   1% /dev/shm
tmpfs          tmpfs       8181592      1860    8179732   1% /run
tmpfs          tmpfs       8181592         0    8181592   0% /sys/fs/cgroup
/dev/sda2      btrfs    1936223232 293723736 1640296008  16% /
/dev/sda2      btrfs    1936223232 293723736 1640296008  16% /usr
/dev/sda2      btrfs    1936223232 293723736 1640296008  16% /home
/dev/sda2      btrfs    1936223232 293723736 1640296008  16% /var
/dev/sda2      btrfs    1936223232 293723736 1640296008  16% /tmp
/dev/sda1      ext2         495844    130296     339948  28% /boot
tmpfs          tmpfs       1636320       604    1635716   1% /run/user/1000

Can someone explain what this error means and how to go about fixing it? It's nothing urgent since I don't think it's actually effecting anything with my computer, but I'd still like to know what's going on.

Thanks a lot!

~ Ev

edit retag flag offensive close merge delete

Comments

1

df has not learned enough about btrfs to be useful here, you should try other methods. https://btrfs.wiki.kernel.org/index.p... seems like a good resource.

randomuser gravatar imagerandomuser ( 2016-03-02 23:08:01 -0600 )edit
1

Yes, df can't show disk usage of btrfs file systems accurately, but it won't show a 100% full disk as 16% full either.

The error message is misleading, it is hardly about your HDD. I'm almost sure that 'device' here is not referring to your hard disk. Considering the manual page of inotify_add_watch, I guess the actual error message is one of these:

ENOMEM Insufficient kernel memory was available.
ENOSPC The user limit on the total number of inotify watches was reached or the kernel failed to allocate a needed resource.

Don't you see any other error messages in system logs?

hedayat gravatar imagehedayat ( 2016-03-03 03:21:03 -0600 )edit

2 Answers

Sort by ยป oldest newest most voted
0

answered 2016-03-08 21:04:52 -0600

cmurf gravatar image
btrfs fi usage /
btrfs balance start -dusage=5 /
btrfs fi usage /

Do that and then report the complete results (both usage outputs and the balance including any error). This should recover allocated chunks with data profile, that aren't actually being used to store data, back into unallocated space. If you get a failure with a value of 5, try a lower value, including 0 if necessary.

edit flag offensive delete link more
0

answered 2016-03-03 09:31:16 -0600

curt1s gravatar image

Try to check what's using up inotify watches and investigate it..Or as a temporary solution increase the max number of inotify watches limit.

It's pretty well explained @ the following link : http://unix.stackexchange.com/questio...

edit flag offensive delete link more

Question Tools

1 follower

Stats

Asked: 2016-03-02 22:03:38 -0600

Seen: 3,571 times

Last updated: Mar 03 '16