NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Solved
Root full due to old volume
So, I need to get better with bash... But in the meantime, I've got a problem. My root is full.
I think I've narrowed down how it happened. I had an eda500 with a volume, eda2 on it. A few of the drives failed, and since I only had raid6, I lost the volume. No big deal, it's all backed up. I put in new drives, but it wouldn't let me use the eda2 name. Strange, but I just named it eda this time, instead.
Well, some of my apps didn't get the memo, and were trying to move files to this volume. I've fixed that part of the problem. Assuming that's actually the problem... Now I need to clean up the mess, but don't know the right sequence or commands.. I'm a windows guy. So here's where I'm at:
root@Hydra:~# df -h Filesystem Size Used Avail Use% Mounted on udev 10M 4.0K 10M 1% /dev /dev/md0 4.0G 3.6G 30M 100% / tmpfs 7.8G 8.0K 7.8G 1% /dev/shm tmpfs 7.8G 3.6M 7.8G 1% /run tmpfs 3.9G 3.5M 3.9G 1% /run/lock tmpfs 7.8G 0 7.8G 0% /sys/fs/cgroup /dev/md127 13T 4.9T 7.9T 39% /data /dev/md127 13T 4.9T 7.9T 39% /apps /dev/md127 13T 4.9T 7.9T 39% /home /dev/md126 11T 4.2T 6.8T 39% /eda shm 64M 0 64M 0% /apps/docker-cli-rnapp/containers/2281928361f944e1ba885f4b49fceb4061e08b46b112e0923327 4193f0a78856/mounts/shm shm 64M 0 64M 0% /apps/docker-cli-rnapp/containers/1e9868e9103924f992bb0fb3d915b2b114be0d025fa7cd9f2b6a e33d7d5e70c0/mounts/shm
root@Hydra:/# du -hsx * | sort -rh | head -10 du: cannot access 'proc/24047/task/24047/fd/3': No such file or directory du: cannot access 'proc/24047/task/24047/fdinfo/3': No such file or directory du: cannot access 'proc/24047/fd/3': No such file or directory du: cannot access 'proc/24047/fdinfo/3': No such file or directory 69G apps 2.5G eda2 539M var 476M usr 68M lib 51M data 30M frontview 11M sbin 11M etc 6.4M bin
root@Hydra:/# rm -rf eda2 rm: cannot remove 'eda2/._share': Operation not permitted
There are a lot of mount points in the root partition, which can conceal the actual space usage in the partition.
So start by entering
mount --bind / /mnt
That mounts sysroot as /mnt.
Then look at the space usage in /mnt. For instance,
cd /mnt
du -hsx * | sort -rh | head -10You can of course then descend into folders and repeat the du command (or use other commands to see what's taking the space).
BTW, anything that is under a mount point can be deleted.
x219c wrote:
root@Hydra:/# rm -rf eda2 rm: cannot remove 'eda2/._share': Operation not permitted
It's hard to be certain from what you posted, but I think it's likely that eda2 is actually the mount point for your new eda volume. So check if the folders/files in eda2 match whats in eda.
If I'm correct, eda2 is a btrfs volume - not the same as an ordinary folder. That would explain why the rm command failed. If it had worked you'd have deleted all the files on eda.
When you're done with /mnt, you unmount it with
cd / umount /mnt
The cd matters, the umount will fail if you are in a folder of /mnt.
3 Replies
Replies have been turned off for this discussion
I could be barking up the wrong tree entirely, of course. This was my only theory, but the 69G next to apps... Well, IDK.
There are a lot of mount points in the root partition, which can conceal the actual space usage in the partition.
So start by entering
mount --bind / /mnt
That mounts sysroot as /mnt.
Then look at the space usage in /mnt. For instance,
cd /mnt
du -hsx * | sort -rh | head -10You can of course then descend into folders and repeat the du command (or use other commands to see what's taking the space).
BTW, anything that is under a mount point can be deleted.
x219c wrote:
root@Hydra:/# rm -rf eda2 rm: cannot remove 'eda2/._share': Operation not permitted
It's hard to be certain from what you posted, but I think it's likely that eda2 is actually the mount point for your new eda volume. So check if the folders/files in eda2 match whats in eda.
If I'm correct, eda2 is a btrfs volume - not the same as an ordinary folder. That would explain why the rm command failed. If it had worked you'd have deleted all the files on eda.
When you're done with /mnt, you unmount it with
cd / umount /mnt
The cd matters, the umount will fail if you are in a folder of /mnt.
Thanks Stephen! Looks like I must have had a docker image in there. All good now!
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
