Why Deleting Large Files Doesn’t Free Disk Space on Linux and How to Fix It

Problem Description

A server in an IDC showed the '/' partition at 100% usage after an 80 GB file was removed with rm -f. The df -h output confirmed the full usage:

# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/VolGroup00-LogVol00
                       58G  7.8G   47G  100% /
tmpfs                 1.9G     0  1.9G   0% /dev/shm
/dev/vda1             190M   72M  108M  40% /boot

Root Cause Analysis

In Linux, deleting a file only removes its directory entry (unlink). If a process still has the file open, the inode and data blocks remain allocated, so the space is not reclaimed. The kernel keeps the file’s metadata until the last reference is closed.

Common Remedies

Use lsof | grep deleted to list deleted files that are still held open, then safely terminate the offending processes. This method can be risky if many critical processes are involved.

Stop or restart the application that was using the deleted file, allowing the OS to free the space automatically.

Reboot the operating system (least preferred).

For log files that are continuously written, truncate the file in place so the process can continue logging while the space is released.

Truncating a Log File In‑Place

# echo " " > /home/wangshibo.log
# cat /dev/null > /home/wangshibo.log
# > /home/wangshibo.log

Another Disk‑Space Symptom

Sometimes df -h shows ample free space, yet creating or writing a file fails with “no space left on device”. This usually stems from deleted files that still occupy space.

Step‑by‑Step Recovery for a Specific Partition (e.g., /data)

Run df -lh to identify the partition with unexpectedly high Used space.

Locate the partition (e.g., /data).

List deleted but still‑open files: lsof -n /data | grep deleted.

Kill the processes holding those files:

lsof -n /data | grep deleted | awk '{print $2}' | xargs kill -9

.

Verify no results remain with the same lsof command.

Note that after killing processes, df -h may temporarily show a spike in used space, which gradually shrinks as the kernel releases the blocks.

Reserved Space on ext Filesystems

Ext2/3/4 reserve about 5 % of the filesystem for privileged processes. On large disks this can waste space. The tune2fs command can adjust the reservation, but setting it to 0 % is discouraged for safety.

# df -T
Filesystem     Type     1K-blocks    Used Available Use% Mounted on
/dev/vda1      ext4      41151808 4962148  34076228  13% /
devtmpfs       devtmpfs   1931468       0   1931468   0% /dev
tmpfs          tmpfs      1941204       0   1941204   0% /dev/shm
tmpfs          tmpfs      1941204     652   1940552   1% /run
tmpfs          tmpfs      1941204       0   1941204   0% /sys/fs/cgroup
tmpfs          tmpfs       388244       0    388244   0% /run/user/0

# df -h
Filesystem      Size  Used Avail Use% Mounted on
/dev/vda1        40G  4.8G   33G  13% /
devtmpfs        1.9G     0  1.9G   0% /dev
tmpfs           1.9G     0  1.9G   0% /dev/shm
tmpfs           1.9G  620K  1.9G   1% /run
tmpfs           1.9G     0  1.9G   0% /sys/fs/cgroup
tmpfs           380M     0  380M   0% /run/user/0

To reduce the reserved space to 2 %:

# tune2fs -m 2 /dev/vda1
Setting reserved blocks percentage to 2% (209704 blocks)

After the change, the '/' partition shows the reclaimed space:

# df -h
Filesystem      Size  Used Avail Use% Mounted on
/dev/vda1        40G  4.8G   34G  13% /
devtmpfs        1.9G     0  1.9G   0% /dev
tmpfs           1.9G     0  1.9G   0% /dev/shm
tmpfs           1.9G  620K  1.9G   1% /run
tmpfs           1.9G     0  1.9G   0% /sys/fs/cgroup
tmpfs           380M     0  380M   0% /run/user/0

These steps help ensure that disk space is correctly reclaimed after file deletions and that reserved blocks are tuned to avoid unnecessary waste.