Hosting Operations
Check Bytes and Inodes Before Cleanup
A host is failing writes and you need to know whether byte usage or inode usage is the tighter limit before deleting anything.
Command
df -h /lab/disk-inode-cleanup && df -ih /lab/disk-inode-cleanupWhat changed
Nothing changes. The command prints filesystem byte and inode summaries.
Danger
safe
When to use it
Use at the start of disk incidents before deciding whether to hunt large files or huge counts of small files.
When not to use it
Do not treat df output as a cleanup plan; it only identifies the pressured filesystem.
Undo or recovery
No undo needed because both checks are read-only.
Expected output
Two filesystem tables: one for byte usage and one for inode usage.
demo script
Disposable terminal steps
df -h /lab/disk-inode-cleanupdf -ih /lab/disk-inode-cleanup
simulated output
What it looks like
::fixture-ready::
$ df -h /lab/disk-inode-cleanup
Filesystem Size Used Avail Use% Mounted on
/dev/vda1 25G 24G 680M 98% /lab/disk-inode-cleanup
tmpfs 982M 12M 970M 2% /run
::exit-code::0
$ df -ih /lab/disk-inode-cleanup
Filesystem Inodes IUsed IFree IUse% Mounted on
/dev/vda1 512K 499K 13K 98% /lab/disk-inode-cleanup
tmpfs 245K 14 245K 1% /run
::exit-code::0
YouTube Short
Check bytes and inodes.
Before deleting anything, check both df views. A filesystem can have free space and still run out of inodes.
LinkedIn hook
No space left can mean full bytes, full inodes, or both.
Question: During disk incidents, do you check inode pressure before cleanup?
experiments
A/B tests to run
Metric: save_rate
A: No space left has two meanings.
B: Check bytes and inodes before cleanup.