Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11If Linux still reports 100% usage after you delete files, stop deleting at random and reconcile the accounting first. df reports blocks allocated by the filesystem, while du totals files that are still reachable through directory entries. A process can keep a deleted file open, so its blocks remain allocated for df even though du cannot see the filename. Inodes, journal retention, reserved space, snapshots, and container layers can produce other mismatches.
Need a server to try this on? Antihost offers Linux VPS plans with NVMe storage, full root access and a 14-day free trial.
The safe objective is not simply to make one command show a lower percentage. It is to identify the exact mount, determine whether bytes or inodes are exhausted, find the owner of any hidden allocation, and verify that the corrective action changed the relevant metric.
As an Amazon Associate I earn from qualifying purchases.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →What 100% usage actually means
Linux has several kinds of capacity, and the percentage shown by a command depends on which one it measures. A filesystem can be out of data blocks, out of inodes, or reporting space that ordinary directory walks cannot account for.
df: asks the mounted filesystem how many blocks are allocated and available. Its percentage includes filesystem-level reservations and unavailable space.du: walks directory entries and adds the apparent usage of files it can reach. An unlinked file has no directory entry, so it is absent from a normalduwalk.df -i: reports inode capacity. Every file and directory consumes an inode, even when the file contains only a few bytes.
Those numbers are related but are not expected to match exactly. The useful signal is the size of the gap and which filesystem is actually full. A percentage at or above 100% can also reflect unavailable or reserved blocks; POSIX semantics allow the reported percentage to exceed 100% when available space is negative.
#1 Best Overall
- BLUE LIGHT BLOCKING GLASSES: Our blue light glasses for women boast UV400 protection, blocking harmful blue rays from computer screens. With these bluelight glasses women, you can enjoy gaming, reading, or watching TV with no worries of eye strain.
- EYE RELIEF: Our blue light glasses are engineered to reduce blurry vision or strain caused by prolonged screen time. These womens blue light glasses aid in better sleep by preventing blue light from disrupting your natural sleep cycle, helping you fall asleep faster and enjoy a more restful night's sleep, making them the perfect christmas present.
- COMFORTABLE: Crafted from durable material, our trendy blue light glasses for woman or blue light glasses men ensure long-lasting comfort. The high-transmittance lenses accessories, gamer gifts for men, help keep your eyes relaxed. Perfect stocking stuffers, gifts for gamers men, christmas gifts for men, halloween costumes.
- LIGHTWEIGHT: These clear blue light glasses feature a lightweight frame, meaning they’ll stay comfy for long-term wear. Perfect as blue blockers glasses for women or mens blue light glasses, these non-prescription fake glasses for men can be worn throughout the day. The blue light blocking glasses women feature polycarbonate HD lenses, which are anti-reflective and restore true color.
- WORRY-FREE PURCHASE: If you experience any issues, our support team is ready to assist you quickly. These are the perfect costume glasses for cosplay dressing up.
First, identify the full filesystem and save a baseline
Run the following commands before changing anything. Save their output so you can compare the filesystem after each remediation.
df -hT /path— replace/pathwith the directory where applications report a full disk, such as/,/var, or a data directory.df -i /path— checks inode usage for the same mount.sudo du -xhd1 /mountpoint— replace/mountpointwith the mount shown bydf. The-xoption prevents the walk from crossing into other filesystems.sudo lsof +L1— lists open files whose link count is below one, the usual signature of an unlinked file still held open.
df -hT gives you the filesystem type as well as human-readable capacity. Record the device, mount point, type, total, used, available, and percentage. Do not assume that a large directory under / belongs to the same filesystem: separate mounts, bind mounts, container storage, and overlays can change the accounting boundary.
Run the commands against the path that is failing, not only against /. A service writing to a dedicated /var or database volume can fail while the root mount still has free space. Conversely, a directory that looks large in a root-level walk may actually be a separate mount and irrelevant to the full filesystem.
Read the results as a decision tree
Case 1: df and du are close
The filesystem is probably genuinely full of reachable data. Drill into the largest directory reported by sudo du -xhd1, run the same command there, and continue until you find the files or directory tree responsible. Check that the files are safe to remove, archived, or moved before deleting them. If the data is required, the durable fix is to expand the correct filesystem or move data to planned storage rather than repeatedly removing unrelated files.
Keep the filesystem boundary fixed while drilling down. Repeating the command without -x can pull in another mount and make a correct diagnosis look inconsistent. Save each level’s output, including the apparent size and ownership, so a cleanup decision can be reviewed later.
Case 2: df is much higher than du
An open deleted file is the leading explanation. Other possibilities include a different mount boundary, a container overlay, a snapshot, filesystem metadata, or reserved space. Run sudo lsof +L1 before deleting more files. The alternative sudo lsof | grep deleted can help on systems where the link-count filter is unavailable, but inspect the complete path and process rather than relying on the word deleted alone.
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →A large gap is a clue, not proof. Compare the device and mount point from df -T with the path you scanned, then consider whether copy-on-write storage or a snapshot retains blocks outside the visible directory tree. The filesystem type determines which follow-up tools are safe.
Case 3: df -i is at 100%
You have inode exhaustion, not necessarily a shortage of data blocks. A tree containing millions of tiny cache, spool, temporary, or mail files can consume every inode while leaving gigabytes of block space unused. Removing a few large files will not create the inodes needed to make new files. Find the directory with the excessive file count, then remove or consolidate those small files using the owning application’s documented cleanup process.
Rank #2
- WINDEX GAMING WIPES: effortlessly clean all your gaming gear and gaming electronics in one easy step with our new Windex Gaming Wipes
- AMMONIA FREE: our ammonia-free formula is safe to use on all of your favorite gaming equipment like consoles, headsets, controllers, screens, keyboards, and more
- PRE-MOISTENED WIPES: each wipe instantly removes dust, dirt, and stubborn smudges and fingerprints leaving a streak-free shine
- ANTI-STATIC: great for cleaning and wiping down your gaming gear and electronics leaving behind no visible residue
- COMPACT RESEALABLE PACKAGE: store your pack of Windex Gaming Wipes anywhere in your home for whenever a quick and easy wipedown is needed
Inode exhaustion can produce the same “No space left on device” error as a full block pool. Treat the two alarms separately: freeing a large video may improve df -h but leave df -i at 100%, while deleting many tiny stale files may improve the inode percentage with only a small byte change.
Why deleting a large file may not free any space
On Linux and other Unix-like systems, deleting a filename unlinks that directory entry. It does not immediately erase the inode or its data blocks when a process still has the file open. The process can continue reading or writing through its file descriptor, and the filesystem must retain the blocks until the last descriptor closes.
That is why a rotated log can disappear from du while its size remains in df. Red Hat describes the behavior this way: deleting an open file unlinks it from the directory structure, but the running process can still access it and it continues to occupy disk space. Alibaba Cloud documentation likewise notes that deleted files still holding space are not visible through ordinary df or du inspection.
Log rotation can therefore appear successful while the process continues writing to the old inode. Renaming a file, unlinking it, or creating a new empty file with the same name does not move the old blocks to the new file. Only closing every descriptor to the unlinked inode releases the allocation.
Find the process that owns the space
- Run
sudo lsof +L1and look for entries with a large size, a path marked as deleted, and a process ID (PID). - Identify the service or application name in the command column. A web server, database, logging daemon, shell, or container runtime may be the owner.
- Confirm the PID and path with the service manager or the application’s own status command. Do not kill a process merely because its entry is large.
- Choose a maintenance window and close the file cleanly, reload the service if it supports reopening logs, or restart only the service that owns the descriptor.
- Run
df -hTagain. The blocks should be released after every descriptor to that inode is closed.
Rebooting closes all descriptors and will usually release the space, but it is a broad recovery action that interrupts unrelated services. A targeted restart is safer and provides a clearer audit trail. Never truncate or remove files under /proc, /sys, database directories, or container layers simply because an open-file listing looks large; first establish what the process is doing and whether the data is recoverable.
If more than one process has the same deleted inode open, restarting only one may not change df. Recheck the listing and continue until the final descriptor closes. If the process is a database, queue worker, or other stateful application, follow its shutdown or reload procedure rather than sending an arbitrary signal.
Check journald and log retention
Systemd’s journal is a common source of growth, especially when a service has been logging repeatedly. Check its current footprint with:
journalctl --disk-usage
Review the retention controls in journald.conf, including SystemMaxUse=, RuntimeMaxUse=, SystemKeepFree=, and RuntimeKeepFree=. The systemd manual documents default keep-free values of 10% for persistent system storage and 15% for runtime storage; each calculated default is capped at 4 GB. These controls limit future journal growth and preserve free space, but they do not guarantee that all existing archives vanish as soon as a filesystem fills.
Rank #3
- Ergonomic & Adjustable Design: Fight "text neck" and improve your posture. Designed for comfort, it offers an adjustable height (7.1in-8.5in) (1.4in adjustable range) and angle (5°-85°) Please note: cell phone stand is only 9 inches tall—the length of a standard pencil
- Without blocking subtitles Perfect for office desk Tiktok Face Time Zoom video—this is the ultimate kitchen desk essential for phones! Perfect viewing experience while reducing neck back strain, A must-have for gamers streamers desk, and anyone who values both style and support!
- Sleek & Stable Design fits Otterbox Case: Say goodbye to bent cables, tangles, and tip-overs! LISEN’s stylish adjustable desk phone holder features a built-in charging hole and smart cable management, keeping your phone secure and your workspace clutter-free cool gadgets for men
- ANTI-SLIP DESIGN free hands: The under 10.00 items easy gadgets back and the bottom of this kitchen essentials for women are fully covered by anti-skid silicone, which LISEN stand up desk decor can provide maximum protection device from any scratches and slides
- Your Perfect Viewing Companion:If your product arrives damaged or breaks within 365 days, reach out to LISEN for a quick and hassle-free solution. Struggling with assembly, like tightening the screw? We’ve got your back! Your experience matters to us. (Product including: phone Holder stand*1, screw*1, hexagon screwdriver*1, specification*1)
When the filesystem later becomes full, journald can stop adding data without immediately deleting every existing archive. Use the retention policy appropriate for your distribution and run its documented journal vacuum command rather than deleting active journal files by hand. Afterward, check both journalctl --disk-usage and df -hT. A policy change prevents recurrence; it does not by itself release an unrelated open deleted file.
Keep persistent and runtime journals in mind: they can live on different storage locations and have separate controls. A successful journal vacuum should be visible in the journal’s reported usage and in the affected mount’s available blocks. If only the policy changed, do not expect an immediate drop until old archives are vacuumed or naturally expire.
Understand inode exhaustion separately from byte exhaustion
df -i exposes the file-slot limit that df -h does not. POSIX defines filesystem reporting in terms that include free file slots as well as free blocks, so an application can fail to create a new file even when the byte-oriented available column still looks healthy.
Typical symptoms include errors such as “No space left on device” while df -h shows available gigabytes, and a 100% inode percentage in df -i. Use a depth-limited du walk to find the broad area, then count and inspect files within that area with tools suited to your distribution. Remove stale per-file caches, temporary files, or queue entries only after confirming the owning application’s retention rules. Consolidating many small records into fewer files can solve an inode shortage without materially changing byte usage.
Do not confuse inode exhaustion with permissions or quotas. First verify the mount’s inode report and identify the directory responsible. If a queue or cache is active, coordinate cleanup so that deleting files does not discard work that another process expects to consume.
Mount points, containers, snapshots, and reserved space
Mount boundaries
A plain du / can cross into mounted filesystems and make the result look larger or smaller than the filesystem reported by df. Use sudo du -xhd1 /mountpoint and repeat the command inside the largest child directory. Compare the mount point and filesystem type shown by df -T before acting on a path.
Recommended Free Tools
Container and overlay storage
Container layers often use copy-on-write and overlay filesystems. A deleted file inside a container, a writable layer, or a host-side log driver may not appear where you expect in the host’s directory tree. Inspect the filesystem and mount layout first, then use the container runtime’s supported prune, log, and storage commands. Do not remove opaque overlay directories or layer files manually; doing so can corrupt images or running containers.
Container cleanup can also affect shared images and other workloads. Confirm which runtime, storage root, and mount are implicated before pruning anything, and verify both host-level df and runtime-reported usage afterward.
Rank #4
- Easy Installation: Our gaming lights have been designed to enhance your entertainment experience. Simply position the led light bar next to your TV or game console to create a game room environment with ease. They come with USB power and a 60 inch cord for direct mounting and mobility
- Audio Syncing & Extensive Mode: The built-in high-sensitivity music sensor allows you to sync the TV led lights to your favorite music. Lights react and move to the beat of music, film and game audio, creating an incredible immersive sound and light experience – and all without the need for an additional hub. You can choose from 64+ patterns, including strobe, illusion, jump, energy, gradient, running, chasing, fluttering mode etc. Each mode can be applied to different scenes
- 64+ Chasing IC+RGB & W Effect: The independent control design of IC+RGB and white all-in-one multifunctional rgbw led bar lights monitor backlight allows you to easily switch the led lights for desk between IC+RGB and white to enjoy a variety of lighting effects, you can choose white lighting mode with high brightness for better illumination when working, etc. Choose IC+RGB lighting mode when you need to create an entertaining atmosphere for the light show, 16 million colours of Color changing rgb light bar let you dance to the beat and create a dynamic party atmosphere
- Dynamic and Dimmable Lights: 1%-full brightness free brightness adjustment gives you the flexibility to adjust dimmable led light bar for room decorations or a variety of occasions
- USB Powered: USB light bars for room can be powered by powernank, usb charger, TV, Monitor, laptop PC mainframe, making behind TV led lighting bar easy to install in any location.TV led lights are powered by the TV usb interface / PC mainframe, can be turned on/off synchronously with TV / PC
Snapshots and copy-on-write metadata
Snapshots can retain old blocks after a working file is deleted. Copy-on-write filesystems may also consume metadata space that ordinary file totals do not show. The correct inspection and cleanup commands depend on the filesystem type from df -T. Do not delete snapshots or alter metadata settings until you know which snapshot, subvolume, or dataset owns the blocks.
Reserved or unavailable blocks
Filesystems can reserve space for the system or report blocks unavailable to ordinary users. POSIX semantics allow the reported percentage to exceed 100% when available space is negative. This is not evidence that deleting one more application file will solve the problem. Identify the filesystem first; changing reservation settings without understanding the filesystem, mount options, and recovery margin can make an already-full system less resilient.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Choose a remedy that matches the failure mode
| Remedy | Releases already allocated blocks? | Requires service restart? | Main risk | Addresses | Filesystem scope |
|---|---|---|---|---|---|
| Close or restart the owner of an open deleted file | Yes, when the last descriptor closes | Usually yes; a reload may suffice | Service interruption if done outside a maintenance window | Hidden data blocks | Any filesystem; verify the owning process |
| Apply journald retention and vacuum policy | Vacuuming can release journal blocks; policy alone prevents future growth | No, when using supported journal controls | Removing logs may reduce diagnostic history | Journal bytes, not inode exhaustion | Persistent and runtime journal locations |
| Remove or consolidate millions of small files | Yes | Only if the application must be stopped | Deleting active queues or caches can lose work | Inodes and, secondarily, bytes | Any filesystem; file-count limits vary |
| Delete or move ordinary large files | Yes, after unlinking and closing descriptors | Not normally | Permanent data loss or breaking an application | Reachable data blocks | Only the identified mount |
| Remove snapshots or clean container layers | Potentially, if they retain the blocks | Sometimes | Irreversible loss or corrupted images | Copy-on-write data and metadata | Filesystem- and runtime-specific |
| Expand the filesystem or move planned data | Creates additional capacity rather than freeing old blocks | Depends on storage layout | Capacity-change and backup risk | Bytes; does not cure inode design limits unless recreated | Must target the correct device and mount |
The comparison matters during an incident. A retention setting prevents future growth but may release no current blocks; a service restart can release hidden blocks but causes interruption; deleting small files addresses inodes, while expanding storage adds bytes but does not redesign an inode-limited filesystem. Choose only after the metric and owner are known.
A safe recovery procedure
- Capture evidence: save
df -hT,df -i,sudo du -xhd1 /mountpoint, andsudo lsof +L1output. - Confirm the boundary: match the path, device, mount point, and filesystem type. Repeat the
duwalk with-xso other mounts do not distort the total. - Classify the shortage: decide whether the immediate limit is blocks, inodes, journal retention, open deleted files, snapshots, overlays, or reserved space.
- Act on the owner: restart or reload only the service tied to an identified open descriptor. For ordinary files, verify ownership and backups before removal.
- Use supported cleanup: apply journald vacuum and retention controls through the distribution’s documented interfaces. Use container, database, and filesystem-specific cleanup tools for their own storage.
- Verify after every change: rerun
df -hT,df -i, and, when relevant,journalctl --disk-usage. Confirm that the available column or inode percentage actually changed. - Escalate capacity: if accounting reconciles and the filesystem is genuinely full, expand that filesystem or move data to planned storage. Recheck both block and inode margins afterward.
Do not combine several destructive actions before checking the result. One measured change at a time preserves evidence and shows whether the remedy released blocks, reduced inode use, or merely changed a policy for future growth.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Prevent the next full-disk incident
- Monitor both
df -hTanddf -i; byte and inode alarms catch different failures. - Alert on individual mounts instead of only the root filesystem. A full
/var, container data mount, or database volume can stop services while/still has room. - Keep log rotation and journald retention aligned with the amount of diagnostic history you actually need. A rotation rule that renames files does not reclaim blocks while a process still holds the old descriptor.
- Include open-deleted-file checks such as
lsof +L1in incident runbooks for any host that writes large logs or temporary files. - Document snapshot, container, and filesystem-specific storage commands before an emergency. Generic deletion from a mount tree is not a substitute for understanding copy-on-write layers.
- Maintain backups before deleting databases, queues, journals, snapshots, or application data. Free space is not a reason to bypass recovery procedures.
Frequently asked questions
Why does df show 100% while du is much smaller?
The usual cause is an unlinked file still open by a process. Its directory entry is gone, so du cannot see it, but its blocks remain allocated until the last file descriptor closes. Different mount boundaries, snapshots, overlays, metadata, and reserved blocks are other possibilities.
Will rebooting always fix the problem?
A reboot closes open descriptors and usually releases space held by deleted files, but it is a broad action that interrupts every service. Identify the owning PID and restart or reload that service when possible. Rebooting does not remove snapshots, cure inode exhaustion, or change a genuinely full filesystem.
Is lsof +L1 safe to run on a production server?
It is a diagnostic listing command and does not delete files or close descriptors. Review the output carefully, especially the PID, command, path, and apparent size. The potentially disruptive step is restarting or stopping the identified service, which should follow its maintenance procedure.
Why does “No space left on device” appear when gigabytes are free?
The filesystem may have exhausted inodes rather than data blocks. Check df -i. Millions of small files can consume all file slots while df -h still reports substantial byte capacity.
Best Value
- Compatible with Wide Screens - To ensure compatibility with the dual monitor mount, your each monitor must meet three conditions at the same time: First, computer screens size range: 13 to 32 inches. Second, screen weight range: 4.4 to 19.8 lbs. Third, the back of the monitor screen must have VESA mounting holes with a pitch of 75x75mm or 100x100mm.
- Regarding the compatibility with desks - Your desk must meet three conditions at the same time: First, desk material: Only wooden desks are recommended, plastic or glass desks cannot be used. Second, desk thickness range: 0.59" - 3.54". Third, the bottom of the desk should not have any cross beams or panels, as this will interfere with installation. We recommend carefully checking that your desk and monitors meets all above conditions before purchasing.
- Dual C-Clamp Hold - Worried your dual monitors might wobble or slip? Our upgraded base uses a larger platform plus a dual C-clamp structure to lock the dual monitor arm firmly to your desk. Each arm safely keeps your screens steady while you type, click and game—no shaking, no sliding, just a clean and secure setup you can trust every day. It also provides Grommet Mounting installation choice, both options ensure stable and secure fixation for your 0.59" - 3.54" desk.
- Full-Motion Adjustment For Comfortable View - Pull the screen closer when you’re deep in a spreadsheet, push it back to watch videos, or rotate to portrait for coding — moving everything smoothly with just one hand. The monitor stand offers +85°/-50° tilt, ±90° swivel and 360° rotation. Raise your monitor up to 15.75″ to support a healthy sitting posture. Whether you’re working from home, gaming through the night, or switching between video calls and documents, getting the screens to your natural line of sight helps relieve neck, shoulder and back strain so you can stay focused longer with less fatigue.
- Keep Your Desk Organized: By lifting both screens off the desktop, this dual monitor stand opens up valuable space for your keyboard, notebook, docking station or a simple, clutter-free work area. Built-in cable management guides wires along the arms, keeping cords out of sight and out of the way. Enjoy a tidy, modern workstation that looks as good as it feels to use.
Can I delete files directly from a journal directory?
Do not remove active journal files by hand. Use the documented journal vacuum and retention interfaces for your distribution, then verify usage with journalctl --disk-usage and the affected df mount. Manual deletion can interfere with logging and may not address an open deleted file elsewhere.
Should I delete container overlay files that look large?
No. Overlay layers and copy-on-write metadata have runtime-specific ownership. Identify the filesystem and container storage root, then use supported runtime cleanup commands. Manual removal can corrupt images or running containers.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsWhen should I expand storage instead of deleting data?
Expand the correct filesystem or move planned data when accounting reconciles and the required data legitimately exceeds capacity. Expansion adds byte capacity; it does not automatically solve an inode design limit. Verify both block and inode margins after the change.
Bottom line
Start with accounting, not more deletion. Identify the exact filesystem, compare df with an -x directory walk, check inodes, and look for open deleted files. Close the process that still owns an unlinked file, apply documented journal retention, and investigate mounts, overlays, snapshots, and reservations according to the filesystem type. Only after the numbers reconcile should you remove data or expand storage.
Frequently Asked Questions
Why does df show 100% while du is much smaller?
An open deleted file is the most common explanation: its name is gone from the directory tree, but the process still holds its blocks. Check sudo lsof +L1, then also verify mount boundaries, snapshots, overlays, metadata, and reserved space.
Will rebooting release the missing space?
Usually, if open deleted files are responsible, because rebooting closes all descriptors. It also interrupts unrelated services and does not solve inode exhaustion, snapshots, or genuinely insufficient capacity; a targeted service restart is preferable when safe.
Free tools Windows power users keep installed
One-click scans. No signup required.
What does df -i tell me?
It reports inode usage, or the number of file and directory slots consumed. A filesystem can have free gigabytes but reject new files when its inode percentage reaches 100%.
Can I remove files directly from journald’s storage directory?
Use journald’s documented vacuum and retention controls instead of deleting active journal files manually. Check journalctl –disk-usage and df afterward.
Is it safe to delete large container overlay files?
No. Identify the runtime and filesystem first, then use its supported storage and prune commands. Manual deletion can corrupt images or running containers.
When is expanding the filesystem the right fix?
After df, du, inode, and ownership accounting reconcile and the required data genuinely exceeds available capacity. Expand the correct device and mount, then verify both block and inode headroom.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problemsThe Bottom Line
Bottom line: Reconcile df, du, and df -i before deleting anything else. Find and close open deleted files, use supported journal and container cleanup, respect mount and snapshot boundaries, and expand storage only after confirming the filesystem is truly full.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
