Free tools Windows power users keep installed
One-click scans. No signup required.
A node can report 58% of its disk space used and still fail to create files because byte capacity and inode capacity are separate filesystem resources. In Sergey Shinder’s reported CI incident, new pods failed with “no space left on device” while the byte-use panel looked healthy; df -i showed the root filesystem’s inodes exhausted. Kubernetes documents inode-free eviction signals separately from available-byte signals, so monitoring bytes alone can miss this failure mode.
How a filesystem can have bytes left but no room for another file
Filesystems track more than the amount of data they can store. They also track inodes: records needed to represent files and directories. A filesystem can therefore have substantial byte capacity remaining while lacking the inode capacity needed to create another file. Many small files can consume inodes while using relatively few bytes.
As an Amazon Associate I earn from qualifying purchases.
In Shinder’s account, the CI nodes’ disk panel correctly showed 58% byte usage, but the root volume had reached 100% inode usage. The error reported to workloads was “no space left on device.” These are incident details from Shinder’s account; the source page’s indexed excerpt does not identify a publication year.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →What happened on the CI nodes
Shinder describes a frontend build image containing node_modules and roughly 400,000 mostly small files. The image was rebuilt several times a day, and nodes unpacked new versions when pulling them. Cleanup was triggered by byte usage at 85%, so the small files could drive inode exhaustion without reaching that byte threshold. One node reportedly retained 43 versions of the image.
#1 Best Overall
- Easily store and access 2TB to content on the go with the Seagate Portable Drive, a USB external hard drive
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop
- To get set up, connect the portable hard drive to a computer for automatic recognition no software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
As pods failed and moved to other nodes, those nodes pulled images too. By the afternoon, Shinder says 9 of 20 nodes were refusing work. These are the author’s reported counts, not general Kubernetes statistics or independently validated study results.
Why Kubernetes inode settings matter
Kubernetes’ node-pressure eviction documentation lists Linux inode-free signals separately from byte-availability signals. The relevant signals include nodefs.inodesFree and imagefs.inodesFree, alongside byte signals for those filesystems. The documented default hard threshold is 5% free inodes for both nodefs and imagefs on Linux.
Rank #2
- Easily store and access 5TB of content on the go with the Seagate portable drive, a USB external hard Drive
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop
- To get set up, connect the portable hard drive to a computer for automatic recognition software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
Those defaults apply only when no threshold parameters are changed. Kubernetes documentation warns that changing a parameter does not automatically preserve the other defaults unless default merging is enabled or the desired thresholds are explicitly configured. Shinder reports that old custom kubelet thresholds had replaced the complete default set, leaving free inodes unwatched. Operators should confirm the behavior for their deployed Kubernetes release rather than assuming a customized configuration inherits every default.
Check the filesystem layout before changing thresholds
Kubernetes documents layouts where nodefs, imagefs, and, where present, containerfs may map to the same underlying filesystem or to distinct filesystems. That affects which resource a signal describes and where image and writable-layer data consume capacity. Do not assume every cluster stores image data in the same place as node files or container writable layers.
Rank #3
- Easily store and access 1TB to content on the go with the Seagate Portable Drive, a USB external hard drive.Specific uses: Personal
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop. Reformatting may be required for Mac
- To get set up, connect the portable hard drive to a computer for automatic recognition no software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
How to investigate an inode-related “no space left” failure
- Check bytes and inodes separately. Compare ordinary byte-usage information with
df -ion the affected node. A filesystem showing available bytes but 100% inode use points to a file-count constraint rather than ordinary byte exhaustion. - Identify which filesystem is full. Establish whether the affected path is on the node filesystem, image filesystem, or another mounted filesystem. Kubernetes’ signal names depend on the node’s filesystem layout.
- Inspect file-heavy workloads and image versions. Look for images that package large dependency trees, caches, generated files, or many small files, and check how many old image versions remain on nodes.
- Review kubelet eviction configuration. Verify that the intended free-inode thresholds are explicitly present or that the deployed release’s default-merging behavior is configured as expected. Include the corresponding byte thresholds that the cluster requires.
- Check monitoring and cleanup policies. Make sure inode use or free-inode signals are visible and alerted on, and determine whether cleanup depends only on byte-use thresholds.
What Shinder says changed
Shinder describes a bundle of changes rather than a controlled comparison of individual fixes:
- The node image was changed to format the container volume as XFS; the author says this allocates inodes as needed. That filesystem claim and its suitability for a particular deployment are not established by the Kubernetes documentation cited here.
- Node panels and matching alerts were updated to show inode use alongside byte use.
- Kubelet configuration was changed to specify each eviction threshold explicitly, including free-inode thresholds.
- The frontend image stopped shipping
node_modules; dependencies were obtained from a cache volume at job time. Shinder reports that the image’s file count fell from about 400,000 to fewer than 9,000. - Unused images were removed after three days regardless of byte usage.
These are the author’s reported remedies, not universal prescriptions. Filesystem choice, image delivery, cache design, kubelet settings, and cleanup rules all depend on the cluster’s storage layout and operational requirements. The cited sources do not provide a benchmark comparing these approaches.
Rank #4
- Easily store and access 4TB of content on the go with the Seagate Portable Drive, a USB external hard drive.Specific uses: Personal
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop
- To get set up, connect the portable hard drive to a computer for automatic recognition no software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
The practical lesson for node capacity planning
Disk capacity is not a single number. Byte utilization, inode availability, image retention, and the number of files in workloads all affect whether a node can continue creating files. Shinder’s account illustrates why a byte-only dashboard and byte-triggered cleanup policy can overlook a node approaching inode exhaustion. Kubernetes’ separate inode signals provide a way to make that resource visible and include it in eviction planning.
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 problemsQuick Recap
Best Value
- [Upgraded Version] - This external hard drive features a mirrored logo stripe combined with a striped anti-slip design, and the rounded corners of the casing make it easier to grip. The stripes also have a heat dissipation function, ensuring stable and fast data transfer.
- 【Ultra-thin and quiet】 - The motherboard adopts JMicron 578 noise-free solution, giving you a quiet working environment. Lightweight and portable size designed to fit in your pocket for easy portability.
- 【Ultra-Fast Data Transfers】 - Pairing this external hard drive with JMicron 578 solution USB 3.0 and USB 2.0 interfaces enables blazing-fast data transfer. It boasts theoretical read speeds of up to 125MB/s and write speeds of up to 103MB/s.
- 【Plug and Play】 - With no software to install, just plug it in and the drive is ready to use.The hard disk chip is wrapped with an aluminum anti-interference layer to increase heat dissipation and protect data.
- 【What You Get】 - 1 x Portable Hard Drive, 1 x USB 3.0 Cable, 1 x User Manual, Gift-type shell packaging ,Three-year manufacturer's warranty and free technical support services.
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.
