https://github.com/maxlaverse/soft-pod-memory-evicter

Gracefully evict Kubernetes Pods before they get OOM killed
https://github.com/maxlaverse/soft-pod-memory-evicter

eviction kubernetes memory oom pod

Last synced: about 1 year ago
JSON representation

Gracefully evict Kubernetes Pods before they get OOM killed

• Host: GitHub
• URL: https://github.com/maxlaverse/soft-pod-memory-evicter
• Owner: maxlaverse
• License: apache-2.0
• Created: 2022-07-02T18:03:52.000Z (almost 4 years ago)
• Default Branch: main
• Last Pushed: 2025-05-19T19:57:18.000Z (about 1 year ago)
• Last Synced: 2025-06-01T15:46:59.923Z (about 1 year ago)
• Topics: eviction, kubernetes, memory, oom, pod
• Language: Go
• Homepage:
• Size: 266 KB
• Stars: 11
• Watchers: 2
• Forks: 4
• Open Issues: 5
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# soft-pod-memory-evicter

![Tests](https://github.com/maxlaverse/soft-pod-memory-evicter/actions/workflows/tests.yml/badge.svg?branch=main)

![Go Version](https://img.shields.io/github/go-mod/go-version/maxlaverse/soft-pod-memory-evicter)

![Releases](https://img.shields.io/github/v/release/maxlaverse/soft-pod-memory-evicter?include_prereleases)

A Kubernetes Controller that proactively evicts Pods approaching their memory limits, enabling graceful shutdowns before OOM (Out of Memory) termination occurs.

## Supported Versions

The controller has been tested and built with the following components:

- Kubernetes >= 1.32.1

- Metric Server >= 0.7.2

## Installation

```bash

helm repo add maxlaverse https://maxlaverse.github.io/helm-charts/

helm repo update

helm install soft-pod-memory-evicter maxlaverse/soft-pod-memory-evicter

```

The controller can be configured by setting Helm values under `extraArgs: []` with any of the options listed in the [Usage](#usage) section below. For a complete list of configurable Helm values, please refer to the [values.yaml][values] file.

## Usage

```

NAME:

   soft-pod-memory-evicter - Gracefully evict Pods before they get OOM killed

USAGE:

   soft-pod-memory-evicter [global options] command [command options]

COMMANDS:

   help, h  Shows a list of commands or help for one command

GLOBAL OPTIONS:

   --dry-run                                              Output additional debug lines (default: false)

   --pod-selector value                                   Evict only Pods matching this label selector

   --eviction-pause value                                 Pause duration between evictions (default: 5m0s)

   --memory-usage-check-interval value                    Interval at which the Pod metrics are checked (default: 3m0s)

   --memory-usage-threshold value                         Memory usage eviction threshold (0-100) (default: 95)

   --channel-queue-size value                             Size of the queue for pod eviction (default: 100)

   --ignore-namespace value [ --ignore-namespace value ]  Do not evict Pods from this namespace. Can be used multiple times

   --loglevel value, -v value                             Log Level (default: 0)

   --help, -h                                             show help

```

## Pod-specific configurations

Pods can be configured individually using annotations to override the default behavior:

- `soft-pod-memory-evicter.laverse.net/memory-usage-threshold`: Specifies the memory usage threshold (0-100) at which the Pod should be evicted. This value overrides the global threshold set by the `--memory-usage-threshold` flag.

Example:

```yaml

apiVersion: v1

kind: Pod

metadata:

  annotations:

    soft-pod-memory-evicter.laverse.net/memory-usage-threshold: "80"

```

## License

Distributed under the Apache License. See [LICENSE](./LICENSE) for more information.

[values]: https://github.com/maxlaverse/helm-charts/blob/main/charts/soft-pod-memory-evicter/values.yaml