k8s-pod-restart-info-collector

k8s-pod-restart-info-collector is a simple K8s customer controller that watches for Pods changes and collects K8s Pod restart reasons, logs, and events to Slack channel when a Pod restarts.

For more information, see the blog on Medium: Automated Troubleshooting of Kubernetes (K8s) Pods Issues

This project is actively used and maintained by Airwallex DevOps team.

Overview of the Data Collected

Here are two Slack screenshots of the example messages.

Brief Alert Message

Detailed Alert Message

As shown below, by clicking “Show more”, we can see the Reason, “Pod Status”, “Pod Events”, “Node Status and Events”, and “Pod Logs Before Restart”.

How to test and develop locally

export SLACK_WEBHOOK_URL=https://hooks.slack.com/services/T0SEVS2SG/xxxxx
go run .

Install using Helm

Replace the slackWebhookUrl, clusterName and slackChannel.

helm upgrade --install k8s-pod-restart-info-collector ./helm \
   --set slackWebhookUrl="https://hooks.slack.com/services/Change-Me" \
   --set clusterName="Change-Me" \
   --set slackChannel="Change-Me"

Check Commands:

# check commands
kubectl get pod,deploy,sa,secret -l app.kubernetes.io/instance=k8s-pod-restart-info-collector
helm status k8s-pod-restart-info-collector
helm get values k8s-pod-restart-info-collector
helm get manifest k8s-pod-restart-info-collector
helm get all k8s-pod-restart-info-collector
# see logs
kubectl logs deployment/k8s-pod-restart-info-collector -f

Run a debug-pod to verify the collector:

kubectl run debug-pod --image=alpine -- date;sleep 30
kubectl get pod debug-pod -w

Uninstall

To uninstall/delete the k8s-pod-restart-info-collector helm release:

helm uninstall k8s-pod-restart-info-collector

The command removes all the Kubernetes components associated with the chart and deletes the release.

Helm Parameters

Name	Description	Value
clusterName	K8s cluster name (Display on slack message)	required
slackWebhookUrl	Slack webhook URL	required
slackUsername	Slack username (Display on slack message)	default: "k8s-pod-restart-info-collector"
slackChannel	Slack channel name	default: "restart-info-nonprod"
muteSeconds	The time to mute duplicate pod alerts	default: "600"
ignoredNamespaces	A comma-separated list of namespaces to ignore	default: ""
ignoredPodNamePrefixes	A comma-separated list of pod name prefixes to ignore	default: ""

FAQ

1. When will the collector send Pod restart messages to Slack channel?

   When a Pod restarts. However, if one of the following conditions is met, the messages are not sent.

   1. Pod restartCount > 30
   2. In the previous 10 minutes, the same Pod restart message was sent

2. How to customize slack channel for each pods

   Adding alert-slack-channel: "your-slack-channel-name" to Pod annotations or labels. For example, a label: alert-slack-channel: "restart-info-nonprod"

3. How to set the kubernetes icon as the Slack profile photo?

   The alert messages use a custom emoji naming :kubernetes: as the profile photo. You need to upload the kubernete.png to Slack to add this custom emoji. Please refer to Add custom emoji

How to write a K8s controller

Please refer to:

• https://github.com/kubernetes/sample-controller/blob/master/docs/controller-client-go.md
• https://github.com/kubernetes/community/blob/master/contributors/devel/sig-api-machinery/controllers.md
• https://github.com/kubernetes/client-go/tree/master/examples/workqueue

Copyright and license

Copyright [2022] [Airwallex (Hong Kong) Limited]

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.