0.8

0.80.70.60.5

›Troubleshooting

Getting Started

  • Quick Start Guide
  • Introduction
  • MayaOnline
  • Minikube
  • Getting Help

Concepts

  • CAS
  • Architecture
  • Features
  • Maya
  • CAS Engines
  • Jiva
  • cStor
  • Storage Policies

Setup

  • Platforms
  • Prerequisites
  • Scheduler
  • Installation
  • Uninstall OpenEBS
  • Provisioning Jiva
  • Provisioning cStor
  • Upgrade
  • Storage Pools
  • Storage Classes
  • Storage Volumes
  • Snapshots

Usecases

  • Blue/Green strategy
  • Snapshots in CI/CD

Integrations

  • OpenShift
  • StackPointCloud
  • AWS
  • IBMCloud
  • Rancher

Databases

  • Cassandra
  • CockroachDB
  • Couchbase Server
  • Crunchy Postgres
  • Jenkins
  • Jupyter
  • MongoDB
  • PerconaDB
  • Percona Galera Cluster
  • RabbitMQ
  • Redis StatefulSet

Tasks

  • mayactl
  • Deploy across AZs
  • Volume provisioning

FAQs

  • FAQ

Troubleshooting

  • Overview
  • ReadOnly Volumes
  • Collecting Logs
  • Installation
  • Persistent Volumes

Releases

  • Changelog

Contributor Resources

  • Open source Meetings
  • Contributor's Guide
EditCreate An Issue

Collecting OpenEBS Logs


Overview of OpenEBS Logger

Logger is a Kubernetes job which can be run on a cluster to extract pod logs and cluster information. It helps in troubleshooting/debugging activities. Logger runs the logger container openebs/logger and it is recommended to run for a specific duration to capture logs while attempting to reproduce issues.

Benefits

Following are the benefits of using Logger.

  • The purpose of Logger is to obtain debug-information/quick logs in clusters where more standard logging frameworks like EFK are not already configured. Logger creates a simple support bundle which can be provided to debug teams.
  • This may be the case with most "non-production/development" infrastructures.

Where is the Logger found?

You can find the Logger file here.

How does Logger work?

Logger uses the following tool/command to work.

  • Logger uses a stern tool to collect the pod logs.
  • It uses kubectl commands to extract cluster information.

Prerequisite

Logger requires the kubeconfig file mounted as a configmap (passed to stern binary). kubeconfig file is generally found under /etc/kubernetes/admin.conf or ~/.kube/config path.

Running Logger

The following procedure helps you run Logger.

  1. In the logger job's command, edit the logging duration (-d) and pod regex (-r) to specify which pods' logs should be captured and for how long.

    For example, in the ./logger.sh -d 5 -r maya,openebs,pvc; command, the logs for pods starting with literals "maya", "openebs" and "pvc" are captured for a period of 5 minutes.

    Note: The duration is arrived depending on the average time taken for the issue/bug to manifest from the time a pod starts.

  2. Create a Kubernetes job to run logger using the kubectl apply -f debugjob.yaml command.

  3. This job will run for the duration specified in the previous steps.

  4. The logs thus collected are placed in a logbundle (tarball) in /mnt directory of the node in which the debug pod was scheduled.

  5. Logs will be available in the node in which the debug pod/logger is scheduled when you run a kubectl get pod -o wide command.

  6. Attach this log support bundle while raising issues on the OpenEBS repository.

    ​

← Troubleshooting readonly volumes in OpenEBSTroubleshooting OpenEBS Installation →
  • Running Logger

Get in touch with OpenEBS community via Slack

mayadata.io
Get OpenEBS support through MayaOnline