You are not logged in. Click here to log in.

codebeamer Application Lifecycle Management (ALM)

Search In Project

Search inClear

Configuring Docker Health Checks Before Upgrading Large Instances

Introduction

During the execution of large upgrades, for example when upgrading to codebeamer release 22.10-LTS (GINA), health checks should be taken into consideration since these checks might report the codebeamer instance as unhealthy.

If an automatic healing mechanism kills the unhealthy codebeamer instance during an upgrade, the upgrade fails. Such an issue can be hard to investigate as the codebeamer instance does not die from a problem in the application, consequently, application logs do not contain any relevant information about the failure.


General Case

In codebeamer, the default health check endpoint is <url>/<contextPath>/hc/ping.spr. For example: https://codebeamer.com/cb/hc/ping.spr.

Automatic healing mechanisms that depend on this endpoint should be disabled before large upgrades, otherwise, the upgrade fails the reasons for which do not show up in the application logs.

Using Docker

In the codebeamer Docker images shipped, there is a health check script at /home/appuser/healthcheck.sh. This script uses the default codebeamer health check endpoint: <url>/<contextPath>/hc/ping.spr.

Automatic healing mechanisms that depend on this script should be disabled before large upgrades, otherwise, the upgrade fails the reasons for which do not show up in the application logs.