When running Jenkins in a managed environment, it is important to know (like all web application) whether it is healthy or not.
Jenkins doesn't offer any dedicated endpoint for that, and most users currently rely on an arbitrary URL like $JENKINS_URL/login or $JENKINS_URL/whoAmI/api/json?tree=authenticated.
The login endpoint is problematic for several reasons:
- It creates a throw-away session which causes memory waste
- In some security realms, this endpoint is actually unused as it doesn't used password-based authentication. But as many people use it for healthcheck, it makes it impossible to removed.
Having a normalized endpoint would allow core or plugin implementations to contribute their own healthchecks. In CloudBees CI for example, we need a clustering subsystem to be online to consider the Jenkins instance as healthy. If it crashes for any reason, we want a probe to fail so that it would automatically reschedule the instance.
Originally reported by 
vlatombe, imported from: Dedicated and pluggable health-check endpoint
- assignee: vlatombe
- status: Closed
- priority: Major
- component(s): core
- resolution: Fixed
- resolved: 2025-04-15T08:04:07+00:00
- votes: 0
- watchers: 2
- imported: 2025-11-24
Raw content of original issue
When running Jenkins in a managed environment, it is important to know (like all web application) whether it is healthy or not.
Jenkins doesn't offer any dedicated endpoint for that, and most users currently rely on an arbitrary URL like $JENKINS_URL/login or $JENKINS_URL/whoAmI/api/json?tree=authenticated.
The login endpoint is problematic for several reasons:
- It creates a throw-away session which causes memory waste
- In some security realms, this endpoint is actually unused as it doesn't used password-based authentication. But as many people use it for healthcheck, it makes it impossible to removed.
Having a normalized endpoint would allow core or plugin implementations to contribute their own healthchecks. In CloudBees CI for example, we need a clustering subsystem to be online to consider the Jenkins instance as healthy. If it crashes for any reason, we want a probe to fail so that it would automatically reschedule the instance.
When running Jenkins in a managed environment, it is important to know (like all web application) whether it is healthy or not.
Jenkins doesn't offer any dedicated endpoint for that, and most users currently rely on an arbitrary URL like $JENKINS_URL/login or $JENKINS_URL/whoAmI/api/json?tree=authenticated.
The login endpoint is problematic for several reasons:
Having a normalized endpoint would allow core or plugin implementations to contribute their own healthchecks. In CloudBees CI for example, we need a clustering subsystem to be online to consider the Jenkins instance as healthy. If it crashes for any reason, we want a probe to fail so that it would automatically reschedule the instance.
Originally reported by
vlatombe, imported from: Dedicated and pluggable health-check endpoint
Raw content of original issue
Having a normalized endpoint would allow core or plugin implementations to contribute their own healthchecks. In CloudBees CI for example, we need a clustering subsystem to be online to consider the Jenkins instance as healthy. If it crashes for any reason, we want a probe to fail so that it would automatically reschedule the instance.