When I try to install the Jenkins Integration for Jira Cloud app using a private listing link on https://marveloud.atlassian.net I get this error message:
The app host returned HTTP response code 500 when we tried to contact it during installation. Please try again later or contact the app vendor.
However, I don’t see any traffic to my app. I have logging in-place that log any error that is returned form the lifecycle install and uninstall hooks.
To make iteven more interesting, other customers of Jira Cloud don’t seem to be effected because new evaluations continue to be started on the app.
Any ideas where I should be able to find the cause of this behaviour?
Cheers,
Mark
What is the baseurl in your descriptor? I would verify that you can execute a POST against the public installation url.
Hi @danielwester,
The baseUrl is https://jjc.marvelution.com, and I’m able to execute POST requests on the lifecycle endpoints.
Another fun addition to debugging this is that I have the same code running in a staging environment which is running just find. The different between staging and production is basically different database, log levels and number of consumers to handle synchronization requests.
Cheers,
Mark
I think I found the cause.
I’m using the Atlassian Connect Spring Boot starter and the issue seems to related to my upgrade to version 1.5.0 to support the GDPR changes.
The request is failing with the message:
Could not find an installed host for the provided client key: 79e31ab7-09ee-3eb1-ae91-654b60789182
And this is thrown by the JwtAuthenticationProvider
before the LifecycleController has a change to install this new host.
Still wird that the staging environment is not effect be this, since its the same code and authz configuration.
I found the cause, the Atlassian Host https://marveloud.atlassian.net IS signing the installed lifecycle callback event, and the Spring boot starter cannot handle this by default for production. Seems like a bug to me, and I have opened a support ticket with Atlassian support for this.
We’re having the same issue. How were you able to resolve this?
Hi @mohammed,
I’m using the ‘atlassian-connect-spring-starter’ and I temporarily enabled the ‘allowReinstallMissingHost’ flag to reinstall the cloud tenant so the future will work again even after disabling the flag.
Was the only thing I could think of to do.
Hope it helps.
Cheers.
4 Likes