I am running Bitbucket, Confluence and Jira Server as Docker containers, all connecting to a Crowd server also as a Docker container.
I recently replaced the Crowd Docker Container, changing it from alvistack/Crowd to Atlassian/Crowd, ran a clean install and then imported my last Crowd backup.
I can log in to Crowd with my user, and can also do the same with Bitbucket.
However, I can't log in via Jira or Confluence. The application configuration in Crowd, and in the crowd.properties for both Jira and Confluence are unchanged and match. The URL to access Crowd is the same (https://crowd.domain.uk).
The Crowd API endpoint is at https://crowd.domain.uk/crowd/rest
I suspect that Jira and Confluence are trying to call the API at https://crowd.domain.uk/rest instead of /crowd/rest, but I have no way of knowing. Where would I change this to check that this is correct?
I have checked authentication of the user for each application within Crowd, and the authentication is successful. The issue appears to be with communicating with Crowd itself.
However, that atlassian-confluence.log displays the following. The jira-security.log is less detailed but also states: user.name tried to login but they do not have USE permission or weren't found. Deleting remember me cookie.
2020-03-15 16:12:50,650 WARN [http-nio-8090-exec-13] [atlassian.seraph.auth.DefaultAuthenticator] login login : 'user.name' tried to login but they do not have USE permission or weren't found. Deleting remember me cookie.
-- referer: https://confluence.domain.uk/dologin.action | url: /dologin.action | traceId: 1996c261e88dda12
2020-03-15 16:12:50,845 INFO [http-nio-8090-exec-13] [confluence.security.login.DefaultLoginManager] recordLoginFailure
Failed login attempt for user 'user.name':
Request URL: https://confluence.domain.uk/dologin.action
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:74.0) Gecko/20100101 Firefox/74.0
Remote Address: 172.18.0.1
X-Forwarded-For: 192.168.1.3
I was correct in realising that the crowd.properties file needed updating, to insert the crowd directory in the file structure:
application.name {application.name}
application.password {application.password}
application.login.url https://crowd.domain.uk/crowd/console/
crowd.server.url https://crowd.domain.uk/crowd/services/
crowd.base.url https://crowd.domain.uk/crowd/
previously the file was configured as:
application.name {application.name}
application.password {application.password}
application.login.url https://crowd.domain.uk/console/
crowd.server.url https://crowd.domain.uk/services/
crowd.base.url https://crowd.domain.uk/
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.