Confluence server not starting up with version 7.19.17

[Webmaster Asha]

We are trying upgrade our confluence data center from 7.19.16 for CVE update.  We are using docker containers for running the confluence.  After the upgrade we are getting HTTP Error 500 and can't find any errors in the log files.  Verified the log file in the docker container /opt/atlassian/confluence/logs.  I cannot find the catalina.out in the directory but the catalina-XXXX-XX-XX.log does not have any errors.  Any troubleshooting tips?

[Alan Stephenson]

Check the version and build numbers. I remember this from previous versions of confluence. I’m not sure if this will be your issue. Can’t hurt to check.

[Nic Brough -Adaptavist-]

500 means that either the service is not running, or that it is not running on the url you are trying.

Start with a look at the process manager, to see if it is running, then look at the application log for errors. 

You say you have no Catalina.out, that suggests Tomcat is not even starting, in which case I would try to start it manually and read the output from the start script.

[Webmaster Asha]

I know that tomcat is running.  nginx is giving a different page when I stop the confluence container.  I did see the following error in the logs ..

2024-01-04 01:45:27,082 ERROR [http-nio-8090-exec-5 url: /500page.jsp] [atlassian.confluence.status.SystemErrorInformationLogger] logException Unhandled exception fa875ffd-e9d9-4161-8230-fd31b16ba2b2: javax.servlet.ServletException: org.apache.velocity.exception.ResourceNotFoundException: Resource name must end with .vm, .vmd, .css or .xml
-- traceId: c25ef389374f5e81
java.lang.RuntimeException: javax.servlet.ServletException: org.apache.velocity.exception.ResourceNotFoundException: Resource name must end with .vm, .vmd, .css or .xml

Seems like jsp servlet did not start 

Before that I see this error

2024-01-04 01:45:27,030 WARN [http-nio-8090-exec-5 url: /, /dashboard.action] [confluence.impl.hibernate.ConfluenceHibernateTransactionManager] doRollback Performing rollback. Transactions:
->[null]: PROPAGATION_REQUIRED,ISOLATION_DEFAULT (Session #1052729554)
-- url: / | traceId: c25ef389374f5e81 | userName: anonymous
2024-01-04 01:45:27,049 ERROR [http-nio-8090-exec-5 url: /, /dashboard.action] [ContainerBase.[Standalone].[localhost].[/]] log Unhandled exception occurred whilst decorating page
-- url: / | traceId: c25ef389374f5e81 | userName: anonymous

This is the first error in the log

2024-01-04 01:45:25,707 ERROR [http-nio-8090-exec-1 url: /, /dashboard.action] [ContainerBase.[Standalone].[localhost].[/]] log Unhandled exception occurred whilst decorating page
-- url: / | traceId: 7062eca51ba15e7b | userName: anonymous
java.lang.RuntimeException: javax.servlet.ServletException: org.apache.velocity.exception.ResourceNotFoundException: Resource name must end with .vm, .vmd, .css or .xml

[Nic Brough -Adaptavist-]

So you do have a catalina.out log and you can find errors in the log files.

The errors in the application log are halting the application (but not Tomcat) because they are terminal.

The error you have first tells us that your database is not configured correctly.  The first error you find in a log is almost always followed by more errors which drill down to the underlying cause.

In this case, you've found the root in your second error message.  Either your database is misconfigured, or you are on a version of Confluence affected by the bug https://jira.atlassian.com/browse/CONFSERVER-56094