atlas-run build failure

Question

[INFO] [talledLocalContainer] Dec 14, 2018 10:54:54 AM org.apache.jasper.servlet.TldScanner scanJars
[INFO] [talledLocalContainer] INFO: At least one JAR was scanned for TLDs yet contained no TLDs. Enable debug logging for this logger for a complete list of JARs that were scanned but no TLDs were found in them. Skipping unneeded JARs during scanning can improve startup time and JSP compilation time.
[INFO] [talledLocalContainer] Dec 14, 2018 10:54:54 AM org.apache.catalina.startup.HostConfig deployDirectory
[INFO] [talledLocalContainer] INFO: Deployment of web application directory [C:\Users\Olivera.Tancheva\olivera3\myPlugin	arget\container	omcat8x\cargo-jira-home\webapps\manager] has finished in [1,332] ms
[INFO] [talledLocalContainer] Dec 14, 2018 10:54:54 AM org.apache.coyote.AbstractProtocol start
[INFO] [talledLocalContainer] INFO: Starting ProtocolHandler ["http-nio-50241"]
[INFO] [talledLocalContainer] Dec 14, 2018 10:54:54 AM org.apache.coyote.AbstractProtocol start
[INFO] [talledLocalContainer] INFO: Starting ProtocolHandler ["ajp-nio-50243"]
[INFO] [talledLocalContainer] Dec 14, 2018 10:54:54 AM org.apache.catalina.startup.Catalina start
[INFO] [talledLocalContainer] INFO: Server startup in 635890 ms
[INFO] ------------------------------------------------------------------------
[INFO] BUILD FAILURE
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 25:26 min
[INFO] Finished at: 2018-12-14T11:04:18+01:00
[INFO] Final Memory: 33M/110M
[INFO] ------------------------------------------------------------------------
[ERROR] Failed to execute goal com.atlassian.maven.plugins:maven-amps-dispatcher-plugin:6.3.21:run (default-cli) on project myPlugin: Unable to execute mojo: Cannot start container [org.codehaus.cargo.container.tomcat.Tomcat8xInstalledLocalContainer@1805b503]: Deployable [http://localhost:50241/cargocpc/index.html] failed to finish deploying within the timeout period [600000]. The Deployable state is thus unknown. -> [Help 1]
[ERROR]
[ERROR] To see the full stack trace of the errors, re-run Maven with the -e switch.
[ERROR] Re-run Maven using the -X switch to enable full debug logging.
[ERROR]
[ERROR] For more information about the errors and possible solutions, please read the following articles:
[ERROR] [Help 1] http://cwiki.apache.org/confluence/display/MAVEN/MojoExecutionException
Exception in thread "Thread-2" java.lang.NoClassDefFoundError: com/google/common/collect/Lists$ReverseList
        at com.google.common.collect.Lists.reverse(Lists.java:729)
        at com.atlassian.maven.plugins.amps.AbstractProductHandlerMojo.stopProducts(AbstractProductHandlerMojo.java:907)
        at com.atlassian.maven.plugins.amps.RunMojo$1.run(RunMojo.java:107)
        at java.base/java.lang.Thread.run(Thread.java:834)
Caused by: java.lang.ClassNotFoundException: com.google.common.collect.Lists$ReverseList
        at org.codehaus.plexus.classworlds.strategy.SelfFirstStrategy.loadClass(SelfFirstStrategy.java:50)
        at org.codehaus.plexus.classworlds.realm.ClassRealm.unsynchronizedLoadClass(ClassRealm.java:259)
        at org.codehaus.plexus.classworlds.realm.ClassRealm.loadClass(ClassRealm.java:235)
        at org.codehaus.plexus.classworlds.realm.ClassRealm.loadClass(ClassRealm.java:227)
        ... 4 more
C:\Users\Olivera.Tancheva\olivera3\myPlugin>

Bryan Turner · Answer

The build failure was because Jira failed to start in time. AMPS has a timeout that it applies when starting products, and if the product hasn't finished starting in within that time, AMPS gives up. That's shown by this output:

[ERROR] Failed to execute goal com.atlassian.maven.plugins:maven-amps-dispatcher-plugin:6.3.21:run (default-cli) on project myPlugin: Unable to execute mojo: Cannot start container [org.codehaus.cargo.container.tomcat.Tomcat8xInstalledLocalContainer@1805b503]: Deployable [http://localhost:50241/cargocpc/index.html] failed to finish deploying within the timeout period

I helped an app developer who was running into a similar error when using AMPS (or the Atlassian SDK) to start Bitbucket Server. In their case, the error was caused by aggressive anti-virus software which was making Bitbucket Server start extremely slowly. Reconfiguring the anti-virus software allowed Bitbucket Server to go from failing to start in 10 minutes to starting without issue in just over 1 minute. (The user in question was on a Mac, and running "top -u" while attempting to start Bitbucket Server to see what process, or processes, was using CPU helped isolate anti-virus as the cause.)

The ClassNotFoundException for ReverseList is an unrelated bug in AMPS (which I fixed as part of AMPS 8.0, which is the default version used by the latest 8.0.4 version of the Atlassian SDK). It is not causing the build failure; it's happening because the build has already failed. When products are started, AMPS registers a JVM shutdown hook to try and ensure they're stopped. It was using a Guava utility to try and shut down products in reverse order, but by that point during JVM execution new classes can't be loaded (since the JVM is shutting down), and so it failed. AMPS 8 uses Collections.reverse instead.

Sarkis Chakarian_xf5j3n · Answer

@Olivera Tancheva See if this thread fixes your issue: https://community.atlassian.com/t5/Jira-questions/java-lang-ClassNotFoundException-ReverseList-During-create/qaq-p/951874#U977381