Uploaded image for project: 'Red Hat Fuse'
  1. Red Hat Fuse
  2. ENTESB-15443

Fabric8 Maven plug-in won't deploy on OpenShift 4.5.22 and later

    XMLWordPrintable

Details

    • Bug
    • Resolution: Duplicate
    • Major
    • fuse-7.8-GA
    • fuse-7.7-GA
    • Fuse on Openshift
    • None
    • False
    • False
    • % %
    • Undefined
    • Hide

      Use any boilerplate Fuse-on-SpringBoot application (I can supply one if needed) based on Fuse 7.7, and using the FMP version specified in the relevant Fuse BOM.

      $ mvn clean package fabric8:deploy 

      A number of alternative error messages will be produced, for example:

      [DEBUG] Connecting websocket ... io.fabric8.kubernetes.client.dsl.internal.WatchConnectionManager@3c2fa57a
[WARNING] Exec Failure: HTTP 500, Status: 500 - invalid upgrade response: status code 200java.net.ProtocolException: Expected HTTP 101 response but was '500 Internal Server Error'
at okhttp3.internal.ws.RealWebSocket.checkResponse(RealWebSocket.java:219)
at okhttp3.internal.ws.RealWebSocket$2.onResponse(RealWebSocket.java:186)
at okhttp3.RealCall$AsyncCall.execute(RealCall.java:153)
at okhttp3.internal.NamedRunnable.run(NamedRunnable.java:32)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)

      All the failure modes have the effect that the FMP quits whilst the builder pod is still at work. We end up with a viable image being pushed to the internal OpenShift registry, but the rest of the deployment does not take place.

       

       

      Show
      Use any boilerplate Fuse-on-SpringBoot application (I can supply one if needed) based on Fuse 7.7, and using the FMP version specified in the relevant Fuse BOM. $ mvn clean package fabric8:deploy  A number of alternative error messages will be produced, for example: [DEBUG] Connecting websocket ... io.fabric8.kubernetes.client.dsl.internal.WatchConnectionManager@3c2fa57a [WARNING] Exec Failure: HTTP 500, Status: 500 - invalid upgrade response: status code 200java.net.ProtocolException: Expected HTTP 101 response but was '500 Internal Server Error' at okhttp3.internal.ws.RealWebSocket.checkResponse(RealWebSocket.java:219) at okhttp3.internal.ws.RealWebSocket$2.onResponse(RealWebSocket.java:186) at okhttp3.RealCall$AsyncCall.execute(RealCall.java:153) at okhttp3.internal.NamedRunnable.run(NamedRunnable.java:32) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617) at java.lang. Thread .run( Thread .java:745) All the failure modes have the effect that the FMP quits whilst the builder pod is still at work. We end up with a viable image being pushed to the internal OpenShift registry, but the rest of the deployment does not take place.    

    Description

      As of OpenShift 4.5.22, the Fabric8 Maven plug-in seems to be unable to deploy Fuse 7 applications, even trivial ones. I've tested with 4.6.6, and the customer has tested with 4.5.22. We've tested with full OpenShift installations and also with CRC. The error messages produced do vary a bit, but the end result is the same – the builder pod is created, and runs to completion. By that time, however, the FMP has failed with an error.

      I don't think this is just a timeout problem – my tests on 4.5.20 don't run noticeably quicker than later version. The problem is completely reproducible – after dozens of attempts, I haven't had one successful deployment.

      In addition, I sometimes have a failure where the exception backtrace displayed by the FMP is interspersed with nonsense text. It looks as if the FMP is receiving complete gobbledegook from OpenShift.

      Attachments

        Issue Links

          is related to

          Bug - A problem that impairs or prevents the functions of the product. ENTESB-15307 FMP not working with OCP4.6

          • Blocker - To be worked above all other priorities. This term is chosen when the severity of the issue is very high, has no workaround, or the effort for the change is comparatively low. Issues that may be very publicly visible and could generate significant media attention may also drive a higher priority.
          • Closed

          Activity

            People

              ldemasi Luigi De Masi
              rhn-support-kboone Kevin Boone
              Votes:
              0 Vote for this issue
              Watchers:
              3 Start watching this issue

              Dates

                Created:
                Updated:
                Resolved: