DX Application Performance Management

 View Only
  • 1.  DX APM 11.1 Installation - deletes the master docker registry

    Posted Aug 27, 2019 04:16 AM
    Hello.
    During the installation process I have my kubernetes cluster messed up.
    Reason is that all the kubernetes images are deleted locally in the docker local repository on the master machine.
    Do you have any idea why this happens
    Both using a private registry for dxi or a new one

    This is installation log............
    at least one uppercase character
    at least one lowercase character
    at least one number, and
    at least one special character): **********
    Re-enter the password for the masteradmin: **********
    The postgres database password has been stored as Base64 encoded value in dxi.input.vars.yml file.
    The password for GA Notify Filter account has been generated and will be saved to dxi.input.vars.yml
    Successfully written properties into /root/./dxi.input.vars.yml file.
    Successfully written properties into /root/./inventory/host_vars/node1.yml file.
    Successfully written properties into /root/./inventory/group_vars/all.yml file.
    Successfully written properties into /root/./inventory/hosts.ini file.
    Console user input collection completed successfully
    Handling registry setup.
    Starting DownloaderApplication on server with PID 1 (/root/bin/console.jar started by root in /root)
    The following profiles are active: registry
    [registry] Non-installation mode.
    Started DownloaderApplication in 2.535 seconds (JVM running for 3.176)
    Pulling docker images.
    Detected offline mode - importing repository.
    registryError processing tar file(exit status 1): write /387269f2a5aec64420fcbca5150ee4fbf22acd9a8f2218f3688c3051295ef16f/layer.tar: no space left on device
    Pushing docker images.
    Unable to find image 'java-runner:2.0' locally
    Trying to pull repository registry.access.redhat.com/java-runner ...
    Trying to pull repository docker.io/library/java-runner ...
    Trying to pull repository registry.fedoraproject.org/java-runner ...
    Trying to pull repository quay.io/java-runner ...
    Trying to pull repository registry.centos.org/java-runner ...
    Trying to pull repository docker.io/library/java-runner ...
    /usr/bin/docker-current: Get https://registry-1.docker.io/v2/: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers).
    See '/usr/bin/docker-current run --help'.
    Importing DXI metadata.
    Unable to find image 'java-runner:2.0' locally
    Trying to pull repository registry.access.redhat.com/java-runner ...
    Trying to pull repository docker.io/library/java-runner ...
    Trying to pull repository registry.fedoraproject.org/java-runner ...
    Trying to pull repository quay.io/java-runner ...
    Trying to pull repository registry.centos.org/java-runner ...
    Trying to pull repository docker.io/library/java-runner ...
    /usr/bin/docker-current: Get https://registry-1.docker.io/v2/: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers).
    See '/usr/bin/docker-current run --help'.


  • 2.  RE: DX APM 11.1 Installation - deletes the master docker registry

    Posted Aug 28, 2019 05:18 PM

    Hello,
    nice to see someone was able to load the installation packages.
    The packages for DX APM 11 are remove since some days from the product download site.
    Broadcom will fix some problems.
    Lutz



    ------------------------------
    Finanz Informatik
    ------------------------------



  • 3.  RE: DX APM 11.1 Installation - deletes the master docker registry

    Posted Aug 29, 2019 03:13 AM
    Hello.
    I understood that:
    • Installation machine must not be a member ofthe kubernetes cluster but can push images from outside kubernetes. In this way I get no problems on the master registry.
    • I used an external registry to start the installation otherwise the private registry willy be started on localhost:5000 failing the installation
    • kubernetes nodes must have 32 GB of memory or pods wont start (MemoryPressure) 
    • kubernetes nodes must have at least 50 GB of disk space on /var because docker will mount containers over there and youll get DiskPressure on the node failing the pods to start
    Anyway I get an error at the end...

    2019-08-28 13:22:53,781 INFO - [create] [ OK ] Ingress ng-acc-configserver-ingress-8088
    2019-08-28 13:22:53,800 INFO - [create] [ OK ] Service ng-acc-repository-service
    2019-08-28 13:22:53,816 INFO - [create] [ OK ] Service ng-acc-configserver-service
    2019-08-28 13:22:53,836 INFO - [create] [ OK ] Service ng-acc-configserver-db-service
    2019-08-28 13:22:53,860 INFO - [create] [ OK ] Deployment ng-acc-configserver-db-deployment
    2019-08-28 13:23:24,003 INFO - [verify] [ OK ] Pod of Deployment ng-acc-configserver-db-deployment is verified successfully
    2019-08-28 13:23:24,031 INFO - [create] [ OK ] Deployment ng-acc-configserver-deployment
    2019-08-28 13:33:25,813 ERROR - [verify] [ ERROR ] Pod ng-acc-configserver-deployment-5b6cbcf48c-4fjtw Initialized: False -> containers with incomplete status: [init-redis]
    2019-08-28 13:33:25,815 ERROR - [verify] [ ERROR ] Pod ng-acc-configserver-deployment-5b6cbcf48c-4fjtw Ready: False -> containers with unready status: [ng-acc-configserver-container]
    2019-08-28 13:33:25,815 ERROR - [verify] [ ERROR ] Pod ng-acc-configserver-deployment-5b6cbcf48c-4fjtw ContainersReady: False -> containers with unready status: [ng-acc-configserver-container]
    2019-08-28 13:33:25,815 ERROR - [verify] [ ERROR ] Pod ng-acc-configserver-deployment-5b6cbcf48c-4fjtw PodScheduled: True -> null
    2019-08-28 13:33:25,815 ERROR - [verify] [ ERROR ] Pod of Deployment ng-acc-configserver-deployment has failed to start in the given timeframe.

    can you help me to understand?
    Thanks 
    Luca Razzi