Exception not raised when install script failed in kubernetes_mgmt

[LiangLu]

While running install script in kubernetes_mgmt, it will find 'Install Failed!\n' full match in stdout

https://opendev.org/openstack/tacker/src/branch/master/samples/mgmt_driver/kubernetes_mgmt.py#L108
        elif type == 'install':
            err = result.get_stderr()
            if 'Install Failed!\n' in result.get_stdout():
                LOG.error(err)
                raise exceptions.MgmtDriverRemoteCommandError(err_info=err)
        return result.get_stdout()

However, we fixed error message in script at last minutes of merge that changed error message:

https://opendev.org/openstack/tacker/src/branch/master/samples/mgmt_driver/install_k8s_cluster.sh#L768

if [[ "$INSTALL_MODE" =~ "master" ]]; then
    result=$(kubectl get nodes -o wide | grep $CURRENT_HOST_IP)
    if [[ -z "$result" ]];then
        echo "Install Failed! The node does not in Kubernetes cluster."
        exit 255
    else

Thus a fix that fits both sides is needed

[Yasufumi Ogawa (yasufum)]

I'm not sure importance of the bug actually because it's just a sample. However, It should be fixed anyway soon or later. I assigned you to the owner of the issue because you proposed the original change and reported it then. Please fix it when convenient. Thanks!

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/tacker 5.0.0.0rc3 release candidate.

[Yasufumi Ogawa (yasufum)]

Fixed: https://review.opendev.org/c/openstack/tacker/+/784519

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/tacker 6.0.0.0rc1 release candidate.