juniper / contrail-helm-deployer

Contrail Helm based deployment

Content was moved to https://github.com/tungstenfabric/tf-helm-deployer

contrail-helm-deployer's Issues

Deploy OpenContrail with external network

Hi, i try do deploy opencontrail and openstack with helm deployer. and it success. but anyone know ho to add external network to openstack? i need external network for connect my vm to external network. thanks before :)

neutron-server failed init

with the latest docker opencontrailnightly/contrail-openstack-neutron-init:latest, opencontrail-neutron-init container failed to init due to following error,

• lib_path=/usr/lib/python2.7/site-packages
  INFO: passed OPENSTACK_VERSION is
  ERROR: OPENSTACK_VERSION is required to init neutron plugin correctly
• echo 'INFO: passed OPENSTACK_VERSION is '
• [[ -z '' ]]
• echo 'ERROR: OPENSTACK_VERSION is required to init neutron plugin correctly'
• exit 1

I'm not able to deploy "make dev-deploy k8s multinode" successfully.

I'm trying to deploy openstack helm and contrail helm and got stuck here in make "dev-deploy k8s multinode" and getting the following error for deploy-kubelet action

ok: [verne] => {
"out.stdout_lines": [
"",
"PLAY [all] *********************************************************************",
"",
"TASK [Gathering Facts] *********************************************************",
"ok: [/mnt/rootfs]",
"",
"TASK [deploy-kubelet : include_tasks] ******************************************",
"included: /opt/playbooks/roles/deploy-kubelet/tasks/support-packages.yaml for /mnt/rootfs",
"",
"TASK [deploy-kubelet : centos | installing epel-release] ***********************",
"skipping: [/mnt/rootfs] => (item=[]) ",
"",
"TASK [deploy-kubelet : centos | installing SElinux support packages] ***********",
"skipping: [/mnt/rootfs] => (item=[]) ",
"",
"TASK [deploy-kubelet : fedora | installing SElinux support packages] ***********",
"skipping: [/mnt/rootfs] => (item=[]) ",
"",
"TASK [deploy-kubelet : installing ceph support packages] ***********************",
"",
"TASK [deploy-package : ubuntu | installing packages] ***************************",
"ok: [/mnt/rootfs] => (item=[u'ceph-common'])",
"",
"TASK [deploy-package : centos | installing packages] ***************************",
"skipping: [/mnt/rootfs] => (item=[]) ",
"",
"TASK [deploy-package : fedora | installing packages] ***************************",
"skipping: [/mnt/rootfs] => (item=[]) ",
"",
"TASK [deploy-kubelet : installing NFS support packages] ************************",
"",
"TASK [deploy-package : ubuntu | installing packages] ***************************",
"ok: [/mnt/rootfs] => (item=[u'nfs-common'])",
"",
"TASK [deploy-package : centos | installing packages] ***************************",
"skipping: [/mnt/rootfs] => (item=[]) ",
"",
"TASK [deploy-package : fedora | installing packages] ***************************",
"skipping: [/mnt/rootfs] => (item=[]) ",
"",
"TASK [deploy-kubelet : installing LinuxBridge support] *************************",
"",
"TASK [deploy-package : ubuntu | installing packages] ***************************",
"ok: [/mnt/rootfs] => (item=[u'bridge-utils'])",
"",
"TASK [deploy-package : centos | installing packages] ***************************",
"skipping: [/mnt/rootfs] => (item=[]) ",
"",
"TASK [deploy-package : fedora | installing packages] ***************************",
"skipping: [/mnt/rootfs] => (item=[]) ",
"",
"TASK [deploy-kubelet : include_tasks] ******************************************",
"included: /opt/playbooks/roles/deploy-kubelet/tasks/hostname.yaml for /mnt/rootfs",
"",
"TASK [deploy-kubelet : DNS | Ensure node fully qualified hostname is set] ******",
"fatal: [/mnt/rootfs]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'dict object' has no attribute 'ipv4'\n\nThe error appears to have been in '/opt/playbooks/roles/deploy-kubelet/tasks/hostname.yaml': line 13, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n\n- name: DNS | Ensure node fully qualified hostname is set\n ^ here\n"}",
"\tto retry, use: --limit @/opt/playbooks/kubeadm-aio-deploy-kubelet.retry",

Can anyone please help me if I need to make any changes in opt/openstack-helm-infra/tools/gate/devel/multinode-vars.yaml file.

Below is my multinode-vars.yaml file

kubernetes:
network:
default_device: enp129s0f1
ipv4: 172.19.2.2
cluster:
cni: calico
pod_subnet: 192.168.0.0/16
domain: cluster.local

Below is my multinode-inventory.yaml file

all:
children:
primary:
hosts:
jules:
ansible_port: 22
ansible_host: xxxx
ansible_user: root
ansible_ssh_private_key_file: /etc/openstack-helm/deploy-key.pem
ansible_ssh_extra_args: -o StrictHostKeyChecking=no
nodes:
hosts:
verne:
ansible_port: 22
ansible_host: xxxx
ansible_user: root
ansible_ssh_private_key_file: /etc/openstack-helm/deploy-key.pem
ansible_ssh_extra_args: -o StrictHostKeyChecking=no

The path ${OSH_PATH}/tools/kubeadm-aio/assets/opt/rbac/dev.yaml does not exsit.

the path ${OSH_PATH}/tools/kubeadm-aio/assets/opt/rbac/dev.yaml does not exsit.
error:
~/contrail-helm-deployer$ kubectl replace -f ${OSH_PATH}/tools/kubeadm-aio/assets/opt/rbac/dev.yaml
error: the path "/home/helm/openstack-helm/tools/kubeadm-aio/assets/opt/rbac/dev.yaml" does not exist

Error running "./tools/deployment/developer/nfs/091-heat-opencontrail.sh"

[ERROR] templates/: render error in "heat/templates/job-rabbit-init.yaml": template: heat/templates/job-rabbit-init.yaml:18:17: executing "heat/templates/job-rabbit-init.yaml" at <include "helm-toolki...>: error calling include: template: no template "helm-toolkit.manifests.job_rabbit_init" associated with template "gotpl"

Change RABBITMQ_PASSWORD in values.yaml

When replace rabbitmq password in values.yaml (contrail-thirdparty), password don't changed
values.yaml
RABBITMQ_PASSWORD: "epnjQuBBuhGZZxK"

rabbitmqctl authenticate_user guest epnjQuBBuhGZZxK
Authenticating user "guest"
Error: failed to authenticate user "guest"

rabbitmqctl authenticate_user guest guest
Authenticating user "guest"
Success

kubectl get pods -n openstack -o wide | grep contrail- returns nothing

the following command
kubectl get pods -n openstack -o wide | grep contrail-
should be
kubectl get pods -n contrail-o wide | grep contrail-

Segfault of systemd when vrouter pod is rescheduled

After "upgrading" from r5.1.0 nightly to r5.0.1 stable images rescheduling of the contrail-vrouter-agent-kernel pod will segfault systemd on the worder node.

Broadcast message from systemd-journald@k8s-node-01 (Tue 2019-01-22 09:54:28 UTC):

systemd[1]: Caught <QUIT>, dumped core as pid 85074.


Broadcast message from systemd-journald@k8s-node-01 (Tue 2019-01-22 09:54:28 UTC):

systemd[1]: Freezing execution.

Distro: Ubuntu 16.04.5 LTS
Systemd: 229-4ubuntu21.15
contrail-helm-deployer: branch R5.0

TF images used:

• build_driver_init: docker.io/tungstenfabric/contrail-vrouter-kernel-build-init:r5.0.1
• contrail_status: docker.io/tungstenfabric/contrail-status:r5.0.1
• node_init: docker.io/tungstenfabric/contrail-node-init:r5.0.1
• nodemgr: docker.io/tungstenfabric/contrail-nodemgr:r5.0.1
• vrouter_agent: docker.io/tungstenfabric/contrail-vrouter-agent:r5.0.1
• vrouter_dpdk: docker.io/tungstenfabric/contrail-vrouter-agent-dpdk:r5.0.1
• vrouter_init_dpdk: docker.io/tungstenfabric/contrail-vrouter-kernel-init-dpdk:r5.0.1
• vrouter_init_kernel: docker.io/tungstenfabric/contrail-vrouter-kernel-init:r5.0.1

How to deploy and manage Kubernetes on Openstack on Kubernetes Stack

Hello,

would it be possibl eto deploy further kubernetes (or openshift) instances (overcloud) in the openstack-cluster that is running on the undercloud kubernetes cluster. Would it be possible to control all stacks with one contrail instance ? How would Pods from overlay, underlay and VMs communicate with each other ? Over a Gateay (MX/vMX,etc.) ?
I suppose it would be, but are there any examples available ? Or ist "just" the normal K8s on Openstack Deployment ?

Thank you

Isn't used cluster_domain_suffix in ingress manifests

When changed cluster_domain_suffix in values.yaml, it don't affect host parameters in ingress manifests

03/19/2018 08:20:21 AM - error: URL [http://172.17.0.1:8082/virtual-networks] returned error ["Name['default-domain', 'admin'] not found"] 03/19/2018 08:20:21 AM - error: REST Server Error: Name['default-domain', 'admin'] not found

03/19/2018 08:19:17 AM - error: Project GET failed 8064b413-d4ab-433d-8605-45cb79b80ac6
03/19/2018 08:20:21 AM - error: URL [http://172.17.0.1:8082/virtual-networks] returned error ["Name['default-domain', 'admin'] not found"]
03/19/2018 08:20:21 AM - error: REST Server Error: Name['default-domain', 'admin'] not found
at APIServer.retryMakeCall (/usr/src/contrail/contrail-web-core/src/serverroot/common/rest.api.js:203:13)
at Request. (/usr/src/contrail/contrail-web-core/src/serverroot/common/rest.api.js:334:18)
at Request.emit (events.js:98:17)
at Request.mixin._fireSuccess (/usr/lib64/node_modules/restler/lib/restler.js:250:10)
at /usr/lib64/node_modules/restler/lib/restler.js:181:20
at IncomingMessage.parsers.auto (/usr/lib64/node_modules/restler/lib/restler.js:414:7)
at Request.mixin._encode (/usr/lib64/node_modules/restler/lib/restler.js:218:29)
at /usr/lib64/node_modules/restler/lib/restler.js:177:16
at Request.mixin._decode (/usr/lib64/node_modules/restler/lib/restler.js:193:7)
at IncomingMessage. (/usr/lib64/node_modules/restler/lib/restler.js:170:14)