[openstack helm] [Ansible] Openstack-helm Ansible inventory.yaml missing some value
Hello today i'm testing installation with completely new enviroment i was following the guide from Openstack-helm document and install k8s cluster using ansible and deploy-env playbook but i encounter this error when setup ssh cluster TASK [deploy-env : Generate ssh key pair] ****************************************************************************************************************** skipping: [node-1] fatal: [primary]: FAILED! => {"msg": "Failed to change ownership of the temporary files Ansible (via chmod nor setfacl) needs to create despite connecting as a privileged user. Unprivileged become user would be unable to read the file."} and the log with -vvv output: debug2: mux_client_hello_exchange: master version 4 debug3: mux_client_forwards: request forwardings: 0 local, 0 remote debug3: mux_client_request_session: entering debug3: mux_client_request_alive: entering debug3: mux_client_request_alive: done pid = 114038 debug3: mux_client_request_session: session request sent debug1: mux_client_request_session: master session id: 2 chown: invalid user: ‘zuul’ debug3: mux_client_read_packet: read header failed: Broken pipe debug2: Received exit status from master 1 And by checking the playbook we can see the playbook is using "{{ client_ssh_user }}" on primary node and "{{ cluster_ssh_user }}" with cluster node which is not defined in inventory.yaml Causing it using default user zuul by adding client_ssh_user & cluster_ssh_user in ansible inventory it works normally Is this meant to be like this? or the document missing some variables in inventory.yaml
Hi, You are right, the doc is missing the recommendation to set the client_ssh_user and cluster_ssh_user variables. They must be defined. I'll fix the doc. All the defaults can be found here [1] [1] https://opendev.org/openstack/openstack-helm-infra/src/branch/master/roles/d... On Mon, Nov 11, 2024 at 7:32 PM <daniel890723@gmail.com> wrote:
Hello today i'm testing installation with completely new enviroment i was following the guide from Openstack-helm document and install k8s cluster using ansible and deploy-env playbook but i encounter this error when setup ssh cluster TASK [deploy-env : Generate ssh key pair] ****************************************************************************************************************** skipping: [node-1] fatal: [primary]: FAILED! => {"msg": "Failed to change ownership of the temporary files Ansible (via chmod nor setfacl) needs to create despite connecting as a privileged user. Unprivileged become user would be unable to read the file."} and the log with -vvv output: debug2: mux_client_hello_exchange: master version 4 debug3: mux_client_forwards: request forwardings: 0 local, 0 remote debug3: mux_client_request_session: entering debug3: mux_client_request_alive: entering debug3: mux_client_request_alive: done pid = 114038 debug3: mux_client_request_session: session request sent debug1: mux_client_request_session: master session id: 2 chown: invalid user: ‘zuul’ debug3: mux_client_read_packet: read header failed: Broken pipe debug2: Received exit status from master 1
And by checking the playbook we can see the playbook is using "{{ client_ssh_user }}" on primary node and "{{ cluster_ssh_user }}" with cluster node which is not defined in inventory.yaml Causing it using default user zuul by adding client_ssh_user & cluster_ssh_user in ansible inventory it works normally Is this meant to be like this? or the document missing some variables in inventory.yaml
-- Best regards, Kozhukalov Vladimir
Hi Vladimir and Daniel, I’m having same issue but inventory is having client_ssh_user and cluster_ssh_user set in vars. Do i need to set them again for primary node? One more question. After running the ansible-playbook second time the error appearing as helm plugin already exists. Ali On Wed, Nov 13, 2024 at 7:58 PM Vladimir Kozhukalov <kozhukalov@gmail.com> wrote:
Hi,
You are right, the doc is missing the recommendation to set the client_ssh_user and cluster_ssh_user variables. They must be defined. I'll fix the doc.
All the defaults can be found here [1]
[1] https://opendev.org/openstack/openstack-helm-infra/src/branch/master/roles/d...
On Mon, Nov 11, 2024 at 7:32 PM <daniel890723@gmail.com> wrote:
Hello today i'm testing installation with completely new enviroment i was following the guide from Openstack-helm document and install k8s cluster using ansible and deploy-env playbook but i encounter this error when setup ssh cluster TASK [deploy-env : Generate ssh key pair] ****************************************************************************************************************** skipping: [node-1] fatal: [primary]: FAILED! => {"msg": "Failed to change ownership of the temporary files Ansible (via chmod nor setfacl) needs to create despite connecting as a privileged user. Unprivileged become user would be unable to read the file."} and the log with -vvv output: debug2: mux_client_hello_exchange: master version 4 debug3: mux_client_forwards: request forwardings: 0 local, 0 remote debug3: mux_client_request_session: entering debug3: mux_client_request_alive: entering debug3: mux_client_request_alive: done pid = 114038 debug3: mux_client_request_session: session request sent debug1: mux_client_request_session: master session id: 2 chown: invalid user: ‘zuul’ debug3: mux_client_read_packet: read header failed: Broken pipe debug2: Received exit status from master 1
And by checking the playbook we can see the playbook is using "{{ client_ssh_user }}" on primary node and "{{ cluster_ssh_user }}" with cluster node which is not defined in inventory.yaml Causing it using default user zuul by adding client_ssh_user & cluster_ssh_user in ansible inventory it works normally Is this meant to be like this? or the document missing some variables in inventory.yaml
-- Best regards, Kozhukalov Vladimir
Hello First question: the client_ssh_user and Cluster_ssh_user will try to create an SSH key on the primary node using the user you provide and try to put the key on all k8s_node so you has to make sure you got access to the user on every node like you wants to make a key for user:ubuntu to access whole Cluster then you should enter Cluster_ssh_user:ubuntu if you trying to let Ansible to create a ssh-key using the wrong user that either it doesn't exist or not having permission the error will popup Second question : well that does happen to me everytime i install so i has to manually uninstall the plugin you can remove plugin by entering "helm plugin uninstall osh" on your terminal Also i think they can put ignore_errors on that task like this : - name: Install osh helm plugin become_user: "{{ kubectl.user }}" shell: | helm plugin install {{ osh_plugin_repo }} ignore_errors: true
participants (3)
-
alidot-subs Mail
-
daniel890723＠gmail.com
-
Vladimir Kozhukalov