On Mar 17, 2016 5:52 AM, "ABDALA Olga" <[email protected]> wrote:
>
> Hi Jason,
>
>
>
> Thank you for your inputs. Please find below my comments.
>
>
>
> De : Jason DeTiberus [mailto:[email protected]]
> Envoyé : mercredi 16 mars 2016 22:25
> À : ABDALA Olga
> Cc : [email protected]
> Objet : Re: [OpenShift 3 deployment] Error response from daemon: Error
setting devmapper transaction ID: Error running SetTransactionID
dm_task_run failed
>
>
>
>
>
>
>
> On Tue, Mar 15, 2016 at 10:00 AM, ABDALA Olga <[email protected]>
wrote:
>
> Hello,
>
>
>
> It’s me again with my openShift deployment issues. Last time, I was
telling that I managed to deploy a non-distributed OpenShift with just 1
Master and 1 Node.
>
> From the basic configuration, I wanted to add the number of Nodes in
order to have 1 Master and 2 Nodes.
>
>
>
> So what I did is, I followed the steps, as indicated here . What changed
from my previous configuration, is :
>
> - I enabled SELinux
>
> - I installed ansible
>
> - I installed the following packages
>
> yum install wget git net-tools bind-utils iptables-services bridge-utils
bash-completion
>
> - I did what is asked for the EPEL repository
>
> - I cloned the openshift-ansible repository
>
> - I generated an SSH key on the master then copied the key on the
other Node hosts
>
> - I then configured my /etc/ansible/hosts file as follow:
>
>
>
> # Create an OSEv3 group that contains the masters and nodes groups
>
> [OSEv3:children]
>
> masters
>
> nodes
>
> #etcd
>
> #lb
>
>
>
> # Set variables common for all OSEv3 hosts
>
> [OSEv3:vars]
>
>
>
> #SSH user, this user should allow ssh based auth without requiring a
password
>
> ansible_ssh_user=root
>
>
>
> # If ansible_ssh_user is not root, ansible_sudo must be set to true
>
> #ansible_sudo=false
>
>
>
> deployment_type=origin
>
>
>
> # uncomment the following to enable htpasswd authentication; defaults to
DenyAllPasswordIdentityProvider
>
> openshift_master_identity_providers=[{'name': 'htpasswd_auth', 'login':
'true', 'challenge': 'true', 'kind': 'HTPasswdPasswordIdentityProvider',
'filename': '/etc/origin/master/htpasswd'}]
>
>
>
> # host group for masters
>
> [masters]
>
> sv5256.selfdeploy.loc openshift_hostname=sv5256.selfdeploy.loc
containerized=true
>
>
>
> Any reason you are choosing a containerized the master and not the nodes?
>
> ð I thought that by setting “containerized to true”, the containerized
services were run on both the Master and the Nodes (please correct me if
I’m wrong, since I am really new to OpenShift).
If you set 'containerized=true' under [OSEv3: vars] that would have been
the case.
With allowing for a mixture of containerized and non-containerized could be
to allow someone to migrate from a non-containerized install to a
containerized install (this would require a bit of extra tooling or manual
process), but the main reason is that we didn't want to impose an arbitrary
limitation.
>
>
>>
>>
>>
>> # host group for nodes, includes region info
>>
>> [nodes]
>>
>> sv5256.selfdeploy.loc openshift_node_labels="{'region': 'infra', 'zone':
'default'}"
>>
>> SV5257.selfdeploy.loc openshift_node_labels="{'region': 'primary',
'zone': 'east'}"
>>
>> SV5258.selfdeploy.loc openshift_node_labels="{'region': 'primary',
'zone': 'west'}"
>
>
>
> Your master should be listed as a node as well, there are some features
that require that the master be part of the sdn and currently that is only
possible with the master also being a node.
>
> ð The master is listed as a node actually. The Master hostname is:
sv5256.selfdeploy.loc . Or maybe is there another place it should be
mentioned as node ?
Ah, I missed that on the first read through. That is sufficient for
defining the master.
>
>
>>
>>
>>
>> - I then run the advanced installation as follow:
>>
>> # ansible-playbook ~/openshift-ansible/playbooks/byo/config.yml
>>
>> The `oc get nodes`, at first showed me that all the hosts were ready
(cfr the attached screenshot),
>>
>>
>>
>> - I also configured a master file
The installation should have left a master config file in
/etc/origin/master/master-config.yaml.
>>
>> - I could not set up the Node configuration file because I was
getting a message saying that ‘ca.crt’ file in
`/var/lib/origin/openshift.local.config/master/` needed to be correct. The
thing is I cannot find the `/openshift.local.config/master` under /var/lib.
It is located somewhere else.
>
> This is because your master is not currently a node, once it is added to
the node inventory, you should see a node directory containing
configuration under /etc/origin/.
>
> ð So is that still is the cause of the error message ?
I missed the part where the master was a node, but the node configuration
would should be present at /etc/origin/node/node-config.yaml. It sounds
like you may have the origin binary installed on the local system though.
>
>
>>
>>
>>
>> I tried to re-run the `oc get nodes` command and any other, but , I am
being displayed this error:
>>
>>
>>
>>
>>
>> Error response from daemon: Error setting devmapper transaction ID:
Error running SetTransactionID dm_task_run failed.
>
>
>
> Sounds like docker is hung on the host. You can try to restart the docker
daemon, and then attempt to run oc again (On a containerized installation
oc is actually a shell script that runs the oc client within a container
rather than natively).
Have you tried to restart Docker? You may have to also start the
origin-master and origin-node services as well if they don't come back on
their own after restarting Docker.
>
>
>>
>>
>>
>> Can please anybody help me with this matter?
>>
>>
>>
>>
>>
>> Thanks
>>
>>
>>
>>
>>
>> ------------
>>
>> Olga
>>
>>
>>
>>
>>
>> _______________________________________________
>> dev mailing list
>> [email protected]
>> http://lists.openshift.redhat.com/openshiftmm/listinfo/dev
>
>
>
>
>
> --
>
> Jason DeTiberus
_______________________________________________
dev mailing list
[email protected]
http://lists.openshift.redhat.com/openshiftmm/listinfo/dev
