A problem encountered while installation

[linuix]

I followed https://github.com/cmusatyalab/elijah-openstack to execute the script.

When the sixth step $ cd ~/elijah-openstack/ansible
$ ansible-playbook -i ./hosts openstackpp.yaml

was executed, the following error occurred,but the dashboard can log in.

This may seems like a question about ansible playbook, but I don't know much about ansible playbook.

Could you please tell me the cause of this issue and how can i solve it?

...
...

TASK [openstack-controller : (OPENSTACK-EXT) add cloudlet panel to dashboard] ***********************
changed: [localhost]

RUNNING HANDLER [openstack-controller : restart apache] *********************************************
changed: [localhost]

RUNNING HANDLER [openstack-controller : restart nova] ***********************************************
changed: [localhost] => (item=nova-api)
changed: [localhost] => (item=nova-cert)
changed: [localhost] => (item=nova-consoleauth)
changed: [localhost] => (item=nova-conductor)
changed: [localhost] => (item=nova-scheduler)
changed: [localhost] => (item=nova-novncproxy)

PLAY [compute] **************************************************************************************

TASK [Gathering Facts] ******************************************************************************
ok: [localhost]

TASK [openstack-compute : (COMPUTE) install nova-compute] *******************************************
changed: [localhost] => (item=[u'nova-compute', u'sysfsutils'])

TASK [openstack-compute : (COMPUTE) install neutron-plugin] *****************************************
skipping: [localhost] => (item=[]) 

TASK [openstack-compute : (COMPUTE) replace nova.conf] **********************************************
skipping: [localhost]

TASK [openstack-compute : (COMPUTE) replace nova-compute.conf] **************************************
changed: [localhost]

TASK [openstack-compute : (COMPUTE) add nova to kvm/libvirtd groups] ********************************
ok: [localhost]

TASK [openstack-compute : (COMPUTE) remove SQLite] **************************************************
ok: [localhost]

TASK [openstack-compute : (COMPUTE) create OpenStack client script] *********************************
ok: [localhost]

TASK [openstack-compute : install networking components] ********************************************
skipping: [localhost] => (item=[]) 

TASK [openstack-compute : (COMPUTE) replace neutron.conf] *******************************************
skipping: [localhost]

TASK [openstack-compute : (COMPUTE) replace linuxbridge_agent.ini] **********************************
skipping: [localhost]

TASK [openstack-compute : (OPENSTACK-EXT) install fabric/openssh/git] *******************************
ok: [localhost] => (item=[u'git'])

TASK [openstack-compute : (OPENSTACK-EXT) checkout elijah-openstack code from github] ***************
ok: [localhost]

TASK [openstack-compute : (OPENSTACK-EXT) copy cloudlet_manager.py] *********************************
changed: [localhost]

TASK [openstack-compute : (OPENSTACK-EXT) copy cloudlet_driver.py] **********************************
changed: [localhost]

TASK [openstack-compute : (OPENSTACK-EXT) copy cloudlet_api.py] *************************************
changed: [localhost]

TASK [openstack-compute : (OPENSTACK-EXT) ensure nova-compute.conf is up to date] *******************
ok: [localhost]

TASK [openstack-compute : (OPENSTACK-EXT) restart core nova service if running everything on a single node] ***
changed: [localhost] => (item=nova-api)
changed: [localhost] => (item=nova-cert)
changed: [localhost] => (item=nova-consoleauth)
changed: [localhost] => (item=nova-conductor)
changed: [localhost] => (item=nova-scheduler)
changed: [localhost] => (item=nova-novncproxy)

TASK [openstack-compute : (OPENSTACK-EXT) ensure symlink to bios.bin exists] ************************
ok: [localhost]

TASK [openstack-compute : (OPENSTACK-EXT) ensure symlink to kvmvapic.bin exists] ********************
changed: [localhost]

TASK [openstack-compute : (OPENSTACK-EXT) ensure symlink to vgabios-cirrus.bin exists] **************
ok: [localhost]

TASK [openstack-compute : (OPENSTACK-EXT) change qemu conf] *****************************************
changed: [localhost]

TASK [openstack-compute : (OPENSTACK-EXT) exclude libvirt from apparmor] ****************************
changed: [localhost]

RUNNING HANDLER [openstack-compute : restart nova-compute] ******************************************
changed: [localhost] => (item=nova-compute)

RUNNING HANDLER [openstack-compute : restart libvirt-bin] *******************************************
changed: [localhost] => (item=libvirt-bin)

PLAY [controller] ***********************************************************************************

TASK [Gathering Facts] ******************************************************************************
ok: [localhost]


ERROR! Cannot reset connection:
Control socket connect(/root/.ansible/cp/346ca06965): No such file or directory

In addition, since the dashboard can log in, whether the following experiment can be carried out normally.

Is this error fatal?

Looking forward to and thank you for your reply.

Regards,
linuix

[teiszler]

@linuix,

This occurs in the following task of the cloudlet-gateway role:

• name: (GATEWAY) sever ssh connection and reconnect so that environment variables are loaded
  meta: reset_connection

I am trying to force Ansible to use a new ssh connection so that the environment variables that were just added to the system can be reloaded. Unfortunately due to the following Ansible bug, reset_connection isn't working properly:

Issue: reset_connection targets wrong ControlPath

If you launch the playbook again it should continue on past that step since you will be using a fresh SSH connection. You can also constrain it to that role by using '--tags gateway' when you launch the playbook. This will skip all of other roles (openstack controller/compute, etc) and only perform the steps for the gateway role. It is not a critical error, but it is necessary if you wish to use the cloudlet gateway functionality which allows you to orchestrate the launching of VMs/containers from the client.

Regards,
Tom

[linuix]

First of all, your reply is really fast.Thank you very much.

My understanding of your answer is, this is an Ansible bug, when perform the installation of the cloudlet gateway.

I see from https://github.com/cmusatyalab/elijah-openstack#cloudlet-gateway that the installation of the cloudlet gateway is optionally.

Q1:

I perform the script probably three or four times, still trapped in this step, i think your solution

is that behind this error steps (temporary skip this error) directly according to

https://github.com/cmusatyalab/elijah-openstack/tree/master/cloudlet-gateway to constrain it

to that role by using '- tags gateway', am i right?

Can i skip this error,and do the next step ?

Q2:

In addition,I also don't understand what you have said roles is to point to what, I will see some related information to learn further.

Q3:

Step 8. Create a pool of floating IP addresses
   
        These can be assigned to VMs to allow public access to them.

        $ nova floating-ip-bulk-create <cidr>

Is the floating IP addresses defined by myself, so the cidr can be like this:10.0.1.XX,same as the network segment of flat_iface?

Please forgive me if I say something unclearly.

regards
linuix