Bug 1514490

Summary: Configuring the host from cockpit is stuck, if ssh public key authentication is not configured
Product: [oVirt] cockpit-ovirt Reporter: SATHEESARAN <sasundar>
Component: GdeployAssignee: Parth Dhanjal <dparth>
Status: CLOSED DUPLICATE QA Contact: SATHEESARAN <sasundar>
Severity: low Docs Contact:
Priority: unspecified    
Version: 0.10.8-2.2CC: bugs, dparth, rhs-bugs, sabose, sasundar
Target Milestone: ovirt-4.3.0Flags: rule-engine: ovirt-4.3?
sasundar: planning_ack?
sasundar: devel_ack?
sasundar: testing_ack?
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: 1514466 Environment:
Last Closed: 2018-11-21 15:11:09 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: Gluster RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1514466    
Attachments:
Description Flags
Ansible throwing error when ssh public key authentication is not configured none

Description SATHEESARAN 2017-11-17 14:37:27 UTC
+++ This bug was initially created as a clone of Bug #1514466 +++

Description of problem:
-----------------------
If ssh public key authentication is not configured, and deployment is initiated, then there is no proper indication of error, and the flow is stuck, unless the user cancels and restarts again.

Version-Release number of selected component (if applicable):
--------------------------------------------------------------
gdeploy-2.0.2-19.el7rhgs
ansible-2.4.0.1-1
cockpit-ovirt-dashboard-0.10.9-1.el7ev.noarch

How reproducible:
-----------------
Always

Steps to Reproduce:
-------------------
1. Skip the step of copying the ssh public key to the other hosts in the cluster
2. Start the deployment

Actual results:
---------------
The cockpit window is stuck in the page.
There is no provision to 'redeploy' unless to cancel the installation

Expected results:
-----------------
1. There should be error thrown at cockpit UI, when unable to connect to the host
2. There should be flexibility to redeploy with the available conf file

--- Additional comment from SATHEESARAN on 2017-11-17 09:21 EST ---



--- Additional comment from SATHEESARAN on 2017-11-17 09:25:15 EST ---

Error message as seen from gdeploy logs:

<snip>


[2017-11-17 18:25:27] INFO gdeploy[186]: **** gdeploy run started with options ['--version'] ****
[2017-11-17 18:25:27] INFO gdeploy[187]: For complete CLI log, set log_path in /etc/ansible/ansible.cfg
[2017-11-17 18:39:28] INFO gdeploy[185]: 


[2017-11-17 18:39:28] INFO gdeploy[186]: **** gdeploy run started with options ['-c', '/tmp/gdeployConfig.conf'] ****
[2017-11-17 18:39:28] INFO gdeploy[187]: For complete CLI log, set log_path in /etc/ansible/ansible.cfg
[2017-11-17 18:39:28] INFO gdeploy[197]: gdeploy hosts: ['10.70.36.73', '10.70.36.74', '10.70.36.75']
[2017-11-17 18:39:28] INFO gdeploy[170]: Playbook directory for the run: /usr/share/gdeploy/playbooks
[2017-11-17 18:39:28] INFO helpers.py[104]: Copied files from /usr/share/gdeploy/playbooks to /tmp/tmprmF1YV
[2017-11-17 18:39:28] INFO core_function_caller.py[40]: No tuning profiles specified, ignoring
[2017-11-17 18:39:28] INFO call_features.py[146]: Parsing json file: /usr/lib/python2.7/site-packages/gdeployfeatures/script/script.json
[2017-11-17 18:39:28] INFO helpers.py[376]: Invoking playbook  /tmp/tmprmF1YV/run-script.yml
[2017-11-17 18:46:52] INFO gdeploy[185]: 


</snip>

Comment 1 Sahina Bose 2017-11-24 12:08:44 UTC
cockpit-gdeploy plugin will only display errors returned from gdeploy. So it looks like a bug is needed in gdeploy to log this error?

Comment 2 SATHEESARAN 2017-12-11 07:11:29 UTC
(In reply to Sahina Bose from comment #1)
> cockpit-gdeploy plugin will only display errors returned from gdeploy. So it
> looks like a bug is needed in gdeploy to log this error?

When I tried the same case with RHVH as platform, I am not seeing this problem.
I will try once again with RHEL platform, if the issue is again seen only in RHEL platform, I will consult with gdeploy dev for more information.

Will update the bug as soon as the testing is complete

Comment 3 Sahina Bose 2017-12-15 12:33:14 UTC
Setting needinfo for the update on bug

Comment 4 SATHEESARAN 2018-01-23 10:16:46 UTC
(In reply to Sahina Bose from comment #3)
> Setting needinfo for the update on bug

Sahina,

Yes, I am seeing this problem with RHVH platform too. The comment2 is incorrect.
Even when I close the 'Hosted Engine with Gluster' wizard, the gdeploy process still runs on the machine.

[root@ ~]# ps aux | grep gdeploy
root     37091  0.4  0.0 219072 15292 ?        S    15:21   0:00 /usr/bin/python /usr/bin/gdeploy -c /tmp/gdeployConfig.conf
root     37343  0.0  0.0 112664   972 pts/0    S+   15:22   0:00 grep --color=auto gdeploy


I feel this should be a gdeploy return as Sahina has suggested.

Comment 5 Parth Dhanjal 2018-11-15 06:03:34 UTC
Created attachment 1505935 [details]
Ansible throwing error when ssh public key authentication is not configured

Hey!

This bug will be resolved as we integrate ansible in place of gdeploy, which is planned for oVirt 4.3
The UI throws an error under the review step itself(refer screenshot for the same)

[vagrant@gd2-node-1 ~]$ ps aux | grep ansible
vagrant   5992  0.0  0.0 112704   972 pts/0    S+   05:58   0:00 grep --color=auto ansible
[vagrant@gd2-node-1 ~]$

Comment 6 Sahina Bose 2018-11-21 15:11:09 UTC

*** This bug has been marked as a duplicate of bug 1651516 ***