Description of problem: When embedded ansible role is activated, pod cfme starts deployment of ansible pod. when ansible is fully setup in pod, pod gets redeployed. I tried to verify ansible tower state after disabling liveness and readiness probes. It looks ok and gives correct responces on 80 and 443 ports. Both probes look good. however, increasing timeouts doesn't help though. Version-Release number of selected component (if applicable): 5.9.2.3 How reproducible: 100% Steps to Reproduce: 1. deploy 5.9.2.3 pod appliance 2. open CloudForms UI and enable Embedded Ansible role 3. wait until ansible pod gets deployed and ready Actual results: ansible pod is constantly redeployed and never gets ready Expected results: correct ansible pod deployment Additional info:
Looks like this is failing the in the new playbook seeding process: [----] I, [2018-04-26T17:07:30.047818 #7:a54879c] INFO -- : MIQ(ContainerEmbeddedAnsible#start) Waiting for Ansible container to respond [----] I, [2018-04-26T17:07:31.498826 #7:a54879c] INFO -- : MIQ(EmbeddedAnsibleWorker::Runner#setup_ansible) Finished starting embedded ansible service. [----] E, [2018-04-26T17:07:33.341609 #7:a54879c] ERROR -- : [ArgumentError]: can't find user for awx Method:[block in method_missing] [----] E, [2018-04-26T17:07:33.341763 #7:a54879c] ERROR -- : /opt/rh/rh-ruby23/root/usr/share/ruby/fileutils.rb:1116:in `getpwnam' /opt/rh/rh-ruby23/root/usr/share/ruby/fileutils.rb:1116:in `fu_get_uid' /opt/rh/rh-ruby23/root/usr/share/ruby/fileutils.rb:1087:in `chown_R' /var/www/miq/vmdb/app/models/embedded_ansible_worker/object_management.rb:94:in `chown_playbooks_tempdir' /var/www/miq/vmdb/app/models/embedded_ansible_worker/object_management.rb:65:in `ensure_plugin_playbooks_project_seeded' /var/www/miq/vmdb/app/models/embedded_ansible_worker/object_management.rb:11:in `ensure_initial_objects' /var/www/miq/vmdb/app/models/embedded_ansible_worker/runner.rb:52:in `update_embedded_ansible_provider' /var/www/miq/vmdb/app/models/embedded_ansible_worker/runner.rb:13:in `do_before_work_loop' /var/www/miq/vmdb/app/models/embedded_ansible_worker/runner.rb:5:in `prepare' /var/www/miq/vmdb/app/models/miq_worker/runner.rb:126:in `start' /var/www/miq/vmdb/app/models/miq_worker/runner.rb:22:in `start_worker' /var/www/miq/vmdb/app/models/embedded_ansible_worker.rb:20:in `block in start_monitor_thread' I think we can probably skip this if we're in a container. We need a separate solution for doing playbook seeding in the container-based installation.
https://github.com/ManageIQ/manageiq/pull/17353
New commit detected on ManageIQ/manageiq/master: https://github.com/ManageIQ/manageiq/commit/9112868a88e2003a83634990dd950ce8aaffa14f commit 9112868a88e2003a83634990dd950ce8aaffa14f Author: Joe Rafaniello <jrafanie> AuthorDate: Thu Apr 26 13:32:58 2018 -0400 Commit: Joe Rafaniello <jrafanie> CommitDate: Thu Apr 26 13:32:58 2018 -0400 Skip playbook seeding in containers Fixes https://bugzilla.redhat.com/show_bug.cgi?id=1571465 app/models/embedded_ansible_worker/object_management.rb | 2 +- spec/models/embedded_ansible_worker_spec.rb | 11 + 2 files changed, 12 insertions(+), 1 deletion(-)
I still see this bug in 5.9.2.4. Will it be included in release ?
Blocked by 1593342
ansible pod doesn't get redeployed in 5.10. Verified in 5.10.0.1