AIO deploy fails for missed no_proxy on step setup-openstack.yml

[Peter Garlic]

25.1.0 - AIO deployment on a single Rocky 8.6 VM with default values inside one infrastructure with proxy. fails on last playbook "openstack-ansible setup-openstack.yml" for a missed no_proxy entry.

How to reproduce:

- check proxy configuration
/etc/profile.d/proxy.sh
export ftp_proxy=http://<my-proxy-ip:port>
export http_proxy=http://<my-proxy-ip:port>
export https_proxy=http://<my-proxy-ip:port>
export no_proxy="127.0.0.1,localhost,<my-server-ip>"

export FTP_PROXY=http://192.168.137.249:8888
export HTTP_PROXY=http://192.168.137.249:8888
export HTTPS_PROXY=http://192.168.137.249:8888
export NO_PROXY="127.0.0.1,localhost,<my-server-ip>"

- follow the quickstart with defauklt values

# clone repo
git clone https://opendev.org/openstack/openstack-ansible /opt/openstack-ansible
cd /opt/openstack-ansible
git checkout 25.1.0

# run scripts
scripts/bootstrap-ansible.sh
scripts/bootstrap-aio.sh

# Run playbooks
cd /opt/openstack-ansible/playbooks
openstack-ansible setup-hosts.yml
openstack-ansible setup-infrastructure.yml
openstack-ansible setup-openstack.yml

The operation fails on:

>TASK [python_venv_build : Install python packages into the venv]
>FAILED - RETRYING: [aio1]: Install python packages into the venv (5 retries left).
> connection broken by 'ReadTimeoutError(\"HTTPConnectionPool(host='ip.of.my.proxy', port=<myport>): Read timed out

- Workaround:
before launching "openstack-ansible setup-openstack.yml" add to no-proxy environment the address "172.29.236.101"

# fix no_proxy
export no_proxy="127.0.0.1,localhost,ip.of.my.proxy,172.29.236.101"
export NO_PROXY="127.0.0.1,localhost,<ip.of.my.proxy>,172.29.236.101"

After exporting the no_proxy values for 172.29.236.101 (re)-launch last playbook

openstack-ansible setup-openstack.yml

and the installation will succesfully complete.

On attachment the last section of playbook log

Kind regards

[OpenStack Infra (hudson-openstack)]

Related fix proposed to branch: master
Review: https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/858385

[Dmitriy Rabotyagov (noonedeadpunk)]

Original bugfix was introduced with https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/852451 However, it brought in another bug, which should be covered with this related fix

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/858385
Committed: https://opendev.org/openstack/openstack-ansible-os_keystone/commit/7868766202b80205574efa66deb217ab2491d5c2
Submitter: "Zuul (22348)"
Branch: master

commit 7868766202b80205574efa66deb217ab2491d5c2
Author: Dmitriy Rabotyagov <email address hidden>
Date: Mon Sep 19 16:08:46 2022 +0200

    Bootstrap when running against last backend

    When deploying keystone for the first time, aliveness check inside
    service_bootstrap can not succeed for multi-node setup, as playbook
    will disable current backend. So we need to bootstrap host only
    when running against last host in play. We also should make sure, that
    following tasks will not fail when running against first ones.

    Closes-Bug: #1990008
    Related-Bug: #1989326
    Change-Id: Ifa9a79c34265b225a5e24c30cae47d3f0fa0739f

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/yoga
Review: https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/859232

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/859232
Committed: https://opendev.org/openstack/openstack-ansible-os_keystone/commit/36d06d4cc3c74fb749ac655f74e35092d6acea07
Submitter: "Zuul (22348)"
Branch: stable/yoga

commit 36d06d4cc3c74fb749ac655f74e35092d6acea07
Author: Dmitriy Rabotyagov <email address hidden>
Date: Mon Sep 19 16:08:46 2022 +0200

    Bootstrap when running against last backend

    When deploying keystone for the first time, aliveness check inside
    service_bootstrap can not succeed for multi-node setup, as playbook
    will disable current backend. So we need to bootstrap host only
    when running against last host in play. We also should make sure, that
    following tasks will not fail when running against first ones.

    With that we also now able to check the service status during
    bootstrap against the internal VIP. This is important in environments
    with limited connectivity.

    Closes-Bug: #1990008
    Closes-Bug: #1989326
    Squashes Change-ID: I1a4aec40618237aa23b4f40b335c141071a56f08

    Change-Id: Ifa9a79c34265b225a5e24c30cae47d3f0fa0739f
    (cherry picked from commit 7868766202b80205574efa66deb217ab2491d5c2)

[OpenStack Infra (hudson-openstack)]

This issue was fixed in the openstack/openstack-ansible-os_keystone yoga-eom release.