AArch64 is getting popular and not all changes working on x86-64 work
properly on aarch64 architecture. Let check them.
Job is running in separate pipeline due to low number of aarch64 nodes.
Authored-By: Xinliang Liu <xinliang.liu@linaro.org>
Co-Authored-By: Jeffrey Zhang <zhang.lei.fly@gmail.com>
Co-Authored-By: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Implements: blueprint aarch64-3rd-party-ci-kolla
Depends-On: https://review.opendev.org/#/c/698606
Change-Id: I7e7802dfa4f4d9c61d01dba4b77c01cfb24301c0
This change modifies the CentOS build CI jobs to use a CentOS 8 host
image.
NOTE: This does not affect the version of CentOS used as the base image
in containers built by kolla.
Partially-Implements: blueprint centos-rhel-8
Partially-Implements: blueprint centos-rhel-python3
Change-Id: Ie25b54a4e6116ed2bcb9c3b5c44eb6edbd2e6b49
Attempts affect pre failures. This means we can increase stability of
jobs by rejecting nodes that fail pre without failing runs at the same
time (unless we are really unlucky and hit b0rken nodes 5 times in a
row).
This change is adapted from I17b7f878c742fa8db66f738526855a02ab9f1905 in
kolla-ansible.
Change-Id: Ied5d2cb8ffa47b90833ee3cf241797601906f9b2
If stestr failed, then no node info was logged.
Logging node info is now handled in pre and post stages of Zuul.
Additionally Zuul v3 migration recommendations were applied
and some unused code removed improving general readability.
Change-Id: Iae9d37cb76997ba8389e21c7e2a3bf5905abe029
Signed-off-by: Radosław Piliszek <radoslaw.piliszek@gmail.com>
We want to migrate to the latest Ubuntu LTS release 18.04 aka Bionic.
This is just a patch to find possible issues before the real migration
starts. See [0] for the big picture.
Also test running tox jobs on Bionic.
[0] https://etherpad.openstack.org/p/devstack-bionic
Depends-On: https://review.openstack.org/610977
Change-Id: I748f26faf6f98e9693b5cd803577df1118121b41
We are seeing a number of POST_FAILURES in the periodic publishing jobs,
with jobs timing out pushing images to Docker hub. Try bumping the
post timeout to 2 hours.
Change-Id: I69b5610b712560e2d617933e639e77041953e21c
The default post-timeout is 1800, which is not enough to push images to
hub.docker.com.
Change-Id: If1e499c247f3eeb8452991cb18aea5b1fdad60a7
Closes-Bug: #1755416
Marcin Juszkiewicz