This commit extends the installation of Python interpreter
from system packages to also support RPM packages (assuming
the playbook is run in an environment with proper repositories).
Also the obsolete note about supporting only Debian distribution
is removed, as the `ensure-python` role already supports other
platforms (via pyenv and stow).
Change-Id: Id6ef8ec3537952348e0a7b233bd26b57613da327
CentOS 8 EOL'd at the end of 2021. OpenDev is beginning the process of
removing the image and starting with some job cleanups like the job
cleanups in this change. Users can use CentOS 8 Stream or some other
RHEL alike.
Change-Id: I9dab417b28a0503e1409e70953ed275f2c6f8c30
Older ansible-lint didn't pin the 'rich' dep and that dep updated and
broke ansible-lint. Newer ansible-lint fixed this. Update ansible-lint
to get around this problem.
Change-Id: Ic798398ab68239d4f13cb806a92038fd4946d25a
To match change I2870450ffd02f55509fcc1297d050b09deafbfb9 in Zuul.
This does not use the versioning feature due to the nature of this repo.
This also corrects a reference which is now an error.
Change-Id: Ia1d31df932b447f11bc588925de9974d4f6dfc7d
When processing subunit streams we attempt to discover where the
(s)testr commands are located. For some reason the output of our scripts
that do this very occasionally emit a newline before the command path.
When this happens we were using the blank line as the command path which
fails as that empty command results in our command arguments being
interpreted as the command path.
Attempt to address this by trimming the stdout instead of taking the
first line. This should remove the leading and trailing newlines and
give us only the command itself.
Change-Id: Id651e019cf3d0b7ab37fdf9df04be249ea7f7af6
OpenDev doesn't currently have an image built for tumblweed. Rather than
try and fix this image up and get it running again OpenDev has proposed
it simply be removed. In preparation for that and to avoid running jobs
for a system that doesn't currently have an image we remove the jobs
here.
It was hoped that tumbleweed would be a good forward looking platform
but it seems to need a lot more care than we are able to give it.
Additionally Zuul is probably the major consumer of the platform and it
doesn't see much consumption by other projects.
Change-Id: Iad351b6da1f08183bc4c49c049c303a9491af844
This reverts commit 10ee6ebf016fa00c9eb8cd15a09832982b08fe48.
Reason for revert: log url get broken
Change-Id: Ifb8db77c2534a92bbb19fde1bfbf31562ed04daf
Since the security update we are not able to modify
zuul.executor.log_root variable in the role testing.
Replace this with explicit log folder creation on the test node.
Change-Id: Ie6cff6f9e9c5594167ddda6cd345f9c3d9f2f470
FIPS needs to be enabled before test-setup is run, as enabling
FIPS requires the node to be rebooted, test-setup needs to run and
setup the environment after the reboot.
Change-Id: I6fecb9c6e917d1a36b2b82c1b02098eed4323ac7
Fedora supports FIPS. There is no version check because the feature is
available on all the supported Fedora releases.
Change-Id: I924fb565a4d70e7c93a215e9e0a5b2b80bced52a
Zuul's use of this role has exceeded the 500M cap. Let's go ahead
and remove it since it's arbitrary anyway, and Zuul is likely the
primary user of the role.
Change-Id: Ib0bc72f6c909e44bc04af7a8fbe40ef8415addcf
This updated branch fixes a few issues and should ensure display with
pcp-dstat (I9234b226242f145db9c235797649202aa530ec74)
Add an override so we pull it
Change-Id: I0d5e1b567c364a9e6c7aa0b95de17abffaef0434
Dstat is dead, but pcp supplies a compatible interface.
Installation can be unreliable, so let's ignore failures since
this isn't generally critical.
Change-Id: I9234b226242f145db9c235797649202aa530ec74
This patch adds a new multinode job definition that enables
FIPS mode prior to multinode configuration.
In order to enable FIPS mode, the OS boot procedure need to be
changed to enable the appropriate kernel flag. This modification
has effect only after system reboot.
The default behavior of this job is to always enable FIPS mode.
Change-Id: I6f1365837d9ed2ba82c391a20f9094c9ef0e6c4e
Signed-off-by: Douglas Viroel <dviroel@redhat.com>
With this patch, Adding repos to install openvswitch
in Centos9 stream.
For Centos9 stream nodes we need to enable 2 repos:-
* nfv repo - Ussuri+ openvswitch is shipped as part of NFV SIG
* rdo xena repo - rdo-openvswitch(which provides openvswitch wrapper)
is available in OpenStack repo
Closes-Bug: #1947665
Change-Id: If5a32b30637296a070f2e75ba8feb6aa45d8ecac
Zuul switched to a new base image, and it seems the new socat puts out
a warning (something like
... socat[489590] W ioctl(5, IOCTL_VM_SOCKETS_GET_LOCAL_CID, ...): Inappropriate ioctl for device
for reference).
Grep the output so we only get the line about what port it is
listening on.
Change-Id: I74fb86a9158b45e6601ee1fbc199ba80cd4991fe
The below error is coming on mol-centos7 jobs while
installing sibling for python2 packeges on centos7
Error: TypeError: initial_value must be unicode or None, not str
This patch handled the TypeError while installing sibling
for python packeges for centos7.
Closes-Bug: #1946641
Change-Id: Ie8058cca92d099e50af19b95b4c417c5a665da0d
build-container-image may ends up pushing large files. Network is not
always as reliable as we want. This commit ensures we retry several
times in case of failure.
Change-Id: Ieaf92a8c3531e24ab5e41783b540de1df806c02f
I think the secondary "rustup default" was intended to make sure the
Zuul user sets up to use the installed rust toolchain
(I32f9b285904a7036f9a80ada8a49fa9cf31b5163) but actually results in a
re-download of components and another local installation. This isn't
really the intention, and also doubles the time spent installing.
From the linked comment, it seems like we're not doing our global
install correctly; even putting it in /usr doesn't avoid the need for
RUST_HOME to be set. Take it's suggestion and install out-of-the-way
in /opt, use a small /usr/local/bin wrapper to call with correct env
vars set and then setup the installed global binary names to be called
via that.
Change-Id: I28ef747b809a17664305bfd9754022251390647b
With the original verbose output support in change
Iafeb88eaf9a596603ad4d2134a4574345d5189ab we looked for lines from
tox --showconfig output starting with an opening '[' but verbosity
also causes output from pip install activity to be included if a
tox.requires entry causes it to install packages before continuing
and these lines are prefixed by a process ID number wrapped in
brackets with the command string after that, which prematurely
triggered our search for the start of the INI content. Add a
stipulation that the first INI output line also end in ']' in order
to skip over those additional prepended lines.
Change-Id: If29e5a9abe3b92a145d87f5efc1b93350ea3908a
Update the alternate tox.ini files in test-playbooks/python to
trigger configuration we saw expose a bug in the siblings lib's
showconfig parser when a tox.requires value resulted in pip
activity. Update its minversion to the earliest tox which provides
the tox.requires configuration option, for clarity.
While we're there, switch to using tox_config_file instead of
overriding zuul_work_dir since that's now an option for this role
and was previously not exercised in the test job.
Change-Id: I521580da978903cd116c872447b1c028c470e63d
This reverts commit 8b1cc73ee30428517f9524624b9eb309e9b14c1f.
Apparently tox.requires adds lines like this to verbose showconfig
output:
using tox-3.24.4 from /.../tox/__init__.py (pid 2919)
Which is confusing the current parser in the siblings module. Roll
this back while we work on a fix and regression test.
Change-Id: If3b1d48b36a5d32fddfdabd9c0ec1b81dd6453f2