Hello and Happy new year!

Here is a summary of what the team did from Dec 15 to Jan 11.

Sprint taskboard:
https://tree.taiga.io/project/morucci-software-factory/taskboard/2017-52


Regarding Software Factory:
We updated the 2.7 release to replace python2-paramiko from rdo-pike
by python-paramiko centos-extras.
We integrated python-zuul-sphinx to add documentation in zuul packages.
We integrated the haveged service to mitigate /dev/random locking
issues on rdo-cloud.
We replaced the local copy of zuul-jobs by a package of the
openstack-infra/zuul-jobs project.
We migrated sf-ci jobs to zuulv3, using the buildset-artifacts role to
enable re-use of sf-rpm-build job repository accross many integration
jobs. Only DCI and rdopkg jobs needs to be migrated now to cutover
zuulv2.
We started the migration of DCI jobs too
We have completed the new pages feature working based on zuulv3 and
roles has been contributed upstream to zuul-jobs
https://review.openstack.org/#/c/531406/
We have fixed some CI issues related to the upgrade from 2.7 to master
and related to artifact retrieval
We have worked on a REST API passthrough for manageSF, to simplify the
v2 API: API calls are forwarded to the target service (zuul, etc),
manageSF handles authorizations. There are open patches for zuul:
https://softwarefactory-project.io/r/#/c/10771/ &
https://softwarefactory-project.io/r/#/c/10714/ and for nodepool:
https://softwarefactory-project.io/r/#/c/10775/ &
https://softwarefactory-project.io/r/#/c/10776/
We started to work on grafyaml integration in SF, a tool used by
openstack infra to allow user to describe grafana dashboard in yaml.
This integration will allow users to manage their dashboard in the
config repo. We will integrate a graphite stack to allow user to use
dashboards based on queries used by openstack infra to provides
grafana dashboard

Regarding upstream contributions:
We continued the work on nodepool-drivers and created a proof of
concept  EC2 driver: https://review.openstack.org/528982
We investigated openapi swagger doc for zuul-web, but it's missing
auto sphinx doc generation: https://review.openstack.org/529013
We enabled sf-project.io as a third-party-ci for
openstack-infra/zuul-jobs to check centos/fedora environment for basic
role and tox jobs.
We contributed the linters job to zuul-jobs: https://review.openstack.org/530682
We investigated a solution to make Zuul more robust when a project
config is corrupted https://review.openstack.org/#/c/529060/
We have contributed the Git driver to Zuul
https://review.openstack.org/#/c/525614/
We have proposed a refactoring of the nodepool webapp to support more
queries, and added an admin endpoint:
https://review.openstack.org/#/c/531718/


Regarding RepoXplorer:
We have continued the effort to remove the mako templating and use
only the rest endpoints. As of now 98% have been done. One of the
prerequisite to release and bump the pkg in SF.

Regarding other things:
We created python-zuul-sphinx and python-sphinx-autodoc-typehints
package for fedora.
We proposed nodepool and zuul package for fedora.

Next public review will happen Thursday Jan 25 at 13:00 UTC, feel free
to join us on #softwarefactory for the event details.


The SF team

_______________________________________________
Softwarefactory-dev mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/softwarefactory-dev

Reply via email to