devel March 2023

devel@lists.centos.org

31 participants
19 discussions

[REPORT] Quarterly Report - CentOS Automotive SIG
by Jeffrey Osier-Mixon 17 Aug '23

17 Aug '23

January 2023 Quarterly report submitted by: Jefro Osier-Mixon, Red Hat - chair _____________________________________ Membership update This SIG does not have a formal membership process. The mailing list currently has 106 subscribers representing at least 32 organizations, though not all subscribers use corporate emails and some are participating as individuals. _____________________________________ Releases in the most recent quarter (or most recent release, if none in that quarter) The Automotive SIG produces three types of artifacts: - AutoSD, a streaming distribution of CentOS designed for in-vehicle automotive use cases. - An Automotive SIG RPM repository that allows the community to expand the content of AutoSD or experiment with some of its parts. - Sample images, built using OSBuild, which provide examples of how to assemble production images based on AutoSD, customized for some hardware, including container images, based on CoreOS/ostree technologies. AutoSD, or Automotive Stream Distribution, is a streaming distribution for automotive in-vehicle software development based on CentOS Stream. It is transparently the upstream project for Red Hat's eventual in-vehicle OS product. AutoSD has been downloaded and used by many organizations who have commented or asked for help, so we know it is getting some traction though of course we don't have exact metrics on usage. In Q4 2022, we released a version of AutoSD compatible with the emerging SOAFEE specification. SOAFEE (https://soafee.io) is an initiative driven by Arm for the purpose of developing a reference specification for Arm-based software-defined vehicles (SDVs) consisting of an operating system, container orchestration strategy, and application layer. In addition, we introduced a very lightweight container runtime based on podman and systemd that does not include Kubernetes, as the latter represented a significant performance hit on all relevant hardware. We presented at the Arm Developer Summit in November on these topics. In the coming quarter, we plan to tighten up documentation, begin a process for accepting and supporting hardware enablement contributions, and continue to develop AutoSD toward the SOAFEE specification while also contributing to that specification within the SOAFEE community. _____________________________________ Health report and general activity narrative The SIG typically has 1 public meeting per month with 25-40 attendees, with visible participation from 7-10 separate organizations. We have largely given up on a separate office hours meeting due to attendance. This SIG is intended to be a community effort with contributions and shared benefits from all participants. All formal meetings are recorded and posted on this page: https://wiki.centos.org/SpecialInterestGroup/Automotive/Meetings Several Red Hat employees made the initial contributions to the project as well as the infrastructure required to build and test it. We occupy a gitlab repository in the CentOS namespace building software regularly using CI, with build instructions provided on the documentation page at https://sigs.centos.org/automotive/ . Sample images are present and downloadable along with customization and build instructions. _____________________________________ Issues for the board to address, if any None, keep up the excellent work :) Jeffrey "Jefro" Osier-Mixon | jefro(a)redhat.com Red Hat Office of the CTO | Sr. Principal Community Architect, Automotive

1 1

Adding bootc images - part of Cloud SIG or separate?
by Colin Walters 29 Jun '23

29 Jun '23

Hello, I'm a developer on Fedora/RHEL and OpenShift. Lately we've been landing a lot of "bootable container" changes in OpenShift core, and there's a lot more to come. However, as we've been doing about this...I've been saying to people that I wish I had a time machine to go back and do bootable containers from the start. There's a lot of things we're doing today that I think we should stop doing, e.g.: - Switching to kernel-rt by fiddling with each node; we should be simply pulling a pre-built bootable container image with that kernel (more on this below) - Getting away from injecting so much persistent state by default (both via Ignition and outside of it) And crucially, I think we should be developing tools and techniques that apply *outside* of Kubernetes/OpenShift and also work well with it. To be direct, I'd like to eventually productize some of what's happening here in RHEL, not in OpenShift. As part of this (potential) re-architecture of how we think of systems management, I created the https://github.com/containers/bootc project. To be direct: If successful, I think bootc will be the successor to (rpm-)ostree. It's also intended to much more closely align with the github.com/containers organization. A simple way to think of this is: One can (build and) run *application* containers with podman; and these containers can also be run in e.g. Kubernetes/OpenShift. One can build *bootable* containers using any tooling (including podman build), but *running* them is via bootc on the end machine. bootc understands kernels etc. But there's a lot to figure out here - and I want to have a space to figure out this stuff and experiment with it outside of a direct-to-product path. I think a CentOS SIG makes sense for this. So what I'd like to do is either: - Add a new effort to the Cloud SIG, which currently (IMO a bit confusingly) hosts OpenStack/RDO and OpenShift/OKD things which would be a 3rd thing. The bootc work would then be the "base OS" split for OKD/SCOS. But of course, nothing stops one from building bootable host images that are instead designed to be RDO/OpenStack hosts. - Or, create a new SIG Personally, I lean towards the latter because honestly I find the naming "Cloud" to be misleading - bootc is also intended to be useful for standalone, non-cloud-infrastructure settings (such as desktops and IoT). Specifically, I'd like to transfer the existing code that lives in https://github.com/cgwalters/bootc-demo-base-images (specifically https://github.com/cgwalters/bootc-demo-base-images/blob/main/c9s.yaml ) into something CentOS-affiliated and explicitly maintained by a team. (Though I'm not super excited to move it to pagure like at least some other SIG content, but let's not get distracted by git hosting too much here). Another way to say it is that I'd love to ship quay.io/centos/centos-boot:stream9 (notice the -boot). Or failing that, it'd be quay.io/centos-boot/centos-boot:stream9 or so. There's a *lot* to discuss in terms of what actually goes in these base images, and also ensuring it's equally ergonomic for users to build their own base images. So really it's very likely there wouldn't be just *one* base image. In fact, I recently introduced a -rt variant with the RT kernel: https://github.com/cgwalters/bootc-demo-base-images/commit/68afb072a5a1396c… - and this was specifically motivated by issues we hit in OCP. But again, I want to have a space where we try to do more of a "clean(er) slate" approach for a while, with notes "not for production use" - for a while. Everything done here though *is* made with that as an explicit goal though (e.g. it's a toplevel design goal too that existing ostree-based systems can be seamlessly switched to be container-based without reprovisioning). At the same time, bootc already introduces some quite new things that need design iteration; for example: https://github.com/containers/bootc#using-bootc-install - we ship tooling such that a container can install itself (without going through a raw disk image as is used by both OCP and Edge deployments today). And at the same time, I'd like to aim to get the Anaconda changes to install these bootable containers in https://github.com/rhinstaller/anaconda/pull/4561 OK this is already too long, so I'm just going to click send =) Thoughts?

7 10

Discussion: Open technical questions concerning SIG content for RHEL releases
by Peter Georg 09 Jun '23

09 Jun '23

Dear all, There has been an inquiry to the CentOS Board to clarify how long SIGs can expect to be able to build against RHEL build targets [1]. This is necessary to allow SIGs themselves to communicate expected lifecycle of their content to users. The outcome of this, in summary, is that SIGs should be able to build against RHEL until it reaches EOL (end of Maintenance Phase) + 30 days (no access to ELS content). This policy follows the one used by EPEL. Note: It is (obviously) still up to every single SIG to decide if and for how long they want to support a particular RHEL (or CentOS Stream) release. This inquiry also included respectively raised some technical questions. These technical questions are not to be decided by the CentOS Board. In [2] it has been recommended to discuss these on centos-devel. So let's do that. I want to start with two open questions which I do know of. In case there are other open question concerning this matter please raise them here. First open question is where to push content produced by SIGs for RHEL8 to. To better understand the issue, let me summarize the current situation, for which we first need to define the used format for tags in CBS: <sig><el>-<project>-<version>-{candidate,testing,release} where <el> can currently be 7,8,8s,9,9s Packages tagged for -testing are pushed to: https://buildlogs.centos.org/centos/<el>/<sig>/<arch>/<project>-<version>/ Packages tagged for -released are push to: If el = 8 or 8s: debuginfo rpm packages: https://debuginfo.centos.org/centos/<el>/<sig>/<arch>/<project>-<version>/ src.rpm packages: https://vault.centos.org/centos/<el>/<sig>/Source/<project>-<version>/ Everything else: http://mirror.centos.org/centos/<el>/<sig>/<arch>/<project>-<version>/ If el = 9 or 9s: http://mirror.stream.centos.org/SIGs/<el>/<sig>/<arch>/<project>-<version>/ For src.rpm <arch> is source. Note: Although it is not relevant for the discussion, in fact on the mirrors the suffix "s" of <el> is replaced by "-stream". This currently is all fine. However it is planned that the infrastructure currently used for packages built for 8 or 8s will disappear once CentOS Stream 8 and CentOS Linux 7 are EOL (June 30th, 2024). So in case any SIG decides to produce content for RHEL 8 after that date packages can not be distributed properly anymore. Only using the -testing tag and thus https://buildlogs.centos.org is not really an option as it would probably produce too much load on these. My proposal: After CentOS Stream 8 went EOL (May 31st, 2024), which is the same date as the end of the Full support phase of RHEL 8, at least the -release tag of all RHEL 8 build targets is locked and SIGs have to opt-in in case they want to build content for RHEL 8 during its Maintenance Support phase (End: May 31st 2029) for a particular build target. For these, tagging for -release will then push content to http://mirror.stream.centos.org/SIGs/8/<sig>/<arch>/<project>-<version>/, i.e. the same location as used for 9 and 9s. This gives SIGs one month for all required changes before the currently infrastructure disappears. Of course this means additional work for SIGs and Infra. However at some point the location where to push content to has to be changed before the currently used infrastructure disappears. Choosing an earlier date would result in work for SIGs who do not want to continue providing any packages once RHEL 8 enters Maintenance Support phase anyway. The second open question concerns the centos-release-* packages provided by SIGs to allow users to easily consume SIGs' content. For 8s and 9s the CBS tags extras<el>-extras-common-{candidate,testing,release} are used to build these packages. This repository is added in CentOS Stream 8 and 9. For packages build for RHEL 8 and 9 there is currently no common way to provide any means of easing the process to consume SIGs' content. My proposal to fix this is by adding extras<el>-extras-common-{candidate,testing,release} for <el> = 8 and 9, i.e., using the same system as currently used for 8s and 9s. To further ease the process I propose to introduce a package named centos-release-extras which contains the repository config pointing to the content of the tags extras<el>-extras-common-{testing,release} (only -release enabled by default) and the CentOS-SIG-Extras GPG key. The centos-release-extras packages itself would be built in the extras<el>-extras-common-el<el> build target. Users of RHEL would then only need to install this single package to allow them to easily install any other centos-release-* packages. Obviously someone needs to maintain the centos-release-extras package. I volunteer to maintain this package. [1]: https://git.centos.org/centos/board/issue/82 [2]: https://pagure.io/centos-infra/issue/1002

4 11

Few questions on cloud images
by Pino Toscano 06 Jun '23

06 Jun '23

Hello, I have few questions related to the cloud images provides for c8s and c9s, i.e.: - https://cloud.centos.org/centos/8-stream/x86_64/images/ - https://cloud.centos.org/centos/9-stream/x86_64/images/ 1) I see the c9s images are regularly updated, while the c8s images not; is this wanted, or simply a mistake/miss somewhere? If so, will they be kept up-to-date just like the c9s images? 2) are the Vagrant images there officially supported? Let's say you use one, and find a bug; then check on a "classic" installation (baremetal/VM), and the bug isn't there; can users report problems in those cases? 3) would it be possible to have either unversioned or "-latest" symlinks for the latest versions of all the files (i.e. images, checksums, etc)? This way, it will be much easier for users to fetch the latest versions with no need to change URLs frequently. Of course, the assumption is that users of this are always willing to use the latest version. Thanks! -- Pino Toscano

7 14

RFC (Request for Comments) : future of bugs.centos.org
by Fabian Arrotin 10 May '23

10 May '23

Let's discuss https://bugs.centos.org ! For a long time, we used to only have one public tracker , powered by MantisBT (https://mantisbt.org/) , and it was https://bugs.centos.org Due to various changes in the CentOS ecosystem in the last years, some categories were removed/migrated elsewhere : - CentOS infra tracker : https://pagure.io/centos-infra/issues - CentOS Stream : https://bugzilla.redhat.com Some other SIGs also started to have (and document) their own tracker (see https://wiki.centos.org/ReportBugs) So the question is : while we can easily migrate bugs.centos.org to el9, should we ? All comments/remarks are welcome, and let's have a plan about what to do Kind Regards, -- Fabian Arrotin The CentOS Project | https://www.centos.org gpg key: 17F3B7A1 | @arrfab[@fosstodon.org]

2 4

python3.11-devel in CS8 (and RHEL8)
by Orion Poplawski 06 Apr '23

06 Apr '23

Pythno 3.11 and some modules have appeared in CentOS Stream 8 appstream. However I don't see python3.11-devel and the other modules that are in CentOS Stream 9 crb in CS8 powertools. I do see them in CS9 koji. Is this just an oversight or some kind of lag? Or is this intentional? It will make it impossible for EPEL to build needed modules for EPEL 8. I've also commented in the python 3.11 RFE here: https://bugzilla.redhat.com/show_bug.cgi?id=2137139 Thanks for any information. -- Orion Poplawski he/him/his - surely the least important thing about me IT Systems Manager 720-772-5637 NWRA, Boulder/CoRA Office FAX: 303-415-9702 3380 Mitchell Lane orion(a)nwra.com Boulder, CO 80301 https://www.nwra.com/

6 11

Are non-Red Hat folks not allowed to post comments to CS Jira?
by Miro Hrončok 03 Apr '23

03 Apr '23

Hello CentOS folks. I've been told "I wanted to leave a comment ..., but I apparently don't have sufficient permissions (it seems to be read only)" about this ticket: https://issues.redhat.com/browse/CS-1502 Is that by design? -- Miro Hrončok -- Phone: +420777974800 IRC: mhroncok

2 1

RFC (Request for Comments) : future of forums.centos.org
by Fabian Arrotin 29 Mar '23

29 Mar '23

Let's chat about forums.centos.org ! IMHO, that service can be easily moved to a newer OS/PHP stack and so continue to run. But on the other hand, I heard rumors that some forums moderators would like to stop maintaining/moderating it after 2024. I also heard that there is now a centos category on fedora discourse (https://discussion.fedoraproject.org/c/centos/71) So while we (centos infra) can easily migrate that server, should we ? Let's spend some time investigating the options and take a decision about it ! Kind Regards, -- Fabian Arrotin The CentOS Project | https://www.centos.org gpg key: 17F3B7A1 | @arrfab[@fosstodon.org]

5 6

[Action Required] : Decommissioning (on premises) OpenShift CI cluster
by Fabian Arrotin 29 Mar '23

29 Mar '23

Hi all, Camilla announced this in december already (https://lists.centos.org/pipermail/ci-users/2022-December/thread.html) and we sent multiple reminders about that. The plan was already knew since it was decided to move the whole CI infra to AWS (announced in June 2022) but the last bit that had to move was the openshift CI cluster (aka *.apps.ocp.ci.centos.org) to AWS (https://console-openshift-console.apps.ocp.cloud.ci.centos.org/) As the new cluster in AWS is deployed since end of November you still have to migrate your workload from old to new cluster, and that *before* end of March 2023 (see https://sigs.centos.org/guide/ci/#migration-to-new-ci-instance) We'll continue to send such reminders as long as we see tenants who haven't migrated (yet) We prefer announce it multiple times so that nobody will ask "where is my app gone ?" when the whole cluster will be powered down end of March. PS: if you have already migrated, please confirm to us so that we can already delete your project/namespace/storage PV in the previous cluster. Kind Regards, -- Fabian Arrotin The CentOS Project | https://www.centos.org gpg key: 17F3B7A1 | twitter: @arrfab

3 6

RFC (Request for Comments) : future of https://planet.centos.org
by Fabian Arrotin 29 Mar '23

29 Mar '23

https://planet.centos.org was deployed a long time ago to aggregate feeds from various centos contributors and it was migrated from centos release to centos newer release, as planet (python2 based app) was easy to just rebuild. Now that CentOS Linux 7 will go EOL next year, and that https://blog.centos.org/ (linked to SSO) seems to be the only source for blog posts, I'd like to consider just decommissioning that planet instance, and the "Around CentOS" section of www.centos.org homepage would fetch rss feed from blog.centos.org, instead of planet.centos.org (just acting as a MITM these days) Opinions, thoughts ? -- Fabian Arrotin The CentOS Project | https://www.centos.org gpg key: 17F3B7A1 | @arrfab[@fosstodon.org]

2 2

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

devel March 2023