dustymabe opened a new pull-request against the project: `fedora-atomic` that you are following:
``
python-docker-py is deprecated
``
To reply, visit the link below or just reply to this email
https://pagure.io/fedora-atomic/pull-request/113
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256
TL;DR: If you use f27 atomic workstation or rawhide atomic host/workstation then you
need to add a new remote and rebase.
[f27 atomic workstation]
# ostree remote add --set=gpg-verify=true --set=gpgkeypath=/etc/pki/rpm-gpg/RPM-GPG-KEY-fedora-27-primary onerepo https://kojipkgs.fedoraproject.org/atomic/repo/
# rpm-ostree rebase onerepo:
[rawhide atomic host/workstation]
# ostree remote add --set=gpg-verify=true --set=gpgkeypath=/etc/pki/rpm-gpg/RPM-GPG-KEY-fedora-29-primary onerepo https://kojipkgs.fedoraproject.org/atomic/repo/
# rpm-ostree rebase onerepo:
Hey Atomic Land,
In Fedora we have migrated to using a single unified repository setup for our
ostree repositories. This means that f28+ all releases will be served out of
the same ostree repo. The repo will be located at https://kojipkgs.fedoraproject.org/atomic/repo/.
If you are a current user of rawhide atomic host/workstation or a current user
of f27 atomic workstation please run the above commands on your machine to
move to the new repository setup. Current users of F27 Atomic Host don't need
to do anything at this time.
Please let us know if you have any questions!
Dusty
-----BEGIN PGP SIGNATURE-----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=x7VK
-----END PGP SIGNATURE-----
ttomecek reported a new issue against the project: `atomic-wg` that you are following:
``
I opened this page: https://fedoraproject.org/wiki/CI/Tests/stat
I was quite shocked, that python-docker-py is on the list. We have renamed the package to python-docker after the big API changes happened in version 2. Right now I'm trying to get rid of python-docker-py.
Would it be possible to include python-docker in Atomic Host compose instead of python-docker-py? The package is probably pulled by atomic CLI, but the fact is that on rawhide atomic RPM requires python3-docker:
```
$ rpm -q --requires atomic | grep docker
python3-docker >= 1.7.2
```
RHEL and CentOS still include python-docker-py because no one requested the newest upstream release yet.
``
To reply, visit the link below or just reply to this email
https://pagure.io/atomic-wg/issue/413
jberkus reported a new issue against the project: `atomic-wg` that you are following:
``
Currently, the main user story for Fedora Atomic Host is as a container infrastructure web host. The OStree ref and ISOs we maintain work well for this user story, and it leads to certain decisions like keeping docker and atomic CLI in the base image (see https://pagure.io/atomic-wg/issue/360)
However, there is a user story we are not satisfying, which is the user who wants a truly minimal install, and is more interested in being able to compose her own ostrees and atomic updates than she is in container orchestration. This is an important user story because we already have demand for it, especially from the IoT sector.
The steps to pursue this would be:
1. work with users to refine and specify the requirements for a "minimal" edition
2. identify and recruit contributors to assist with building it
3. create and publicize new minimal ref and associated tools
Based on my conversations at conferences, the main requests are these:
* reduce the size of the base ostree by stripping things out, preferably to less than 300MB installed.
* auto-rollback on rpm-ostree upgrade reboot fail
* a complete, well-documented toolchain for building and redistributing custom ostrees, based on either the minimal ref or "from scratch".
* support for ARM32 (unlikely at best, but people have asked)
``
To reply, visit the link below or just reply to this email
https://pagure.io/atomic-wg/issue/403
dustymabe opened a new pull-request against the project: `fedora-atomic` that you are following:
``
Remove old python-docker-py library.
``
To reply, visit the link below or just reply to this email
https://pagure.io/fedora-atomic/pull-request/107
dustymabe reported a new issue against the project: `atomic-wg` that you are following:
``
In a few previous issues ([1](https://pagure.io/atomic-wg/issue/379) [2](https://pagure.io/atomic-wg/issue/381) [3](https://pagure.io/atomic-wg/issue/383) we had discussed our strategy regarding automated container builds in dockerhub. Now that we have friends with a hosted container registry we should consider pivoting and using quay.io instead.
``
To reply, visit the link below or just reply to this email
https://pagure.io/atomic-wg/issue/430
clcollins reported a new issue against the project: `atomic-wg` that you are following:
``
The vSphere Cloud Provider for Kubernetes/OpenShift requires that "disk.EnableUUID = True" be set for the node VMs to be able to identify the dynamically provisioned VMDKs that are mounted into the containers for dynamically created persistent storage volumes. (See: https://kubernetes.io/docs/getting-started-guides/vsphere/#enable-vsphere-c… ; and: https://docs.openshift.org/latest/install_config/configuring_vsphere.html#v…)
Setting "disk.EnableUUID = True" allows disks to show up in /dev/disk/by-uuid when they're added to the VM.
We can set it manually each time we create a new node, but the potential to forget is high, and the consequences would be unfortunate (pods would no longer be able to mount storage when scheduled on nodes without the setting).
Considering a lot of the focus on Atomic Host is for Kubernetes & OpenShift, and folks using the vSphere OVA for Atomic would probably like to use the vSphere provisioning with it, this might make sense to be a default setting.
Ideally this would apply to the Red Hat and CentOS Atomic OVAs as well.
Included for more info:
I asked about this in the #Atomic channel, and the replies mentioned -
(03:10:36 PM) walters: it's going to involve changes to https://github.com/redhat-imaging/imagefactory/ most likely
(03:11:35 PM) walters: specifically https://github.com/redhat-imaging/imagefactory/blob/master/imagefactory_plu… I believe
``
To reply, visit the link below or just reply to this email
https://pagure.io/atomic-wg/issue/431
cverna reported a new issue against the project: `atomic-wg` that you are following:
``
The memcached and mariadb container were never built for f27.
The last build for mariadb is mariadb-10.1-13.f26container and for memcached memcached-0-2.f26container_modular.
Also memcached Dockerfile on the f27 is using the rawhide base image. https://src.fedoraproject.org/container/memcached/blob/f27/f/Dockerfile#_1
``
To reply, visit the link below or just reply to this email
https://pagure.io/atomic-wg/issue/424