Support ignition in simplified-installer and raw-image #3130
Conversation
|
/packit build |
|
sorry about that ^ Friday morning infrastructure problems -_- |
runcom
force-pushed
the
ignition-poc
branch
4 times, most recently
from
fbf95a3
to
f74bfbb
Compare
There was a problem hiding this comment.
Manifest-diff job [0] as manifests.diff.
There was a problem hiding this comment.
Manifest-diff job [0] as manifests.diff.
There was a problem hiding this comment.
Manifest-diff job [0] as manifests.diff.
There was a problem hiding this comment.
Manifest-diff job [0] as manifests.diff.
runcom
force-pushed
the
ignition-poc
branch
4 times, most recently
from
ede5e0a
to
66b12b1
Compare
There was a problem hiding this comment.
Manifest-diff job [0] as manifests.diff.
runcom
force-pushed
the
ignition-poc
branch
2 times, most recently
from
f1919fb
to
16d58ab
Compare
|
blueprint customization options on #3161 |
There was a problem hiding this comment.
Thanks, this all looks good. Some minor comments and one potentially important one below.
I'd also like a more descriptive commit message on the last commit ("wire ignition").
Also, in the kernel options in the manifests I'm seeing ...modprobe.blacklist=vc4,rw,,coreos.no_persist_ip,... (an empty option between rw and coreos.no_persist_ip) but I can't pinpoint where it's coming from.
There was a problem hiding this comment.
Manifest-diff job [0] as manifests.diff.
that's one last WIP I have - I'll polish everything up for a final review anyway |
There was a problem hiding this comment.
Manifest-diff job [0] as manifests.diff.
There was a problem hiding this comment.
Manifest-diff job [0] as manifests.diff.
|
tests are green - I'm gonna polish this all tomorrow, pending a tiny decision about |
runcom
force-pushed
the
ignition-poc
branch
2 times, most recently
from
07c84f6
to
8c1ea62
Compare
@achilleas-k I found it, comes from |
There was a problem hiding this comment.
Manifest-diff job [0] as manifests.diff.
|
should be good for another review now 🤔 tests are green |
Signed-off-by: Antonio Murdaca <antoniomurdaca@gmail.com>
Signed-off-by: Antonio Murdaca <runcom@linux.com> Signed-off-by: Antonio Murdaca <antoniomurdaca@gmail.com> Signed-off-by: Irene Diez <idiez@redhat.com> Co-authored-by: Irene Diez <idiez@redhat.com> Signed-off-by: Antonio Murdaca <antoniomurdaca@gmail.com>
Signed-off-by: Antonio Murdaca <antoniomurdaca@gmail.com>
Signed-off-by: Antonio Murdaca <antoniomurdaca@gmail.com>
Signed-off-by: Antonio Murdaca <antoniomurdaca@gmail.com>
Signed-off-by: Antonio Murdaca <antoniomurdaca@gmail.com>
runcom
force-pushed
the
ignition-poc
branch
2 times, most recently
from
1f00476
to
e16b825
Compare
Signed-off-by: Antonio Murdaca <antoniomurdaca@gmail.com>
There was a problem hiding this comment.
Manifest-diff job [0] as manifests.diff.
This patchset is the basic work needed to support ignition in r4e - it doesn't include any blueprint or file injection change as I wanted to maintain it easy to review and since this won't break anything, I think it'll be beneficial to merge this before other work to plumb ignition files/config in the blueprints. This PR also includes a commit that enables kargs injection from the installer and raw-image blueprint to effectively enable passing the artifact an
ignition.config.urland test it out. Lastly, to make the work bound to the sysroot.ro=true patch, this also includes #3053Things missing:
ignition-edgeRPM isn't yet packaged anywhere (working with the coreos team to ship it), the source lives https://github.com/fedora-iot/ignition-edge - https://gitlab.com/redhat/centos-stream/rpms/ignition/-/merge_requests/13