quine-global-labs/mirror-coreos-fedora-coreos-config
1
0
Fork 0
mirror of https://github.com/coreos/fedora-coreos-config.git synced 2026-11-03 22:59:47 -09:00
Code Issues Projects Releases Packages Wiki Activity
mirror-coreos-fedora-coreos.../image-base.yaml

33 lines
1.1 KiB
YAML
Raw Normal View History

Move next-devel to cgroupsv2 for new nodes All we need to do is to make the associated karg be specific to `testing-devel` by uplisting it to `image.yaml`. Then `next-devel` will naturally shed it when `image-base.yaml` gets synced over. See: https://github.com/coreos/fedora-coreos-tracker/issues/292
2021-03-26 07:12:58 -08:00
# This file is shared by all streams. For a stream-specific change, use
# image.yaml instead.
image.yaml: split out image-base.yaml The `image.yaml` file will be unique to each stream (like `manifest.yaml`), while `image-base.yaml` will be shared by all streams (like `manifests/fedora-coreos-base.yaml`). Prep for dropping the `systemd.unified_cgroup_hierarchy=0` karg on the next stream.
2021-03-24 13:48:10 -08:00
# Target disk size in GB.
# Make it at least 10G because we want the rootfs to be at least 8G:
# https://github.com/coreos/fedora-coreos-tracker/issues/586
size: 10
extra-kargs:
# Disable SMT on systems vulnerable to MDS or any similar future issue.
- mitigations=auto,nosmt
# Disable networking by default on firstboot. We can drop this once cosa stops
switch default networking config to ip=auto Upstream NM changed the behavior of `ip=dhcp,dhcp6` to also wait 20 seconds for DHCPv6. We don't want to require all users by default to wait for DHCPv6 especially when it's doubtful most users even have a ipv6 DHCP server. We can stick with our current behavior by utilizing `ip=auto` instead. Fixes https://github.com/coreos/fedora-coreos-tracker/issues/1000
2021-11-02 10:51:36 -08:00
# defaulting to `ip=auto rd.neednet=1` when it doesn't see this key.
image.yaml: split out image-base.yaml The `image.yaml` file will be unique to each stream (like `manifest.yaml`), while `image-base.yaml` will be shared by all streams (like `manifests/fedora-coreos-base.yaml`). Prep for dropping the `systemd.unified_cgroup_hierarchy=0` karg on the next stream.
2021-03-24 13:48:10 -08:00
ignition-network-kcmdline: []
# Optional remote by which to prefix the deployed OSTree ref
ostree-remote: fedora
image-base: specify VMware OS and HW versions Stick with hardware version 13 for now, but switch the OS ID from RHEL 7 to 64-bit Fedora. For https://github.com/coreos/coreos-assembler/pull/2762.
2022-03-16 22:10:00 -08:00
vmware-os-type: fedora64Guest
# VMware hardware versions: https://kb.vmware.com/s/article/1003746
# We use the newest version allowed by the oldest non-EOL VMware
# Workstation/Player/Fusion/ESXi release: https://lifecycle.vmware.com/
image-base: bump VMware hardware version to 17 The last VMware products requiring older hardware versions are now EOL. Fixes https://github.com/coreos/fedora-coreos-tracker/issues/1141.
2022-10-17 07:43:06 -08:00
vmware-hw-version: 17
image-base.yaml: set `compressor: xz` This will be picked up by `cosa compress` to know to compress with xz.
2022-11-21 13:39:09 -09:00
compressor: xz
image-base.yaml: compress applehv with gzip, hyperv with zip These platforms are already compressed using those compressors today, but we're moving that policy out of cosa and into here since it's not a requirement from the platform itself and more an opinion (which others may want to differ from). This uses a new `platform-compressor` key which `cosa compress` will learn to read.
2024-12-03 08:15:32 -09:00
platform-compressor:
# On Windows, it's the only compression tool we can rely on.
hyperv: zip
# On macOS, it's the only compression tool we can rely on.
applehv: gzip
Reference in a new issue Copy permalink