Open leograba opened 6 months ago
I suspect it happens due to the dependency of docker-binfmt
on Docker:
RDEPENDS:${PN} += " bash docker"
One possible alternative would be to use VIRTUAL-RUNTIME_container_engine
instead:
RDEPENDS:${PN} += " bash ${VIRTUAL-RUNTIME_container_engine}"
This is possible due to such variable being set on the images:
layers$ grep -nrie "VIRTUAL-RUNTIME_container" .
./meta-common-torizon/recipes-images/images/torizon-core-container.inc:2: ${VIRTUAL-RUNTIME_container_engine} \
./meta-common-torizon/recipes-images/images/torizon-core-podman.bb:8:VIRTUAL-RUNTIME_container_engine = "podman"
./meta-common-torizon/recipes-images/images/torizon-core-docker.bb:7:VIRTUAL-RUNTIME_container_engine = "docker-ce"
Even though the package depends on the systemd docker.service
, it might work. Inspecting a device with torizon-core-podman
, there is a symlink:
root@verdin-imx8mm-07276363:/var/rootdirs/home/torizon# ls -l /etc/systemd/system/docker.service
lrwxrwxrwx 1 root root 38 Dec 27 00:17 /etc/systemd/system/docker.service -> /usr/lib/systemd/system/podman.service
root@verdin-imx8mm-07276363:/var/rootdirs/home/torizon# ls -l /usr/etc/systemd/system/docker.service
lrwxrwxrwx 5 root root 38 Dec 20 11:01 /usr/etc/systemd/system/docker.service -> /usr/lib/systemd/system/podman.service
root@verdin-imx8mm-07276363:/var/rootdirs/home/torizon# systemctl status docker.service
● podman.service - Podman API Service
Loaded: loaded (/usr/lib/systemd/system/podman.service; enabled; vendor preset: enabled)
Active: active (running) since Wed 2024-01-03 20:57:47 UTC; 53min ago
TriggeredBy: ● podman.socket
Docs: man:podman-system-service(1)
Main PID: 639 (podman)
Tasks: 10 (limit: 1740)
Memory: 84.4M
CGroup: /system.slice/podman.service
└─ 639 /usr/bin/podman --log-level=info system service -t 0
Jan 03 20:57:51 verdin-imx8mm-07276363 podman[639]: 2024-01-03 20:57:51.788971125 +0000 UTC m=+4.033118001 container init 342fbcfaf20157a647f5a69deccd43500a23492902b83274c5e8105bb694a8cf (image=docker.io/leograba/torizon>
Jan 03 20:57:51 verdin-imx8mm-07276363 podman[639]: 2024-01-03 20:57:51.826243125 +0000 UTC m=+4.070389876 container start 342fbcfaf20157a647f5a69deccd43500a23492902b83274c5e8105bb694a8cf (image=docker.io/leograba/torizo>
Jan 03 20:57:51 verdin-imx8mm-07276363 podman[639]: @ - - [03/Jan/2024:20:57:51 +0000] "POST /v1.41/containers/342fbcfaf20157a647f5a69deccd43500a23492902b83274c5e8105bb694a8cf/start HTTP/1.1" 204 0 "" "Docker-Client/unkn>
Jan 03 20:57:51 verdin-imx8mm-07276363 podman[1170]: [INFO netavark::firewall] Using iptables firewall driver
Jan 03 20:57:51 verdin-imx8mm-07276363 podman[1170]: [INFO netavark::network::netlink] Adding route (dest: 0.0.0.0/0 ,gw: 10.89.0.1, metric 100)
Jan 03 20:57:52 verdin-imx8mm-07276363 podman[639]: time="2024-01-03T20:57:52Z" level=info msg="Running conmon under slice machine.slice and unitName libpod-conmon-9b09b187b7d3583c01973e395040d9dadfcea598983ec1b8313c5c0e>
Jan 03 20:57:52 verdin-imx8mm-07276363 podman[639]: time="2024-01-03T20:57:52Z" level=info msg="Got Conmon PID as 1217"
I didn't test this, though.
The following error happens when trying to build
torizon-core-podman
for the machineqemu86-64
: