Containers not running in TorizonCore with evaluation containers for iMX8M-Plus

vix · March 27, 2023, 1:25pm

I’ve just downloaded the last TorizonCore with evaluation containers image (build 220 for Verdin iMX8M-Plus) and installed using TEZI.
TorizonCore starts, and I can connect through ssh.
But I can’t see portainer page on HDMI monitor.
If I connect ssh and I run docker image ls I see

torizon@verdin-imx8mp-14777722:~$ docker image ls
REPOSITORY               TAG       IMAGE ID       CREATED                  SIZE
torizon/weston-vivante   3         779d54f7f4d1   Less than a second ago   510MB
portainer/portainer-ce   2.17.1    ada025d39772   Less than a second ago   267MB
torizon/chromium         2         cb7ac4913734   6 months ago             609MB

And docker container ls gives


torizon@verdin-imx8mp-14777722:~$ docker container ls
CONTAINER ID   IMAGE                    COMMAND                  CREATED         STATUS         PORTS
                                        NAMES
537452a778ee   portainer/portainer-ce   "/portainer --templa…"   6 minutes ago   Up 6 minutes   8000/tcp, 9443/tcp, 0.0.0.0:8840->9000/tcp, :::8840->9000/tcp   torizon-portainer-1

It seems that weston-vivante and chromium containers are not running as expected.

If I try the connection with ApolloX on WSL (I have version 0.0.67 and I think it’s the last one), the connection seems ok and I can see the logs of the containers
immagine
For chromium


 *  Executing task: docker logs --tail 1000 -f da5140c384d1ad9f6b64a527360e3bef158ae1446ed07501906f1981eaec043a 

[1:1:0223/003833.873126:ERROR:wayland_connection.cc(209)] Failed to connect to Wayland display
[1:1:0223/003833.873262:ERROR:ozone_platform_wayland.cc(226)] Failed to initialize Wayland platform
[1:1:0223/003833.873279:ERROR:env.cc(225)] The platform failed to initialize.  Exiting.

For weston-vivante

 *  Executing task: docker logs --tail 1000 -f 4f8c0eef313349fb069c92dbf3f8baa02a15fe46a71c540f245c22c2bfca98c6 

NXP EULA has already been accepted.
SoC is: 'i.MX8MP'
SoC has GPU: true
SoC has DPU: false
g2d implementation: viv
Couldn't open /dev/tty1
/usr/bin/entry.sh: line 86: [: =: unary operator expected
Removing previously created '.X*-lock' entries under /tmp before starting Weston. Pass 'IGNORE_X_LOCKS=1' environment variable to Weston container to disable this behavior.
dos2unix: converting file /etc/xdg/weston/weston.ini to Unix format...
dos2unix: converting file /etc/xdg/weston-dev/weston.ini to Unix format...
00:00:00.000 [INFO] [seatd/seat.c:39] Created VT-bound seat seat0
00:00:00.000 [INFO] [seatd/seatd.c:194] seatd started
Date: 2023-02-23 UTC
[00:38:32.159] weston 10.0.1
               https://wayland.freedesktop.org
               Bug reports to: https://gitlab.freedesktop.org/wayland/weston/issues/
               Build: lf-5.15.52-2.1.0-10-g9452feba
[00:38:32.160] Command line: weston -Bdrm-backend.so --current-mode -Swayland-0
[00:38:32.160] OS: Linux, 5.15.77-6.2.0-devel+git.3f45ee6bd117, #1-TorizonCore SMP PREEMPT Thu Mar 23 13:16:48 UTC 2023, aarch64
[00:38:32.160] Flight recorder: enabled
[00:38:32.160] Using config file '/etc/xdg/weston/weston.ini'
[00:38:32.161] Output repaint window is 7 ms maximum.
[00:38:32.161] Loading module '/usr/lib/aarch64-linux-gnu/libweston-10/drm-backend.so'
[00:38:32.169] initializing drm backend
[00:38:32.169] Trying libseat launcher...
00:00:00.043 [INFO] [seatd/server.c:145] New client connected (pid: 27, uid: 1000, gid: 1000)
00:00:00.043 [INFO] [seatd/seat.c:170] Added client 1 to seat0
00:00:00.043 [ERROR] [common/terminal.c:162] Could not open target tty: Operation not permitted
00:00:00.043 [ERROR] [seatd/seat.c:72] Could not open terminal for VT 1: Operation not permitted
00:00:00.043 [ERROR] [seatd/seat.c:461] Could not open VT for client
00:00:00.043 [ERROR] [common/terminal.c:162] Could not open target tty: Operation not permitted
00:00:00.043 [ERROR] [seatd/seat.c:86] Could not open terminal to clean up VT 1: Operation not permitted
[00:38:32.170] libseat: session control granted
00:00:00.046 [ERROR] [seatd/seat.c:222] Could open device: client is not active
00:00:00.046 [ERROR] [seatd/client.c:238] Could not open device: Operation not permitted
00:00:00.047 [ERROR] [seatd/seat.c:222] Could open device: client is not active
00:00:00.047 [ERROR] [seatd/client.c:238] Could not open device: Operation not permitted
[00:38:32.176] no drm device found
00:00:00.050 [ERROR] [common/terminal.c:162] Could not open target tty: Operation not permitted
00:00:00.050 [ERROR] [seatd/seat.c:86] Could not open terminal to clean up VT 1: Operation not permitted
00:00:00.050 [INFO] [seatd/seat.c:524] Closed client 1 on seat0
00:00:00.050 [INFO] [seatd/seat.c:192] Removed client 1 from seat0
00:00:00.050 [INFO] [seatd/client.c:471] Client disconnected
[00:38:32.176] fatal: failed to create compositor backend
Internal warning: debug scope 'drm-backend' has not been destroyed.
00:00:00.054 [INFO] [seatd/seatd.c:218] seatd stopped
Switching back to vt 1

Is there something wrong?

EDIT: I can add that the Monthly TorizonCore with evaluation containers
6.2.0-devel-202303+build.6.container for iMX8M-Plus shows Portainer page on HDMI monitor.
And I can see the following images

torizon@verdin-imx8mp-14777722:~$ docker image ls
REPOSITORY               TAG       IMAGE ID       CREATED                  SIZE
portainer/portainer-ce   2.17.1    ada025d39772   Less than a second ago   267MB
torizon/chromium         2         cb7ac4913734   6 months ago             609MB
torizon/weston-vivante   2         eb17fa1e613f   6 months ago             424MB

I see that all the containers share the major tag 2.
Not sur eif mixing 2 and 3 like the lates noghtly builds is a good idea or not.

lucas_a.tx · March 27, 2023, 2:51pm

Hi @vix ,

For the next TorizonCore 6 releases we’re currently in the process of porting our Debian container images from Bullseye (tag 2) to Bookworm (tag 3). Given that our nightly builds are used for active development of new features and bug fixes, it’s expected that these type of releases will not be necessarily in a working state, as we don’t test them. You can see more details about our support strategy here: Toradex Embedded Linux Support Strategy | Toradex Developer Center

I see that all the containers share the major tag 2.
Not sur eif mixing 2 and 3 like the lates noghtly builds is a good idea or not.

The plan is to port all of our current Debian container images to tag 3 (Bookworm), but not all images are ported yet. I’d recommend using either our Monthly or Quarterly releases for the development of your project, always keeping in mind our support strategy for these kind of releases as well.

Best regards,
Lucas Akira

vix · March 27, 2023, 3:12pm

The problem is that I’m actively working with Toradex engineers to have new features implemented (basically Cortex-M firmware and 3D acceleration using GPU).
All these features are enabled in the nightlies and so I’m forced to follow them.
I know the bookworrm migration path, but for me it would be great if the migration can be finalized asap.

I cannot use the quarterly, I give a try to the monthly but I hope in the nightly…

lucas_a.tx · March 27, 2023, 9:45pm

Hi @vix ,

I know the bookworrm migration path, but for me it would be great if the migration can be finalized asap.

I understand your concern about the migration, but for the moment there isn’t too much that can be done in this regard other than to wait for the team here to finish their work. I’ll let the team know about your concern though.

Best regards,
Lucas Akira

vix · March 29, 2023, 10:26am

Hi @lucas_a.tx
thinking a little bit more on what happens, I think that there is another underlying issue:
the only container that has already been ported to bookworm is weston-vivante, and this is the container that fails.

*  Executing task: docker logs --tail 1000 -f 4f8c0eef313349fb069c92dbf3f8baa02a15fe46a71c540f245c22c2bfca98c6 

NXP EULA has already been accepted.
SoC is: 'i.MX8MP'
SoC has GPU: true
SoC has DPU: false
g2d implementation: viv
Couldn't open /dev/tty1
/usr/bin/entry.sh: line 86: [: =: unary operator expected
Removing previously created '.X*-lock' entries under /tmp before starting Weston. Pass 'IGNORE_X_LOCKS=1' environment variable to Weston container to disable this behavior.
dos2unix: converting file /etc/xdg/weston/weston.ini to Unix format...
dos2unix: converting file /etc/xdg/weston-dev/weston.ini to Unix format...
00:00:00.000 [INFO] [seatd/seat.c:39] Created VT-bound seat seat0
00:00:00.000 [INFO] [seatd/seatd.c:194] seatd started
Date: 2023-02-23 UTC
[00:38:32.159] weston 10.0.1
               https://wayland.freedesktop.org
               Bug reports to: https://gitlab.freedesktop.org/wayland/weston/issues/
               Build: lf-5.15.52-2.1.0-10-g9452feba
[00:38:32.160] Command line: weston -Bdrm-backend.so --current-mode -Swayland-0
[00:38:32.160] OS: Linux, 5.15.77-6.2.0-devel+git.3f45ee6bd117, #1-TorizonCore SMP PREEMPT Thu Mar 23 13:16:48 UTC 2023, aarch64
[00:38:32.160] Flight recorder: enabled
[00:38:32.160] Using config file '/etc/xdg/weston/weston.ini'
[00:38:32.161] Output repaint window is 7 ms maximum.
[00:38:32.161] Loading module '/usr/lib/aarch64-linux-gnu/libweston-10/drm-backend.so'
[00:38:32.169] initializing drm backend
[00:38:32.169] Trying libseat launcher...
00:00:00.043 [INFO] [seatd/server.c:145] New client connected (pid: 27, uid: 1000, gid: 1000)
00:00:00.043 [INFO] [seatd/seat.c:170] Added client 1 to seat0
00:00:00.043 [ERROR] [common/terminal.c:162] Could not open target tty: Operation not permitted
00:00:00.043 [ERROR] [seatd/seat.c:72] Could not open terminal for VT 1: Operation not permitted
00:00:00.043 [ERROR] [seatd/seat.c:461] Could not open VT for client
00:00:00.043 [ERROR] [common/terminal.c:162] Could not open target tty: Operation not permitted
00:00:00.043 [ERROR] [seatd/seat.c:86] Could not open terminal to clean up VT 1: Operation not permitted
[00:38:32.170] libseat: session control granted
00:00:00.046 [ERROR] [seatd/seat.c:222] Could open device: client is not active
00:00:00.046 [ERROR] [seatd/client.c:238] Could not open device: Operation not permitted
00:00:00.047 [ERROR] [seatd/seat.c:222] Could open device: client is not active
00:00:00.047 [ERROR] [seatd/client.c:238] Could not open device: Operation not permitted
[00:38:32.176] no drm device found
00:00:00.050 [ERROR] [common/terminal.c:162] Could not open target tty: Operation not permitted
00:00:00.050 [ERROR] [seatd/seat.c:86] Could not open terminal to clean up VT 1: Operation not permitted
00:00:00.050 [INFO] [seatd/seat.c:524] Closed client 1 on seat0
00:00:00.050 [INFO] [seatd/seat.c:192] Removed client 1 from seat0
00:00:00.050 [INFO] [seatd/client.c:471] Client disconnected
[00:38:32.176] fatal: failed to create compositor backend
Internal warning: debug scope 'drm-backend' has not been destroyed.
00:00:00.054 [INFO] [seatd/seatd.c:218] seatd stopped
Switching back to vt 1

Chromium has not been ported yet, but it can’t run since it depends on weston-vivante.

The root error seems

Couldn't open /dev/tty1
...
00:00:00.043 [ERROR] [common/terminal.c:162] Could not open target tty: Operation not permitted

The question is: why the bookworm container weston-vivante fails on iMX8M-Plus?

lucas_a.tx · March 30, 2023, 8:52pm

Hi @vix ,

The question is: why the bookworm container weston-vivante fails on iMX8M-Plus?

I’ll pass this issue in particular to the TorizonCore team. For now the team is actively in the process of porting our Debian containers to Bookworm, which includes the ones you tested, so this problem will eventually be analyzed internally.

Keep an eye out at our News Announcement section for any updates on these containers.

Best regards,
Lucas Akira

leon.tx · March 31, 2023, 7:17pm

Hi @vix, I’m from the TorizonCore team.

why the bookworm container weston-vivante fails on iMX8M-Plus?

This is a known problem to us and the fix is at the end of our release pipeline, so it will be available shortly.

Best Regards,

Leonardo Held.