Container keeps restarting after update ResinOS to 2.13.6

raspberrypi3
docker

#1

Hi there,

i’ve had this before; after upgrading an older image to the latest ResinOS, I can’t get it to run correctly. My previous solution was to install the module new with a fresh image download, but now I have a device where I can’t do this so easily.

The logs:

15.08.18 08:30:11 (+0200) Installing service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:16 (+0200) Installed service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:17 (+0200) Starting service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:19 (+0200) Systemd init system enabled.
15.08.18 08:30:19 (+0200) systemd 215 running in system mode. (+PAM +AUDIT +SELINUX +IMA +SYSVINIT +LIBCRYPTSETUP +GCRYPT +ACL +XZ -SECCOMP -APPARMOR)
15.08.18 08:30:19 (+0200) Detected virtualization 'other'.
15.08.18 08:30:19 (+0200) Detected architecture 'arm'.
15.08.18 08:30:19 (+0200) Set hostname to <0709d27>.
15.08.18 08:30:20 (+0200) Started service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:21 (+0200) Killing service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:22 (+0200) Sending SIGTERM to remaining processes...
15.08.18 08:30:22 (+0200) Sending SIGKILL to remaining processes...
15.08.18 08:30:22 (+0200) Unmounting file systems.
15.08.18 08:30:22 (+0200) Unmounting /sys/kernel/debug.
15.08.18 08:30:22 (+0200) Unmounting /lib/modules.
15.08.18 08:30:22 (+0200) Unmounting /dev/mqueue.
15.08.18 08:30:22 (+0200) All filesystems unmounted.
15.08.18 08:30:22 (+0200) Halting system.
15.08.18 08:30:26 (+0200) Service exited 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:27 (+0200) Killed service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:27 (+0200) Installing service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:33 (+0200) Installed service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:33 (+0200) Starting service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:35 (+0200) Started service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:35 (+0200) Systemd init system enabled.
15.08.18 08:30:35 (+0200) systemd 215 running in system mode. (+PAM +AUDIT +SELINUX +IMA +SYSVINIT +LIBCRYPTSETUP +GCRYPT +ACL +XZ -SECCOMP -APPARMOR)
15.08.18 08:30:35 (+0200) Detected virtualization 'other'.
15.08.18 08:30:35 (+0200) Detected architecture 'arm'.
15.08.18 08:30:35 (+0200) Set hostname to <0709d27>.
15.08.18 08:30:36 (+0200) Killing service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:37 (+0200) Sending SIGTERM to remaining processes...
15.08.18 08:30:37 (+0200) Sending SIGKILL to remaining processes...
15.08.18 08:30:37 (+0200) Unmounting file systems.
15.08.18 08:30:37 (+0200) Unmounting /sys/kernel/debug.
15.08.18 08:30:37 (+0200) Unmounting /lib/modules.
15.08.18 08:30:37 (+0200) Unmounting /dev/mqueue.
15.08.18 08:30:37 (+0200) All filesystems unmounted.
15.08.18 08:30:37 (+0200) Halting system.
15.08.18 08:30:43 (+0200) Service exited 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:44 (+0200) Killed service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:44 (+0200) Installing service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:50 (+0200) Installed service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:50 (+0200) Starting service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:52 (+0200) Started service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:52 (+0200) Systemd init system enabled.
15.08.18 08:30:52 (+0200) systemd 215 running in system mode. (+PAM +AUDIT +SELINUX +IMA +SYSVINIT +LIBCRYPTSETUP +GCRYPT +ACL +XZ -SECCOMP -APPARMOR)
15.08.18 08:30:52 (+0200) Detected virtualization 'other'.
15.08.18 08:30:52 (+0200) Detected architecture 'arm'.
15.08.18 08:30:52 (+0200) Set hostname to <0709d27>.
15.08.18 08:30:52 (+0200) Killing service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:53 (+0200) Sending SIGTERM to remaining processes...
15.08.18 08:30:53 (+0200) Sending SIGKILL to remaining processes...
15.08.18 08:30:53 (+0200) Unmounting file systems.
15.08.18 08:30:53 (+0200) Unmounting /sys/kernel/debug.
15.08.18 08:30:53 (+0200) Unmounting /lib/modules.
15.08.18 08:30:54 (+0200) Unmounting /dev/mqueue.
15.08.18 08:30:54 (+0200) All filesystems unmounted.
15.08.18 08:30:54 (+0200) Halting system.
15.08.18 08:30:58 (+0200) Service exited 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:59 (+0200) Killed service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'
15.08.18 08:30:59 (+0200) Installing service 'main sha256:b5d3ac42849aa3b6d560dc0f3e64cfd65b54430ca9f9bf01b1b3d8fa5af936bf'

Even stranger; it seems that it will start working WHILE it’s downloading a pushed update, w(I tested allot to get it up and running).

My steps where:

  1. I updated an existing device to ResinOS 2.13.6, that crippled the Application (because of a driver with incorrect kernel headers)
  2. Created new application, moved Device there to test stuff, Changed Dockerfile for new kernel headers, from then on; only this reboot situation.
  3. Removed ins mod driver load, still no luck
  4. Changed application to be a infinite CMD loop, still no luck

How should I proceed?


#4

Hi @salorob, will send you a private message for the device link, and we’ll take a look.


#7

Hi, thanks, we’ve received the link, and looking into the issue!


#9

Hi, sorry that it took this long, but figured it out.

  • The device name had a stray newline at the end.
  • The latest supervisors have an issue with newlines where they shouldn’t be (as it’s being more strict than the previous versions)
  • Removed the newline from the end of the name, and the device is running now.

Let us know if you have any other issues with the device! We are putting more safeguards in place for this not to happen in the future.