wait for 10 seconds after sending SIGTERM

[buzz0r]

It's better to wait until the main ioBorker controller process is finished with it's shutdown.
If we don't wait for the clean shutdown, the adapter hm-rpc can not disconnect from the ccu gateway and is not working after a container restart.
See:
ioBroker/ioBroker.hm-rpc#595

[buanet]

Thank you. But this is not what I had planned for optimizing the graceful shutdown.
I'm now testing a different timeout method (inspired by the rewritten maintenance script) to shorten the timeout time.
Maybe you can take a look at b8c67b7 and check if the solution will also fix the hm-rpc issue.
The solution will be available with new beta version within the next days.

BTW: I'm using Homematic and rpc/ xmlrpx too. I've never faced that issue so far...

Regards,
André

[buzz0r]

Hi,
if I understand the proposed pull request correctly, so far, it’s only handling the maintance.sh script.
But if a user does, for example, a "docker-compose stop" pkill is till instantly killing the main process without giving the running adapters some time to do a clean shutdown.

I could do another pull request based on the timeout implementation used in the maintance.sh script and adapt it in the iobroker_startup.sh, if that would be better fitting in the project.
What do you think?

[buanet]

Hi,
please take a closer look here: b8c67b7#diff-d7a77b1b100f0e35f6ffadd80381e880c07fc52f4db3dc347d5af9079c87122f
I already implemented the solution from the rewritten maintenance script so the startup script with yesterdays commit.
So no more pr needed.

However. I will publish the latest adjustments as new beta version soon. Please check if this helps you with your issue.

Regards,
André

[buzz0r]

Ah sorry, I missed this file change! Looks good, but of course I will test it as soon as a new beta release is available!
Thx again!