CroatianMeteorNetwork / RMS

RPi Meteor Station
https://globalmeteornetwork.org/
GNU General Public License v3.0
169 stars 47 forks source link

reboot_try in range(4*60) causing failure to reboot #158

Closed g7gpr closed 1 year ago

g7gpr commented 1 year ago

https://github.com/CroatianMeteorNetwork/RMS/blob/f93dba60f7aa6df89981ff54452fd2f4e8b85731/RMS/StartCapture.py#L794

Commit e0f2b02 made target of RunExternalScript an external process.

I think this means that the 4 hour timer starts running from the end of the observation, rather than the end of RunExternalScript.

On (only) one occasion the external script took a very long time, about 6 hours. This happened on at least two separate systems, both with good internet connections. I believe that the iStraStream server was not accepting connections, and the 6 hours was the accumulated timeout of each stage of the iStraStream script.

Can you make the timer longer, so that a failed external script does not block the reboot. My fear is that if iStraStream was down for an extended period of time, it would prevent any automatic updates of the code until operator intervention.

An ideal length for the timer would be so that it timed out a defined number of minutes before observation start, killed the external script, and then behave as though the external script had run to completion.

g7gpr commented 1 year ago

Fix no longer required.