too much log file generated

Hi all,

I just wanted to post a problem met by one of my colleagues.

When the settings.json file is "broken" (if you miss a comma at the end of the line for example) Orthanc is generating continuously many log files of 1Kb saying that the startup procedure has failed, and this log generating process is never ending.

My colleague broke something in his settings.json file and didn't fix it for several weeks.

When I gone to his computer to fix it, I found 1.4 million of log files, eating almost 600 Mb of hard disk space (during all these weeks the Orthanc service generated multiple log each seconds....).

The log folder had so many files that windows explorer took almost half an hour to display them and the erase process was also very long (80 files deletion per second for 1 400 000 files...).

Best regards,

Salim

Hello,

I’ve just filed an issue about this topic:
https://bitbucket.org/sjodogne/orthanc/issues/38/limit-the-number-of-automatic-restarts-by

Regards,
Sébastien-

I am in the process of removing over 75,000 log files. Just realized that I left it autostart for a few months when I was not actually using it.

There is an open source workaround or solution to this, to use NSSM Doc which has the automatic back-off feature if the service crashes or does not run for more than X number of milliseconds not to hold off the restart cycle.

Perhaps, Orthanc may want to model after that.

I think you'd be best served by tools dedicated to these kinds of
functionalities like journald or logrotate.

Discussion on alternatives for Windows:
https://serverfault.com/q/358172

Hello,

I am in the process of removing over 75,000 log files. Just realized that I left it autostart for a few months when I was not actually using it.

There is an open source workaround or solution to this, to use NSSM Doc which has the automatic back-off feature if the service crashes or does not run for more than X number of milliseconds not to hold off the restart cycle.

There is already a pending issue in our bug tracker about improving this behavior:
https://bitbucket.org/sjodogne/orthanc/issues/38/limit-the-number-of-automatic-restarts-by

Sébastien-