Bug 7690 - Fatal crashes aren't sent to syslog
Summary: Fatal crashes aren't sent to syslog
Status: NEW
Alias: None
Product: ThinLinc
Classification: Unclassified
Component: Other (show other bugs)
Version: trunk
Hardware: PC Unknown
: P2 Normal
Target Milestone: LowPrio
Assignee: Bugzilla mail exporter
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2021-04-20 13:14 CEST by Pierre Ossman
Modified: 2024-04-16 13:59 CEST (History)
0 users

See Also:
Acceptance Criteria:


Attachments

Description Pierre Ossman cendio 2021-04-20 13:14:58 CEST
If one of our services crashes, then the traceback is automatically printed by Python. However it is printed to stderr, which means that it does not go through the proper log system.

Right now that mostly works anyway as we set up stderr to point to the proper log file as a fallback. However the formatting is broken.

If ThinLinc is configured to use syslog though then the lines will be lost (or end up in a log file that normally doesn't exist). This would be the norm if we always use syslog (e.g. bug 5452).


So we should set up sys.excepthook for all our services, just like we do for tl-setup already.
Comment 1 Pierre Ossman cendio 2024-04-16 13:59:02 CEST
This is less of a problem after bug 7636. Very few crashes are now fatal, and will instead by caught by asyncio and logged by our exception handler.

Note You need to log in before you can comment on or make changes to this bug.