Is the problem that you're getting too many logs, or too few? Have you considered connecting Vault's standard output to a log filter before writing to disk?
We are having issues with vault & consul, and sometimes they are going down & we don't see much information on the logs to find the root cause. So we have to change our systemd service file to add a log-level command line argument, but as our servers are maintained by chef they get overridden on the next chef run. Then I found log_level configuration on consul and thought it would be nice to provide same option for vault too.
Would it be better to do this via the API (much like CORS) so that you could adjust the log level without having to restart the server?
@naunga - That would be great or a reloadable (not restart) configuration would also works.
Hello - this is a rather old issue at this point, and Vault now offers a log_level configuration flag as well as live configuration reloading with SIGHUP signals, so I'm going to close this. Thanks!
Most helpful comment
Would it be better to do this via the API (much like CORS) so that you could adjust the log level without having to restart the server?