To let you analyse the validated ROA payload (VRP) data set as well as its overall health, Routinator logs an extensive amount of information. The log levels used by syslog are utilised to allow filtering this information for particular use cases.
The log levels represent the following information:
Information related to events that prevent Routinator from continuing to operate at all, as well as all issues related to local configuration even if Routinator will continue to run.
Information about events and data that influences the set of VRPs produced by Routinator. This includes failures to communicate with repository servers, or encountering invalid objects.
Information about events and data that could be considered abnormal but do not influence the set of VRPs produced. For example, when filtering of unsafe VRPs is disabled, the unsafe VRPs are logged with this level.
Information about the internal state of Routinator that may be useful for debugging.
When running interactively logging
information will be printed to standard error by default. You can redirect
logging to syslog using the
--syslog option, or to a file with the
--logfile option. You can influence the amount of information returned
with these options:
Print more information. If given twice, even more information is printed. More specifically, a single
-vincreases the log level from the default of warn to info, specifying it twice increases it to debug.
Detached Server Mode¶
When running Routinator detached in server mode
logging to syslog is implied. Using the
--syslog-facility option you
can specify the syslog facility to use, which is daemon by default. You also
redirect logging output to a file using the
Though almost all settings are available as command line options, you would likely want to configure logging options in the configuration file.
When you run the HTTP service logging information is also available at the /log path. This will produce logging output of the last validation run. The log level matches that set upon start. Note that the output is collected after each validation run and is therefore only available after the initial run has concluded.