Note: With release of ISL Conference Proxy 4.4.1816.25 - Release Info the logs were redesigned and new settings added offering you a much grater control over which events get logged and which get reported via email.
Logs are separated in the following parts:
- Disable legacy debug logs: Logs were changed to the AKV format and the previous way of logging was disabled
- Report module errors to system email address(s): Set if the error reports should be sent out
- (legacy debug) Log debug counters: Enable legacy logging
- (legacy debug) Log debug transports: Enable legacy logging
- (legacy debug) Log XMLMSGS request/responses: Enable legacy logging
- Max. log file size in MB: Limit the maximum log size
- Max. number of log files: Limit the maximum number of log files
- Create new log file interval (1d 2h 3m 4s): Set log rotation period
- (legacy debug) Log quick statistics (interval in seconds): Enable legacy logging
- Max crash report age in seconds: Limit the maximum crash report age
- Crash report name filter (<Regex>): Specify crash report name to be used by ICP
- Record activity log: Enable activity log. This will log all user and admin actions. e.g. If you have admin activity enabled you can see who opened and changed which setting
- Limit activity log files (in days): Specify how long the activity logs should be kept
Enable system activity log: Specify if activity should be logged
- Enable admin activity log: Specify if activity should be logged
- Enable user activity log: Specify if activity should be logged
- Enable mail log reports: Specify if activity should be logged
Subsystem settings are located underneath the general settings and come in pairs relating to each subsystem of ISL Conference Proxy. The pairs of settings have the following layout:
- Log subsystem [<Subsystem name>] <Event name> severity report level:
- Log subsystem [<Subsystem name>] <Event name> severity mail report level:
Each event in ISL Conference Proxy is assigned a severity level, from 0 to 7, by the server itself. Severity is classified in the following way:
- 0 - Emergency
- 1 - Alert
- 2 - Critical
- 3 - Error
- 4 - Warning
- 5 - Notice
- 6 - Info
- 7 - Debug
With the first setting in the pair (severity report level) you control the minimal severity for which the events get logged to the ISL Conference Proxy log. E.g. if you set the severity to 3 (Error) then all events with a severity of (3,2,1 and 0) will get logged.
Second setting in the pair (severity mail report level) works in the same way, except events get sent to the system email addresses.
By default the minimal severity report level is set to 4 (Warning) and the severity mail report to 2 (Critical).
It is possible to disable the logging for a certain item by setting it to -1 (disabled).
Set how the counter behaves and what it should notify, example bellow:
#Comment starts with '#' char #[destination counter name or email name] #Create counter that is difference of current and previous counter value. #trend=counter_name #Create counter that is difference of current and value from 10 seconds ago. #trend=counter_name #interval=10 #Create counter that is average of value over 10 seconds. #average=counter_name #interval=10 #Create counter that is difference of countes. You can put multiple counters separated by comma. #diff=counter1,counter2,counter3 #Create counter that is sum of countes. You can put multiple counters separated by comma. #sum=counter1,counter2,counter3 #Force this counter to be dumped to log file every 30seconds #dump=30 # #Key notify_text will make email notification when counter gets to certain limit. Notifications checks are performed every 10s. Limit must be more than 1. #notify_text=number of transports seems high (transports=%1%) #notify_limit=1000 #Email will not be sent during silence period from since notification. Silence period is ignored if undefined or 0. This parameter is useful for non-increasing counters. #notify_silent=60 #When limit is reached and notify offset is enabled the new limit will be offset by current value. #notify_offset=true #Instead of silent period you can set hysteresis. The emails will not be sent until the value drops below current limit and hysteresis value. #notify_hysteresis=10 #Reset internal state every notify_reset seconds. #notify_reset=3600 [WEBTOKEN:err_pop] notify_text=we are getting valid but strange webtokens (count=%1%) notify_limit=1 notify_silent=3600 notify_reset=3600 [WEBTOKEN:err_crypto] notify_text=we are getting invalid webtokens (count=%1%) notify_limit=10 notify_silent=3600 notify_reset=3600 [WEBTOKEN:err_version] notify_text=we are getting unknown version (count=%1%) notify_limit=1 notify_silent=3600 notify_reset=3600