Debugging the kernel locally
Locally debugging the kernel allows you to diagnose the kernel directly on the host computer. Debugging information from different kernel components are saved to a kernel debugging file, kernelDbg-#.log, where it is numbered depending on how many times the kernel is run.
The structure of the local debugging log is like the below picture:
The local debug logs contain two versions of formatting:
Each of these fields have their own values, as follows:
date: The date of the event
time: The time of the event
level: One character error level, which is one of:
T: Trace verbose message
D: Debug verbose message
I: Informational message
W: Warning message
E: Error message
F: Fatal error message
method: The method name in which the message was posted
source: The source code file where the method is found
linenum: The line number from the source file
message: The message
Each of these fields have their own values, as follows:
date: The date of the event
time: The time of the event
level: One character error level, which is one of:
T: Trace verbose message
D: Debug verbose message
I: Informational message
W: Warning message
E: Error message
F: Fatal error message
fully-qualified-method: The full method name in which the message was posted
message: The message
To debug your mods, they must call the debug functions in order for the kernel to acknowledge your message. There are useful functions listed below that may help you debug your routines in your mods.
Calling the debug function below will post your debug message to the kernel debugger normally. There's a function for you to call below:
Found in the DebugWriter module under the Nitrocid.Kernel.Debugging namespace.
Calling the debug function below will post your debug message to the kernel debugger if the condition that you've set within the function is satisfied. There's a function for you to call below:
Found in the DebugWriter module under the Nitrocid.Kernel.Debugging namespace.
Calling the debug function below will post your debug message to the kernel debugger normally. However, it also filters every variable you've selected to be censored in the debug log. For example, if you provide two variables (A, B) and B contains sensitive info, you may want to create an array of indexes which holds B's index (in this case, 1) when calling the below function.
Found in the DebugWriter module under the Nitrocid.Kernel.Debugging namespace.
Calling the debug function below will post the stack trace of an exception, including its inner exceptions, to the kernel debugger. There's a function for you to call below:
Found in the DebugWriter module under the Nitrocid.Kernel.Debugging namespace.
Calling the debug function below will post the stack trace of an exception, including its inner exceptions, to the kernel debugger if the condition that you've set within the function is satisfied. There's a function for you to call below:
Found in the DebugWriter module under the Nitrocid.Kernel.Debugging namespace.
If you want to rotate logs each nth message, such as the 10000th message, you can enable this feature by going to the kernel settings and enabling it there. You can also set the number of messages before the log rotation occurs.
If this session's debug quota is exceeded, the new debug file is created with the increasing number. For example, if kernelDbg-0.log exceeds 10000 messages, the debugger creates the kernelDbg-1.log and writes further messages there, and so on.
You'll need to restart the kernel to make use of this feature.
