Document how to use the validation tools

docs/troubleshooting.md

@@ -0,0 +1,53 @@
+# Troubleshooting issues with binaries
+
+The agent exposes some validation tools to help troubleshoot issues with a binary.
+This allows the user to debug and fix errors that could prevent a job from running.
+
+## Using the validation tools
+Download a zip file containing the agent binaries:
+
+```
+onefuzz tools get <destination_folder>
+```
+
+Extract the zip file in a folder of your choice.
+Navigate to the folder that matches your os.
+Run the following command to see the tools available:
+`onefuzz-agent.exe validate --help`
+The current list of commands are:
+   - run_setup : Run the setup script
+   - validate_libfuzzer:  Validate the libfuzzer target
+   - execution_log: Get the execution logs to debug loading issues
+
+   More tools might be added in the future so please refer the help command to get the most up to date list.
+
+
+## In a docker container
+
+It could also be helpful to run the those command in an environment to closely match the vm where the agent is deployed.
+A docker container can help with that scenario.
+
+Make sure [docker](https://docs.docker.com/desktop/) is installed and runs properly.
+
+Navigate to the folder that matches your os in the tools folder created earlier and build the docker container:
+
+```cmd
+docker build --t <container_name> .
+```
+
+Use the container interactively to execute the validation command:
+
+windows
+
+```
+docker run --it --rm --entrypoint powershell <image_name>
+```
+
+linux
+
+```
+docker run --it --rm --entrypoint bash <image_name>
+```
+
+From there you can navigate to the onefuzz directory and execute the validation commands
+

docs/unmanaged-nodes.md

@@ -75,7 +75,7 @@ Under the `client_credential` section of the agent config file, update `client_i
 
 Save the config to the file.
 
-### Start the agent.
+### Start the agent
 
 Navigate to the folder corresponding to your OS.
 Set the necessary environment variable by running the script `set-env.ps1` (for Windows) or `set-env.sh` (for Linux).
@@ -111,7 +111,7 @@ From here you will be able to schedule jobs on that pool and they will run.
 
 ## Troubleshooting
 
-### increase the verbosity of the logs
+### Increase the verbosity of the logs
 
 It can help when investigating issues to increase the log verbosity. you will need to set the [RUST_LOG](https://docs.rs/env_logger/latest/env_logger/#enabling-logging) environment variable when starting docker
 
@@ -127,7 +127,7 @@ log_level can be any of
 - debug
 - trace
 
-### use the container interactively
+### Use the container interactively
 
 you can use the container interactively by with the following command
 
@@ -143,7 +143,7 @@ linux
 docker run --it --rm --entrypoint bash <image_name>
 ```
 
-### mount a local folder in the container
+### Mount a local folder in the container
 
 docker allows you to [mount](https://docs.docker.com/storage/bind-mounts/#mount-into-a-non-empty-directory-on-the-container) a local folder when running a container
 

src/agent/onefuzz-agent/src/validations.rs

@@ -11,9 +11,9 @@ use uuid::Uuid;
 pub enum ValidationCommand {
     /// Run the setup script
     RunSetup { setup_folder: PathBuf },
-    /// Validate the libfuzzer target
+    /// Validate the libfuzzer target by attempting to run the target by itself and with some of the supplied seeds if provided
     ValidateLibfuzzer(ValidationConfig),
-    /// Get the execution logs to debug loading issues
+    /// Get the execution logs to debug dll loading issues
     ExecutionLog(ValidationConfig),
 }
 
@@ -30,8 +30,6 @@ where
         .find('=')
         .ok_or_else(|| format!("invalid KEY=value: no `=` found in `{s}`"))?;
 
-    println!("******** pos: {}", pos);
-
     Ok((s[..pos].parse()?, s[pos + 1..].parse()?))
 }