Getting the lay of the land
The first thing to understand about the .NET agent (Framework or Core) logs is that there are two kinds of logs: Profiler logs and Agent logs
- Profiler logs get generated every time a compatible and correctly configured .NET process starts up on the system.
- They have the format
- There will be a separate log for each qualifying process ID that starts on your system.
- The profiler checks the process environment and agent configuration to answer questions like “Should this application be monitored?”, “Is the process/environment supported by the agent?”. and “What instrumentation is in place to monitor this application?”
- These logs are generated only after the Profiler decides that the process meets the requirements for monitoring.
- They have the format
- There is one log per process. If the process is restarted new data is appended to the end of the agent log file.
- This log contains information about what happens when the agent actually attempts to monitor your application.
Before we dive in to troubleshooting with the agent logs, lets make sure we have a clean slate and are working with fresh data. So, do the following:
- Remove all current files from the agent logs folder. You can either delete them, or move them elsewhere if you’d like to keep them.
- Restart your application (or IIS if you’re trying to monitor an IIS-hosted application).
- Generate some traffic/activity in your application for at least 1 minute.
- Identify the process ID of your application’s currently running process.
- Check your New Relic account after 5 minutes to see if data is reporting.
What type(s) of logs do you have?
No logs at all - You have an installation issue. Please see the installation troubleshooting guide for more info.
Only Profiler logs - The profiler is recognizing .NET processes starting up on your system, but either making the determination that it should not be monitored, finding no instrumentation present to define HOW to monitor it, or running into a configuration/compatibility error.
Profiler and agent logs - The agent saw the application process start up on the system, determined it should be monitored and bootstrapped the agent to attempt to monitor the application.
Find the profiler log for your application
You should have already identified the process ID the application is running under. Now, find the profiler log that corresponds to that process ID. The log should be named
NewRelic.Profiler.<process ID>.log. We’ll get in to the “meat” of the file in a bit, but for now, just make sure that it exists, since you want to make sure you’re looking at the log that actually corresponds to the process you’re trying to monitor.
If no profiler log exists for the process ID your app is running under, either the app’s environment is not configured to allow the agent to attach to it, or the app itself is not recognized by the agent as a .NET application. In either case, please refer to the links provided in the “Is this post for you?” section at the top of this post.
I’ve got the profiler log, now what?
If you’re only getting profiler logs - Head over to <Common .NET Agent Profiler Log Issues> to troubleshoot the profiler logs.
If you’re getting both profiler and agent logs:
Find the agent log that contains the string “
started (pid <your_process_ID>)”, making sure to replace <your_process_ID> with the actual PID of your process. For example: if your application is running under PID 2531, search your agent files for the string “
started (pid 2531)”.
If no agent log files contain that string, go over to <Common .NET Agent Profiler Log Issues>.
If you found both the profiler and agent logs that correspond to your application, then for now go ahead and reach out to support, or ask your specific question in these forums. I’ll be expanding these posts in the future to include common issues in the agent logs.