Blog

Parsing windows IIS logs

March 10, 2024 by Prabhat Sharma
OpenObserve

Introduction

Internet Information Services (IIS) is a flexible, secure and manageable Web server for hosting anything on the Web. Monitoring and analyzing IIS logs can provide valuable insights into the performance and health of your web applications. In this blog, we will walk through the process of capturing IIS logs using the OpenTelemetry Collector (otel-collector) and parsing them using VRL functions in OpenObserve for analysis. You could use fluentbit or vector as well to capture the logs and send them to OpenObserve.

Step 1: Setting Up the otel-collector

The OpenTelemetry Collector offers a vendor-agnostic implementation on how to receive, process, and export telemetry data. In our case, we want to capture IIS logs and send them for analysis. Here's how you can set it up:

  1. Install the otel-collector: You can install the otel-collector on your server to capture the IIS logs using the one liner command available from your OpenObserve Ingestion page.

otel-collector

This will install the otel-collector as a Windows service with the binary and the configuration file in c:\otel-collector

  1. Configure the filelog receiver: The otel-collector configuration file should include a filelog receiver. This receiver will watch for log files in the specified directory and read new log entries as they are written.

Open the configuration file c:\otel-collector\otel-config.yaml and add the following configuration snippet:

receivers:
  filelog/iis:
    include: ['c:\\inetpub\\logs\\LogFiles\\**\\*']
    start_at: beginning
  1. Configure the OpenTelemetry exporter: The otel-collector configuration file should also include an OpenTelemetry exporter to send the captured logs to OpenObserve.
exporters:
  otlphttp/openobserve_iis:
    endpoint: http://localhost:5080/api/default/
    headers:
      stream-name: iis
      Authorization: "Basic cm9vdEBleGFtcGxlLmNvbTpXbTNtdTFaS2Vma3pCa1p1"
  1. Wire them together: Ensure that the service pipelines are correctly wired to use the filelog/iis receiver and the appropriate exporter.

Example configuration snippet for the service:

service:
  pipelines:
    logs/iis:
      receivers: [filelog/iis]
      processors: [resourcedetection/system, memory_limiter, batch]
      exporters: [otlphttp/openobserve_iis]

At this point, the otel-collector is set up to capture IIS logs and send them to OpenObserve for analysis. For the changes to take effect, restart the otel-collector service.

otel-collector restart

Once the otel-collector is running, it will start capturing IIS logs and sending them to OpenObserve. You should see a new stream iis under streams/logs menu

OpenObserve with iis stream

Step 2: Parsing IIS Logs with VRL

After setting up the otel-collector to capture the IIS logs, the next step is to parse these logs into a structured format that can be easily analyzed. VRL functions can be created in OpenObserve to parse the logs.

The IIS logs are typically written in a space-separated format.

Here is a sample log entry from an IIS log file:

{
  "_timestamp": 1710067224558454,
  "body": "2024-03-10 10:39:35 ::1 GET / - 80 - ::1 Mozilla/5.0+(Windows+NT+10.0;+Win64;+x64)+AppleWebKit/537.36+(KHTML,+like+Gecko)+Chrome/121.0.0.0+Safari/537.36+Edg/121.0.0.0 - 304 0 0 0",
  "dropped_attributes_count": 0,
  "host_name": "EC2AMAZ-FI57MQI",
  "log_file_name": "u_ex240310.log",
  "os_type": "windows",
  "severity": ""
}

The body contains detailed information about a single HTTP request in a common log format. Let's break down the components:

  • 2024-03-10 10:39:35: The date and time when the request was received.
  • ::1: The source IP address of the request, which in this case is the loopback address (IPv6 format), indicating the request was made locally on the server.
  • GET: The HTTP method used for the request.
  • /: The path of the requested resource, which in this case is the root directory.
  • 80: The port number on which the request was received.
  • -: This dash often represents the referrer URL, which is missing here, indicating the request was not made through a link on another page.
  • ::1: Again, the source IP address, also indicating a local request.
  • Mozilla/5.0 ... Edg/121.0.0.0: The user agent string, which provides information about the client making the request, including the operating system, browser version, and rendering engine.
  • 304: The HTTP status code sent in response, indicating that the resource was not modified and the client can use the cached version.
  • 0: Substatus code (specific to the server software) which sometimes provides more detail on the response.
  • 0: Win32 status code, which can provide information about the status of the request at the operating system level.
  • 0: The time taken to serve the request, in milliseconds.

This structured logging is useful for automated processing and analysis, allowing systems and administrators to filter, search, and analyze log data efficiently.

Here's an example of how you can parse the logs using VRL:

pattern = r'^(?P<timestamp>[^\s]+ [^\s]+) (?P<client_ip>[^\s]+) (?P<http_method>[^\s]+) (?P<requested_path>[^\s]+) (?P<placeholder1>[^\s]+) (?P<server_port>[^\s]+) (?P<placeholder2>[^\s]+) (?P<referrer_ip>[^\s]+) (?P<user_agent>.*?) (?P<referrer_url>.*?) (?P<http_status>[^\s]+) (?P<other_codes>[^\s]+ [^\s]+) (?P<response_time>[^\s]+)(?: (?P<forwarded_ips>.*))?$'

.body = parse_regex!(.body, pattern)

Step 3: Viewing Parsed Logs in OpenObserve

Once the logs are parsed, they should look like below:

{
  "_timestamp": 1710067224558454,
  "body_client_ip": "::1",
  "body_forwarded_ips": null,
  "body_http_method": "GET",
  "body_http_status": "304",
  "body_other_codes": "0 0",
  "body_placeholder1": "-",
  "body_placeholder2": "-",
  "body_referrer_ip": "::1",
  "body_referrer_url": "-",
  "body_requested_path": "/",
  "body_response_time": "0",
  "body_server_port": "80",
  "body_timestamp": "2024-03-10 10:39:34",
  "body_user_agent": "Mozilla/5.0+(Windows+NT+10.0;+Win64;+x64)+AppleWebKit/537.36+(KHTML,+like+Gecko)+Chrome/121.0.0.0+Safari/537.36+Edg/121.0.0.0",
  "dropped_attributes_count": 0,
  "host_name": "EC2AMAZ-FI57MQI",
  "log_file_name": "u_ex240310.log",
  "os_type": "windows",
  "severity": ""
}

OpenObserve with parsed IIS logs

Step 4: Save the function and associate it to the stream

You should save the VRL function and associate it to the stream. This will ensure that the logs are parsed and structured as they are ingested into OpenObserve.

OpenObserve with parsed IIS logs

Associate function to the stream

Once you have associated the function with the stream, the logs will be parsed and structured as they are ingested into OpenObserve.

Final result of parsed IIS logs

Step 5: Analyzing the Data in OpenObserve

With the logs now in OpenObserve, you can use dashboards and alerts to:

  • Monitor access patterns and potential security breaches.
  • Analyze the performance of your web applications.
  • Set up alerts for specific log patterns that might indicate errors or other issues.

Conclusion

By leveraging the power of otel-collector and VRL, you can efficiently capture and send IIS logs to OpenObserve and parse them for comprehensive analysis. This approach allows you to gain deeper insights into your web applications, helping you to maintain performance and reliability. Remember to check the OpenTelemetry and Vector documentation for the most up-to-date instructions and configurations.

Author:

authorImage

Prabhat Sharma is the founder of OpenObserve, bringing extensive expertise in cloud computing, Kubernetes, and observability. His interests also encompass machine learning, liberal arts, economics, and systems architecture. Outside of work, Prabhat enjoys spending quality time playing with his children.

OpenObserve Inc. © 2024