Log Analyser

A simple utility to read, parse and flag the evens from a log file.

To run the app

1. Clone the git repo

git clone https://github.com/chandanv89/log-analyser.git

2. Open the project in an IDE of your choice.
3. Run the main application - com.github.chandanv89.loganalyser.LogAnalyser.

   1. Create/Update the Run Configuration for the class to provide the program argument.
      

      
      A sample log file is available in `resources/samples/logfile.txt`

   2. You can generate a sample log file using the utility com.github.chandanv89.loganalyser.utilities.LogFileGenerator. You can specify the number of events to be generated for the sample file.

Summary of task

Our custom-build server logs different events to a file named logfile.txt. Every event has 2 entries in the file - one entry when the event was started and another when the event was finished. The entries in the file have no specific order (a finish event could occur before a start event for a given id).

Every line in the file is a JSON object containing the following event data:

• id - the unique event identifier
• state - whether the event was started or finished (can have values STARTED or FINISHED)
• timestamp - the timestamp of the event in milliseconds

Application Server logs also have the following additional attributes:

• type - type of log
• host - hostname

Example contents of logfile.txt:

{"id":"scsmbstgra", "state":"STARTED", "type":"APPLICATI ON_LOG", "host":"12345",
"timestamp":1491377495212}
{"id":"scsmbstgrb", "state":"STARTED", "timestamp":1491377495213}
{"id":"scsmbstgrc", "state":"FINISHED", "timestamp":1491377495218}
{"id":"scsmbstgra", "state":"FINISHED", "type":"APPLICATION_LOG", "host":"12345",
"timestamp":1491377495217}
{"id":"scsmbstgrc", "state":"STARTED", "timestamp":1491377495210}
{"id":"scsmbstgrb", "state":"FINISHED", "timestamp":1491377495216}
...

In the example above, the event scsmbstgrb duration is 1491377495216 - 1491377495213 = 3ms. The longest event is scsmbstgrc (1491377495218 - 1491377495210 = 8ms).

The program should:

• Take the path to logfile.txt as an input argument
• Parse the contents of logfile.txt
• Flag any long events that take longer than 4ms
• Write the found event details to file-based HSQLDB (http://hsqldb.org/) in the working folder
  • The application should create a new table if necessary and store the following values:
    • Event id
    • Event duration
    • Type and Host if applicable
    • Alert (true if the event took longer than 4ms, otherwise false)