Gilles Darold


Gilles is a PostgreSQL expert and developer. He is involved in the world of PostgreSQL and Linux for over 25 years and has worked on many different projects. He is also an expert in Oracle to PostgreSQL migrations and the creator of the well known Ora2Pg tool. Gilles started using PostgreSQL in 1995, and since then he has created some well known tools around PostgreSQL like pgBadger for log analysis, pgCluu for performance analysis or pgFormatter. He has also contributed to PostgreSQL and developed various extensions.

Talk Description: pgBadger - Advanced used

pgBadger has been a widely used log analyzer that is also available as part of the PGDG repository. Most of the information used for analyzing a PostgreSQL database is available in log files. This is one of the biggest reasons why pgBadger was developed. It can parse PostgreSQL log files and generate a report that can be used to get the details of errors, vacuum and locks activity, slow running queries that can be used for further optimizations. By default, pgBadger is able to process a PostgreSQL log completely as long as the appropriate format is used. But some log formats are more suitable than others, especially in terms of performance. Similarly, certain options are essential to refine the reports. I would like to take this opportunity to discuss some of the use cases to help you understand how you can effectively use pgBadger for a deeper analysis of PostgreSQL.

I am going to discuss the following topics in this talk :

  • PostgreSQL log formats and the parameters to generate a good pgBadger report.
  • Analyzing a pgbadger report to understand the areas to tune.
  • Optimizing the report generation time using incremental reports.
  • Generating your customized pgbadger reports such as errors or top N slow queries.
  • Using pgBadger for databases hosted on Cloud.