GoCardless is a payment company, and they had a long outage in October. This is one quote from the blog post.
As a payments company, we take reliability very seriously. We hope that the transparency in technical write-ups like this reflects that.
As a developer, this adds a lot of trust for me to the company. Now, since I’m not a customer of GoCardless I was still extremely interested in the content of the post. In the article, they are explaining in detail what failed in their structure and what they did to find out the problem (took much longer than expected) and what they improved to make sure it would not happen again.
Spoiler: most of the article is about their database infrastructure, particularly in PostgreSQL, with a quite standard redundant structure. I think this article will be interested to all users of PostgreSQL.