I’m Lawrence lawrencejones, a Product Engineer working at incident.io, building an incident response tool that can help respond and learn from incidents.
Before incident I worked as a Principal Engineer at GoCardless, where I worked on a number of projects within both the SRE space and the product.
This blog is about sharing what I’ve learned so other people can benefit, and getting better at doing that sharing. The themes will be drawn from my work, so expect technical subjects like Postgres and Kubernetes and broader themes like incident response or career development.
If you enjoy what I write and want to hear more, follow me @lawrjones where I’ll post any new content. Similarly, if you have any questions about my writing then don’t hesistate to message- I love hearing other people’s perspective and am always up for a (remote or otherwise!) coffee.
For an idea of what I’ve worked on, here’s a list of projects that I’ve worked on which I’ve been able to open-source.
These include individual projects, personal efforts that have been adopted by my company, and work that was done purely within my capacity as an employee.
pgsink is a Postgres change-capture device that supports high-throughput and low-latency capture to a variety of sinks.
Simplicity: by default, everything will just work
Performance: even extremely high volume or large Postgres databases should be streamable, without impact to existing database work
Durability: no update should be lost
Not yet complete, but on a slow and steady march to v1. I expect other tools to speed past me before I ever get there, but I’ll only make this public when it’s really worth considering using.
Watch this space!
stolon-pgbouncer extends a stolon PostgreSQL setup with PgBouncer connection pooling and zero-downtime planned failover of the PostgreSQL primary.
This project contains GoCardless Kubernetes extensions, mostly in the form of operators and admission controller webhooks. The aim of this project is to provide a space to write Kubernetes extensions where:
Doing the right thing is easy; it is difficult to make mistakes!
Each category of Kubernetes extension has a well defined implementation pattern
Writing meaningful tests is easy, with minimal boilerplate
pgsql-cluster-manager extends a standard highly-available Postgres setup (managed by Corosync and Pacemaker) enabling its use in cloud environments where using using floating IPs to denote the primary node is difficult or impossible.
In addition, pgsql-cluster-manager provides the ability to run zero-downtime failover of the Postgres primary with a simple API trigger.
The goal of Diggit is to provide a tool capable of generating insights about code changes in the context of all those that came before them. This tool would be run in the code review process to aid decisions about whether the proposed change will have a positive impact on the system.
Coach improves your controller code by encouraging:
Modularity: No more tangled before_filter’s and interdependent concerns. Build Middleware that does a single job, and does it well.
Guarantees: Work with a simple provide/require interface to guarantee that your middlewares load data in the right order when you first boot your app.
Testability: Test each middleware in isolation, with effortless mocking of test data and natural RSpec matchers.