Microservices

Caching in general is a good practice to improve the responsiveness of the system. But sometimes it can go horribly worng and cause cascading failures leading to bringing the entire system down. Below were such incidents that I encountered: Backend service caches M2M auth tokens which it gets by making a API calls to an Authservice. Auth service introduced a bug in which it returns empty token and 200 status code. The backend service checks the response code and cache the token for few hours and the rest is history :) Service caches DNS entry to make API call for the resource. DR hits and the cached DNS entry no longer works. In the absence of dynamic service discovery, needs entire cluster restart. Caching a /GET call and not respecting the expiration. Well this is just silly and should have been caught in the code review Happy caching!

Companies have a never ending quest to continuously adopt new tools, processes and new architectures with the goal of enabling teams to move fast and build high performing teams. What is a high performing team? If you ask a product manager, they will say "It’s a team that can deliver new value faster". If you ask a Team lead they will say along the lines of "A high performing team can pay down its tech debt faster, spends majority of the time building new value to the product and optimizes their work for improved quality and faster delivery." "so what makes these teams move fast?" Now this is a great question! I have been fortunate to work with some of these high performing teams.  Juggling down the memory lane, I was trying to find the answer to this question in a way that discounts the personal traits of any individual in the team. I stumbled upon three unique qualities that distinguished a high performing team from others: Ability to make sweeping chang...

Terminology Observability : Observability is the property of a system to answer either trivial or complex questions about itself. How easily you can find answers to your questions is a measure of how good the system’s observability is. Monitoring : observing the quality of system performance over a time duration Instrumentation : refers to metrics exposed by the internals of the system using code (a type of white-box monitoring) Why IOM?  Analyzing long-term trends like User growth over time User time in the system System Performance over time Comparing over time or experiment groups How much faster are my queries after new version of a library Cache hit/miss ratio after adding more nodes Is my site slower than it was last week? Alerting - Something is broken, and somebody needs to fix it right now!  Building dashboards - Answer basic questions about your service - Latency, Traffic & Errors. Conducting ad hoc retrospective analysis - Our latency just shot up; what else ha...

Here is the link to my #Dockerccon2020 talk "Monolithic to Microservices + Docker = SDLC on Steroids!" 

EWF is another microservice decomposition pattern that I like to use when building services. It stands for Entity, Workflow & Functional services. In simplest terms, it’s an extension of Single Responsibility principle and allows for building independent testable services. To understand this pattern better, let’s start with an example. Imagine I want to write a signup service that allows user to signup to a website using their Google account. The service would perform the following:  Let user login into google account and fetch user’s basic information from Google upon successful sign-in. Apply more default settings associated to creating new user.  Persist the user in the database.  After successful registration, redirect the user to the homepage.  At a high level, this works just fine and gets the job done. But if you think about testability first and wants to test this service in isolation we have a problem! The entire testing paradigm relies on ...

Recently I attended an RCA meeting for an incident that happened in prod few weeks back. The issue was somewhat of a 2nd degree failure. To put it in simple terms, lets say a workflow composed of two services - A and B. When user performed an action from the UI, it made a call to Service A which then internally called service B with some parameters based on user preferences. In this case, due to some unusual combination of preferences, Service-A ended making call to Service-B with parameters that Service-B didn’t fully understood. Solution - Write an integration test to cover this scenario. Reasoning - Since this involved 3 different components (User Preferences, Service A and Service B), it should be covered by an integration test. More Reasoning: This is sort of the usual way of thinking about the integration tests - meaning whenever there are multiple services involved, we immediately think about writing an integration test. But is this the right approach specially in microserv...

Microservice architecture paradigm fits well within the Unix’s first principle/philosophy: “Make one program do one thing well”.  We love creating small services which has identity of its own (within a bounded context). These services have their own CI/CD pipelines and provides loose coupling b/w components which significantly improves our ability to deliver code to production. In reality, a workflow can span multiple services. Its very common to find scenarios like below: In the above example, a workflow spans 3 services (Service-1, 2 and 3). Service1 receives the original request. In order to process this request, it GET(s) some data from its dependent services(d1/d2/d3). It further runs some business logic and produces an intermediate response #A. It then makes a state mutable call (POST/PUT/DELETE/PATCH) to Service-B using #A as input and waits for its response. Service-B follows the same paradigm and the workflow ends when all services returns.  The scena...