For better or worse, most companies—including their execs and developers—see SREs as superheroes who’ll save them from the evils of downtime and service degradation with their boundless superpowers.
SREs are expected to constantly perform dangerous stunts like production debugging or communicating highly technical issues to angry VPs. They must also be able to manage infrastructure, networks, databases, pipelines, operating systems and much more.
Because there’s so much to work with, it can be a challenge to make sense of all the terms you come across with as you navigate through your career as an SRE. That’s why we’re putting together a list of the common tools and what to look out for in each of them.
We’ve separated the tools in concentric circles from closest to you as an SRE to further, but still relevant to your function.
Incident Response Tools
A big part of being an SRE is dealing with incidents. Whether they occur discreetly during work hours, or as a big scandal during rush hour on a weekend. The tools you use to digest alerts and coordinate a response will be a constant in your.
Rootly
Rootly is a modern on-call and incident response tool, used by teams like LinkedIn, Cisco, Canva and Elastic.
Key features:
- A single, streamlined, solution to manage incidents from alert to retrospective.
- Compatible with most observability solutions
- Flexible On-call schedules that let responders request coverage when needed.
- Beautiful mobile app that provides more context after paging.
Best suited for:
- Startup and medium sized companies getting started with their SRE practice
- Enterprise customers with advanced security needs and flexibility requirements.
PagerDuty
PagerDuty is the most popular legacy on-call management tool, founded in 2009. For a few years, PagerDuty was the only reliable alerting solution that could provide enterprise customers.
However, SRE managers often report their frustration on the amount of manual work they put in to configure and manage PagerDuty due to its assumptions on how software is shipped. Its high costs and aggressive upselling tactics make customers question if they’re getting the value of their investment back.
Key features:
- Compatible with most observability solutions
- On-call schedules
- Mobile app with prioritized push notifications for alerts
Best suited for:
- Large enterprise teams with monolithic practices
OpsGenie
OpsGenie is another legacy on-call management tool. Owned by Atlassian, OpsGenie is more often found in organizations using other products from the company, like Jira or Confluence. However, OpsGenie has been known for its multiple and constant outages. OpsGenie is also often critized by its customers for the lack of investments made by Atlassian to improve the product over the years.
Key features:
- Compatible with most observability solutions
- Basic on-call schedules
- Native integrations with Atlassian products
Best suited for:
- Organizations who already use Atlassian’s software and can negotiate a good deal to add OpsGenie to the bundle. That is given the team is comfortable with downtime in their alerting solution.
{{cta-checklist}}
Monitoring & Observability Tools
You can only respond and resolve an incident if you detect it. Monitoring and observability tools constantly look at data coming from your system to understand if everything is working as expected. If they detect an anomaly, they’ll trigger an alert.
Prometheus
Prometheus is a popular open-source system monitoring suite. Prometheus is known for its scalability, reliability, and strong community support (54k stars on GitHub).
Key features:
- Self-hosted. so it’ll take more commitment from your team to use it.
- CNCF project, which means it’s often used in Kubernetes and Cloud native environments.
- Very versatile thanks to its multidimensional data model.
- Collects time series from instrumented jobs, mainly through HTTP endpoints that expose metrics.
- A powerful query language, PromQL, to search and aggregate time series data.
- Prometheus is typically used with Grafana for visualizations.
Best suited for:
- Teams with a mature system and bandwidth to implement Prometheus. The learning curve can be steep, so it’s more popular among experienced SREs.
Datadog
Datadog one of the most used observability vendors. It offers a wide set of products aimed at getting visibility of the performance and health of your applications, infrastructure and environments.
Key features:
- Fully managed SaaS
- Offers support for Kubernetes and other deployment strategies.
- All-in-one platform for observability: from metrics collection and querying to alerting and synthetic monitoring.
- Comes out of the box with dozens of integrations with a wide range of tools.
- Subscription-based pricing, which often grows rapidly, as pointed out by a lot of users.
Best suited for:
- Teams who prefer to use a managed observability solution instead of maintaining it themselves.
Container Orchestration Tools
Most teams these days rely on containers to deploy their software because they make composability easier. However, you’ll end up having dozen of containers flying around when you account for every component of your system. Thus, you need a platform to help you automate, manage, scale, and network containers.
Even though Container Orchestrators usually offer some kind of ‘auto-heal’ feature, as an SRE, you’ll rapidly get accustomed to having to ‘heal’ things yourself.
Kubernetes
Kubernetes is the most widely used container orchestration platform. It’s the founding project of the Cloud Native Computing Foundation (CNCF), which means being open-source is at its core.
Key features:
- Can manage automated deployments at a large scale, but can be difficult to learn and set up properly.
- Monitors the health of nodes and containers and can automatically restart failed ones.
- Can support dynamic storage provisioning
- Very vibrant ecosystem
Best suited for:
- Large and complex deployment setups. Best suited for enterprise teams with multi-cloud strategies, microservices architectures, and sophisticated CI/CD pipelines.
Docker Swarm
Before Kubernetes, Docker was pretty much the default way of deploying containers. The Docker ecosystem isn’t fully open source, but that brings the benefits of a more polished product that is easier to setup.
Key features:
- Simpler setup, which results in a relatively straightforward learning curve.
- A seamless experience from development to deployment through various Docker products
- Supports rolling updates
Best suited for:
- Small and medium companies who don’t need very complex deployments.
CI/CD Tools
Continuous Integration and Continuous Delivery are pretty much the standard way of developing and shipping software these days. CI/CD pipelines execute a set of steps to build and run software within specified parameters and environments.
Yes, you’ll likely become besties with a wide range of CI/CD pipelines as you try to figure out what or why something went wrong at 3am when you tried to rerun the workflows to bring a service back online.
Jenkins
Infamous and feared, Jenkins has been around since 2011. As a robust and highly-customizable open-source solution to build pipelines, Jenkins dominated the market for a few years. However, configuring it and its plugins can rapidly become a challenge for a lot of teams.
Key features:
- You can set up pipelines as code using their DSL (Domain Specific Language) called Groovy.
- Over 1500 plugins that can be mixed and matched. The caveat is that plugins don’t have widespread standards so their way of working varies widely.
- Can be used for CI/CD, but also automated testing and other types of pipelines.
Best suited for:
- Teams with very complex CI/CD needs and with enough resources to maintain a Jenkins solution.
CircleCI
CircleCI got into the market around the same time as Jenkins, but it’s more of an open-core solution. Their main offering is a fully-featured Cloud hosted version, but a self-managed free to use version is also available.
Key features:
- Provides a nicer UI and lets you use YAML to describe configurations and pipelines.
- It’s optimized for speed, allowing parallel execution and caching to reduce build times.
- Comes with built-in “Orbs” which cover common use cases for CI/CD pipelines.
Best suited for:
- Small to medium teams that need a quick and easy CI/CD pipeline setup.
GitHub Actions
GitHub shook the CI/CD world with their release of GitHub Actions in 2019. The clear caveat is that it only works for GitHub customers. But the ergonomics built into GitHub Actions make it the easiest-to-use CI/CD solution in the market at the moment.
Key features:
- Simple YAML configuration to set up pipelines taking advantage of native GitHub events like commits or pull requests.
- A marketplace to get pre-built actions and workflows for common tasks, tools, or vendors.
- Matrix builds to run workflows across multiple environments and configurations simultaneously.
Best suited for:
- Teams already using GitHub
Configuration Management Tools
A small misconfiguration can cause major outages, so most organizations opt to manage their configurations through a dedicated tool. When you have multiple environments and systems, you definitely need some form of automation around your configs.
As an SRE, you won’t be stranger to the configuration management tool your team uses. Whether if it’s fishing for that problematic provisioning or figuring if the updates were applied uniformly across services and environments.
Ansible
Ansible is an open source automation engine that helps teams set up configuration management, among other processes. It’s commercialized by RedHad through a broader offer platform offering.
Key features:
- Ansible uses SSH for communication with the target systems instead of having to install agents on each of them.
- Lets you define playbooks with configurations using YAML
- Guarantees that the result of running a playbook is always the same.
Best suited for:
- Ideal teams for a simple configuration management tool.
Puppet
Released in 2005, Puppet is still relied on by many organizations due to its maturity and extensive ecosystem. Similar to Ansible, it offers an open-source core and a product around it, as well as services and training.
Key features:
- Lets you define system configurations using a declarative language.
- Employs an agent model to guarantee communication and configuration enforcement across different types of systems.
- Features a large collection of pre-built modules for common use cases, tools, and vendors
Best suited for:
- Large enterprises with complex requirements.
- Puppet is also great for organizations operating in highly-regulated markets due to its advanced compliance and auditing capabilities.
Application Performance Monitoring Tools
A full service unavailability is not the only kind of incident you’ll deal with as an SRE. You’ll likely also want to address any performance degradation in any of your components. And for that you need, yes, more tools.
Application Performance Monitoring tools provide real-time data on response times, error rates, and transaction throughput.
New Relic
New Relic offers full-stack observability capabilities, including application performance, user experience, and infrastructure health.
Key features:
- Monitors backend and frontend performance
- It can help identify bottlenecks by tracking and visualizing how requests are coming to the app
- New Relic recently added AI and machine learning to their systems to detect anomalies and predict potential issues.
Best suited for:
- Even though New Relic mainly serves enterprise customers, it can also be used at startups and SMBs without much hassle.
Dynatrace
Launched in 2005, Dynatrace is one of the most established application performance monitoring solution. The offer a large suite of products to cover several use cases within the observability space.
Key features:
- Tracks how users are interacting with the application
- It can automatically discover dependencies within the application
- Can be set up to provide code-level visibility and find root causes to the issues it discovers.
Best suited for:
- Dynatrace can be best leveraged by larger enterprises with complex architectures.
Rootly integrates with tools you’re already using
SREs rely on a wide range of tools to keep systems running smoothly. Rootly integrates with all the tools you already use so your incident response experience is as streamlined as possible.
Book a demo with one of our reliability experts to see how Rootly can help your reduce your MTTR.
{{cta-demo}}