Comparing Rundeck and Jenkins (And Why They Work Great Together)

Picture of Paul Lambert
Paul Lambert     April 21, 2018 IT Operations, Rundeck

A common question that comes up from time-to-time in the many places where the community discusses Rundeck:

“What’s the difference between Rundeck and Jenkins?”

In short, Rundeck is designed for Operations to work directly with existing tools, services, and systems and to allow secure self-service access to automation for every kind of critical business workflow. Jenkins is a continuous integration server designed for the primary purpose of building artifacts.


Differences in more detail…

Rundeck is built specifically to turn any operations procedure into a repeatable and secure service that can be accessed securely via a Web GUI or API. Because it is not focused specifically on building software artifacts, Rundeck jobs can make real changes, update network and infrastructure configuration, and simplify all types of business workflows.

Rundeck use doesn't need to be limited to developers or expert staff because it has been designed from the start to allow Operations teams to offer self-service access to anyone in the company while meeting and even exceeding established security and compliance requirements. Fine-grained access controls ensure that sensitive credentials, such as firewall passwords, can be restricted to qualified and trusted staff, while very limited use cases (such as a particular set of firewall changes for a particular product) can be made available as a self-service job to a larger audience. This allows Operations to transition from a bottleneck to the critical provider of the tools other teams use to accomplish their work uninterrupted.

Of Rundeck’s many features that support that operations-centric purposes, the one that stands out as the most unique from Jenkins is Rundeck’s inherent knowledge of the detail of your environments (nodes, services, etc.)




Why is it so powerful that Rundeck knows the details of your environments?

A key concept in Rundeck’s design is the idea of a Node. Nodes let you describe your infrastructure and environments, giving you a view of hosts and services.

This is important because most operations tasks today need to be executed across any number of servers from single digits to thousands. Working in a dynamic, distributed environment demands a tool that knows the details of that environment to execute actions on the right subset of servers at that moment, understands the results of those actions, and takes further action based upon that understanding.

Rundeck will combine the data from multiple sources (such as Amazon EC2, Puppet, an Ansible inventory, and Device42) to allow a single view of all the information available, without requiring the maintainer of each source to adopt yet another "new standard."

Rundeck offers out-of-the-box functionality that is difficult to accomplish in Jenkins

Rundeck can execute steps across any remote nodes/servers

Because Rundeck supports many types of workflows the moment it's installed, beyond those offered by a build tool, you can execute workflows across any remote nodes/servers. There is no need to install an agent on the nodes, or for a “Rundeck team” to build and own them. In Rundeck, you define both the workflow to be executed as well as where you want the execution to occur. Filters can be used to dynamically choose servers where Rundeck executes the job based on attributes like environment, data center, operating system, application installed, and other meaningful differences. Jobs can be dispatched via the default SSH (unix/linux) or WinRM (Windows) executors, Rundeck Pro's PowerShell executor, or one of many plugins for other executors such as Ansible, SaltStack, Puppet (Bolt or MCollective), Chef (Knife), etc.

Rundeck helps you to better understand the job output coming back from remote servers. Not only will Rundeck collect the raw output from each step in a job, it organizes all of that output and presents it back to the user with full context about where, when, and why it happened. Each step's status and output are clearly visible with ANSI 256-color support, and you can format structured information like tables and JSON to be easily read and compared. Toggle easily between a workflow view or a log view collated by node and step. Even decorate the output with colorful highlights and unmistakable callouts which make it obvious even to a new employee how to find the log messages which matter the most.

These are immense benefits to both the person designing a job and others in your organization who need to quickly understand the job output. Many users achieve great value from Rundeck in its role as an information radiator, which can often be as game-changing as Rundeck’s utility as a workflow tool.

Rundeck enables you to execute, track, and audit ad-hoc commands in addition to saved jobs.

In a troubleshooting situation, whether it's an emergency or you're being proactive, you will often need to execute commands and scripts in an exploratory fashion. If you have been granted appropriate permissions via Rundeck's comprehensive access control lists (ACLs), you can use Rundeck’s ad-hoc commands feature to execute actions where and when necessary. These ad-hoc commands are logged and tracked just like a Rundeck job, allowing easy collaboration, and answering questions like "what did netstat look like 10 minutes ago?" If you like, you can later save these ad-hoc commands as jobs for future use, to share with others in your organization. In this scenario, Rundeck provides a similar capability that you would otherwise need a network shell tool to achieve.

Rundeck makes it easy to define and execute multiple types of workflows.

By default, you can choose the execution strategy you want a job’s workflow to follow in order to match the real-world operations challenge you are trying to solve. You can have the workflow be step-oriented (execute a step on all of the nodes selected before continuing to next step) or node-oriented (execute all steps on a nodes before moving on to the next node). You can also decide if you want steps to run in parallel and set thread count limits. With Rundeck Pro, you can also use advanced workflows to have both sequential and parallel steps in the same workflow, and to conditionally execute steps based on job settings and the output of previous steps.

Rundeck jobs have built-in error handling features.

When you define your job’s workflow in Rundeck, you can take advantage of built-in error handling that lets you gracefully handle the various types of failures that a successful Operations team must be prepared for. When a job fails on some nodes, Rundeck gives you the option to re-run it only on the failed nodes.

Rundeck logs all activity and sends out notifications.

The usefulness of Rundeck’s built-in ability to add context to a job’s output (not just what ran, but where and what the status was) doesn’t end with just the person actively running or watching the job execution. Both logging and notifications are plugin points in Rundeck. This means you can send all activity information to whatever logging and notification tools your organizations uses (e.g. Jabber, IRC, HipChat, Splunk, Log4J, etc). Rundeck’s contextualized output is useful for auditing purposes.

Rundeck gives you access control policies that know about your environments.

Rundeck gives you the expected capability of being able to specify who can edit or execute what jobs and when. However, because of Rundeck’s resource model's in-depth understanding of your nodes, you can also use Rundeck’s fine grained ACL policies to create rules controlling access to nodes and actions based on roles. With its LDAP integration, Rundeck can manage login access via directories like Active Directory or OpenLDAP.

How are Rundeck and Jenkins often used together?

Development and Operations are equally important parts of a technology organization. If you want to have high quality throughout in your application lifecycle, you need tools that do their specific tasks well and work well together.

Jenkins and Rundeck integrate well and are commonly used by organizations who are looking to achieve DevOps-style results as a part of the same delivery toolchain that spans from Development to Operations. Jenkins will manage software builds and Rundeck will manage any and all operations tasks.

Jenkins can trigger Rundeck to do deployments in a Continuous Deployment scenario, which allows direct Jenkins access to production environments to be restricted. The Rundeck Jenkins plugin plays a key role in this. The diagram below shows a high-level view of a common integration scenario for Rundeck and Jenkins.




In Conclusion

Jenkins and Rundeck are often used together by high performing Development and Operations teams. Using both Jenkins and Rundeck together connects repeatable build, test, and integration processes with infrastructure and production operations. With this integration, you will be able to meet security and compliance obligations while reducing the toil that impacts your ability to deliver meaningful value to the business.

Paul Lambert

Paul is Platform Solutions Architect at Rundeck and wants operations professionals to get the most out of their tools.

Related Posts in Topic