Perfect balance between low-level close-to-the-metal container-based pipelines and high-level super-user-friendly UI, GitOps support, pipeline debugging with breakpoints, K8s/Docker/Helm as first-class citizens, built-in secrets, a multitude of integrations, great doc and support material/videos/webinars, nice people... (well, there's so much more, do you really want me to keep going?!? ;) Review collected by and hosted on G2.com.

Set of commands that can be invoked while troubleshooting a pipeline in debug mode is very limited (cd, cat, ls...) Review collected by and hosted on G2.com.

Lots of OOB Integrtations with multiple 3rd party tools and services Review collected by and hosted on G2.com.

Better documentation and more instructions/tutorials would be great Review collected by and hosted on G2.com.

After searching and evaluating a number of different platform for CI/CD we discovered Codefresh to be the most cost effective, easy to use and customisable for all our workflows. With a direct Kubernetes API integration, we can control each cluster directly from the Codefresh dashboard. This has provided us with a faster feedback on making some quick changes like increasing replicas. The Helm Dashboard that allows for easy Helm control such as visibility of which revision is deployed to managing rollbacks makes Codefresh a far superior platform to work with.

When running the workflows it is important that we have an easy and flexible system to manage environment variables. Codefresh not only allow easy pipeline editing the variables but also the ability to share variables among projects and pipelines.

Some of the best features are -

- Multiple Environment Workflows

- Shared Environment Variables

- Exportable Variables

- Customisable Steps

- Kubernetes API integration

- Unified Log Output

- Easy rollbacks with Helm dashboard Review collected by and hosted on G2.com.

It would be nice to be able to split the users into teams with better granular RBAC settings. At present all users have access to all projects. Having user groups attached to projects would give more security control to team leaders.

At first we used Codefresh for both CI & CD but we discovered that Github actions was was much faster at building, running and testing our code than Codefresh was so we recently opted to use GitHub actions for all CI and just use Codefresh for CD. Review collected by and hosted on G2.com.

I like that once my build is setup it just works. It's easy to add custom configurations. Review collected by and hosted on G2.com.

It could be a little faster, but the tests take up most of the time. I could speed things up if I spend some time optimizing tests. Review collected by and hosted on G2.com.

Codefresh doesn't have a single way of setting up the CI. You can set it up and get it running how you want, you can run steps in parallel, or you can run them synchronously, or you can do both. You can trigger other pipelines from the current pipeline and act on their success or failure status. Review collected by and hosted on G2.com.

The only self-hosted runner option requires kubernetes. Review collected by and hosted on G2.com.

Codefresh makes it insanely easy to integrate my CI/CD pipeline with my K8s infrastructure, without requiring that my K8s infra be married to the CI/CD pipeline or owned by it.

The pipeline debugging feature, as the title indicates, is a HUGE game changer. Too many pipelines end up as just a complicated mess of abstractions over the actual logic that is being executed at a given moment, but with the ability to enter a debugging state before, during (for freestyle steps, at least), or after a step has executed allows you to make certain that what you think is happening is actually what is happening.

The variety of nearly flawless Codefresh customized containers that they provide as steps you can easily implement in your pipeline while safely abstracting away implementation details and maintaining security and integrity within the pipeline makes my job of implementing a pipeline for the various projects within our company so much easier. Review collected by and hosted on G2.com.

The default implementation of the Codefresh deployments does require your pipeline to be responsible for the app from initial deployment onward in order to work. However, with the abundance of other step types they provide and crafty use of freestyle steps, you can easily have your own deployment requirements met if the Codefresh defaults don't work for you, for whatever reason. Review collected by and hosted on G2.com.

Managing Kubernetes is very complicated and requires someone with a highly specialized skillset, one that's only just now starting to become more popular. Being a software engineer and business owner, I don't have time to manage my daily workload and stay up to date with whatever is happening in the Kubernetes world but at the same time I can't afford dedicated staff to handle that. Between Google Cloud Platform and CodeFresh, I can get by with my limited knowledge very easily and I rarely ever have to get into the Kubernetes plumbing which makes CodeFresh worth the price. Not to mention the insane amount of time it saves me automating my deployments and notifying clients via Slack of their tasks being delivered to production or being ready for review, so I don't have to write a ton of emails. We've just established a simple system so they know when to start looking. Review collected by and hosted on G2.com.

Had some hiccups with support not answering a few of my inquiries at first, but my recent couple experiences have been great and they even personally reached out to learn about the problem when I was a bit blunt about the lack of a response in further support inquiries. Everything was resolved, so really there's nothing I dislike about CodeFresh at the moment. Review collected by and hosted on G2.com.

The UI gives more context to my clusters and pods and the product gives seemeless integration for auto deployment, and notifies me of failures. So, its a complete poackage and I highly recomend the product. Review collected by and hosted on G2.com.

The loading times can be improved, and I have faced some down times at late night deployments. Review collected by and hosted on G2.com.

Docker based model which avoids the pitfalls of using Jenkins with conflicting plugins. YAML configuration is modern, readable, and easily accepted by different software development groups. Hybrid installation for on-prem runners and SaaS hosted UI allowed us to keep our data internal while getting updates and product improvements with no upgrades on our part. Docker containers with each step gives granular control of software version for the task. Integrations with Kubernetes allow for complete CI/CD with native steps and less scripting. Really liked the CLI capabilities in addition to intuitive user interface. Support has been amazingly responsive and is invested in our success. Review collected by and hosted on G2.com.

Not much to dislike. It is a growing product in the landscape of CI/CD so it not the first product to come to mind for an evaluation. With some research, it quickly moves to the top of a product comparison list with more established tools. Review collected by and hosted on G2.com.

Lots of freedom to define steps in a pipeline, easy to create alerts, good integrations (github, slack). Can reuse pipeline definitions through YAML specs. Review collected by and hosted on G2.com.

Occasional failures can interrupt vital CI/CD pipelines. Sometimes steps fail without a clear cause, restarting them seems to work but false negatives can cause distraction. Review collected by and hosted on G2.com.