Changelog intro, Changelog punchline, feature tease, and another joke.

Everyone, welcome, to another predictable Changelog. 

Your cron jobs may be predictable, but this edition is anything but. All new features, with plenty more in the pipe, let’s dive in.

Passengers on Railway really love cron jobs. So much so, that they have lovingly told us that they wanna know when and how they executed, so they can love the fact that it ran.

Alberto, Software Engineer, and Jitachi, one of our Designers, spent this week rethinking the Cron Jobs experience.

First, they poured through all the emails, one by one, wrote thank you notes and then got to work. Now, when you enable Cron Jobs via the setting page UI, the service experience will use the dedicated Cron Job UI.

When you click into a Cron Job service, the Deployments pane will now navigate you to the new Schedule pane. Here, we added a few pieces of UI to make the cron experience that much more tractable.

You will now see an attached tooltip to the service indicating the cron schedule.

You will now see a banner showing the cron schedule, with the option to edit it.

You will see successful and failed prior executions, as well as the incoming executions.

It’s never been easier for your workloads to stay on schedule.

, we shipped Dockerhub and GHCR deploys to Priority Boarding. We then proceeded to take every hammer to every nail. After ensuring the experience was… nailed-down, we then hit the General Availability button harder than usual at Railway HQ. 

As a reminder, you can deploy images via the CMD+K menu and going to: New Service. Then you can slam in a public Dockerhub or GHCR url harder than usual. (Private registries coming soon)

Considering that the feature was only available to Beta users, you all managed to deploy registry images 9000+ times. Commence hockey stick adoption. Go!

We already hear you yelling: “Docker Compose when?” To which we reply: no comment. 

 (Joke explainer: this links to Canny, it’s planned.)

We’re getting serious about the debug and monitor flow for running deployments so, this week, we introduced an “Observability” toggle! 

Now when debugging, or admiring at how well your app is working, you can switch back and forth with ease from service view to logs. It even stores the context of where you were at so switching back takes you to where you were.

The response to the the formerly named “Log Explorer” has been monumental. Apparently, the next version of the Observability pane is now booked to screen at Cannes.

This toggle currently switches between the traditional Canvas, and the “Log Explorer” we released 

. However, we have much more planned than just logs so let us know what you’d like to see. 😏 (Our eyes must be acting up again.)

Nixpacks: Node 18 will soon be the default for all Node builds moving forward

Fixed a bug when you used the bulk variable editor it broke the Changeset experience and prevented you from being able to merge a forked environment.

Fixed a bug where it was possible to register for Railway with a case-sensitive email. Yea… we know.

Fixed a bug where template function variables would not always be collapsed

Fixed a bug where reference variables would not be resolved for the first template deploy of image-only services

Made the plan cards in the account page responsive to for users with Samsung S10s. No detail unscathed. 

Every week the Go-to-Market team chants: “Business! Business! Business!” but this week it was different, they chanted “Maturity! Maturity! Maturity!” - we now have a dedicated docs page for CISOs and Infra people alike explaining Railway for production use-cases. 

Fixed cases where the email link in our docs went to a dead email. 

https://res.cloudinary.com/railway/image/upload/v1693613064/cron-docker-switch_ypjaoc.png

Cron Jobs UI, Image Deploys, Observability Mode

Changelog #0149