Show HN: I built an open source job scheduler to help script everything (exograd.com)
So I built a program for that. Eventline schedules, runs and monitors any script. Jobs can be executed on various runtimes (currently locally, with SSH, in Docker or in Kubernetes), manually or in reaction to various events. You then get observability (script output, history, execution stats…), full control (with the web UI, HTTP API or command line tool) and the ability to share scripts with others.
Eventline 1.0 is out. I use it myself, so does a friend and a first client. The roadmap is packed with features which will make Eventline even more useful: global registry for value injection, sub-jobs, scratchpad to share data between jobs, artifacts, ACL (and SSO for Eventline Pro), a lot more connectors, identities and events…
I quitted my job in 2020 and created a company, Exograd, hoping to become financially independent. This has been quite a ride, both exhilarating and scary: I made lots of mistakes and learned more in 22 months than in my entire career. I would like to bring enough revenue to be able to live from it, and then work on expanding Eventline and developing new technical products. I currently offer a support contract, a proprietary extension (Eventline Pro) for enterprise use, and cloud hosting to use Eventline without having to host it.
Happy to answer any question!
3 comments
[ 3.7 ms ] story [ 16.0 ms ] threadWhat is the advantage of this? For example, an "integration script" is a kind of re-occurant QA test? wouldn't that be better run in a CI (that has repo context) or even a tool like sonarqube that can you graphs of change over time, or a tool that provides a UI specific to the test framework?
To take your example: "CI/CD platforms make it easy to script tasks related to a repository, but are limited when trying to do anything else"
Isn't that a good thing? keep technical repo related stuff (e.g. build a package/image) in the CI - even if it's a call to some other system, or ansible script.
On that note, you could run Ansible AWX/tower and run a lot of things from that - how does Eventline compare?
Since you can add parameters to your jobs and run them manually with an auto-generated form, it becomes very simple to create complex tasks and let colleagues use them. You could have tasks to create feature environments, prepare onboarding for new employees, process GDPR requests, all kinds of jobs really.
I do know anything about Ansible AWX, but looking at the documentation, the only thing I can think is that it seems awfully complex when I just want to be able to run script and have a platform to monitor what happens and alert me when it fails.
can you give an example of the added complexity (other than the dependency on ansible script language itself)? For example you mention "access to credentials" which is one of the complexity points.