Kristian Lake
03/26/2022, 4:48 PMFrank
Frank
Frank
Kristian Lake
03/27/2022, 6:21 PMOmry Hay
03/27/2022, 7:57 PMKristian Lake
03/27/2022, 9:13 PMRoss Coundon
03/27/2022, 9:39 PMKristian Lake
03/27/2022, 9:49 PM====================
➡️ Post-Deploy
====================
$ cd frontend && npm install --save-dev start-server-and-test
81.67 s
$ npm install --save-dev http-server
2.35 s
$ npm install --save-dev cypress
3.94 s
$ npm cypress run
0.32 s
Unknown command: "cypress"
To see a list of supported npm commands, run:
npm help
ERROR: Failed to run: npm cypress run
Kristian Lake
03/27/2022, 9:53 PMRoss Coundon
03/28/2022, 6:20 AMOmry Hay
03/28/2022, 7:51 AMIs there any reason you did not use seed.run for it?@Kristian Lake - IMO Seed is a CD tool and running Cypress is a bit missing the point there. As other pointed out you can basically run everything in Seed as you have the custom hooks and can run anything, but it’s not a tool that was built for it - also in terms of pricing a minute in GHA is less expensive than Seed’s minute. But also Github action offers a wide range of open source actions you can leverage like Slack integration, visual test results, saving the output of the run like videos and snapshots, creating reports, and also run those e2e in a schedule job in Production. For example, we run e2e in every PR, every merge to master, every deploy to Production and also as scheduled job twice a day on Prod. If there is a failure we save the video recording and images snapshot for debugging in the github action, and create a report using
mochawesome
(save it in S3, but you can also save it in Github action) and send it out in Slack whenever there is a failure.
We could do that with Seed, but we will have to develop and write a bunch of code to do it, and it basically not the right tool for the job IMO.Haider Abbas
03/28/2022, 10:25 AMKristian Lake
03/28/2022, 10:29 AMKristian Lake
03/28/2022, 10:33 AM