johnboyes / example-synthetic-monitor Goto Github PK

pro: less chance of people being confused between test_helper and spec_helper

con: test_helper is the standard name

Also, is it feasible to have a Reek check (either on GitHub and/or a pre-commit hook)?

Add a pronto check?

And a status badge?

Will have to have a good look at what other checks are available.

At the moment the automated comments are by me, which is confusing.

• Create new GitHub bot account
• Create a new personal access token on the new account with repo access
• Add the new account as a collaborator to this repo
• Change the environment variable in CircleCI to use the new token

• Better to have the config explicit in the codebase, for documentation/findability

• Wait until the config is stable

When building using Docker, the build errors saying Errno::EADDRNOTAVAIL: Failed to open TCP connection to localhost:7001 (Cannot assign requested address - connect(2) for "localhost" port 7001) e.g. see https://circleci.com/gh/johnboyes/synthetic-monitor/16

This link suggests that there can be networking issues when using docker: https://discuss.circleci.com/t/cant-access-to-elasticsearch-after-plugin-install/20578

Benefits would be:

• better example, shows the full end-to-end journey
• functional tests would have better coverage of the end-to-end journey
• set up a new slack team just for this purpose
• ensure that only the team owner can post

This will be useful as it will give feedback on how stable a simple monitoring app is in production.

• The app should alert me on a real Slack channel if there are test failures.

Depends on #29

lots of things to do including:

• remove blurb in README about SMS notifications coming soon

• link to foreman article is out of date

• link to tests needs to be updated

• Add some info about Poltergeist/Phantom JS?

will probably be more things, too