Daniel Sanche

We [recently had an issue where a terraform dependency updated and broke our released code](https://github.com/GoogleCloudPlatform/cloud-ops-sandbox/pull/864). One of the brainstormed solutions to this problem would be to create an alert to...

[We recently had an issue where a terraform dependency updated and broke our released code](https://github.com/GoogleCloudPlatform/cloud-ops-sandbox/pull/864). I believe this repo should have a bot to catch these issues and auto-update dependencies...

pin terraform provider versions

2

Make sure we pin specific versions, so upstream updates don't break sandboxes

Add test that deploys through the Cloud Shell UI

1

We have E2E tests that build and test our custom cloud shell image in Docker. Real users use the Cloud Shell UI to deploy our image, which often has errors...

Automate checks against cloud shell

1

We sometimes run into issues where the Cloud Shell environment can't be provisioned (https://github.com/GoogleCloudPlatform/cloud-ops-sandbox/issues/583). It would be useful to have checks in place so we can catch these early, instead...

If the ` Test Using Existing Project` e2e test is moved above monitoring tests, it will fail, due to multiple resources being created. If we can integrate the tfstate files,...

There have been reports of the Loadgenerator being hit by bots, trying to modify the url field to find a potential log4j opening. The lodgenerator is not vulnerable to log4j,...

define and track uptime SLOs for sandbox project

1

Cloud Ops Sandbox is a live service that users are starting to depend on. We should consider tracking up time metrics, and possibly define SLOs around them

When we install sandbox over top of an existing sandbox, we expect the old sandbox to be updated to use the new code. Instead, the cluster seems to continue to...

Add tests to make sure services are healthy

3

We want to make sure fresh sandboxes don't have any errors. Either error logs, or failing SLOs