Vadim Gedz
Vadim Gedz
It might be worth adding an optional flag enabling resulting manifest validation, for example, using [kubeconform](https://github.com/yannh/kubeconform).
Currently, we notify clients only about manual set/release operations. There is no feedback to the user about scheduled lockdown activities, so the users are not aware about the fact that...
We will need to support both old and new setups for some time, as it will be quite hard to change all projects in one go. This will potentially allow...
We already have Prometheus metrics. It might be worth preparing a Grafana dashboard for visualization.
Currently, we are cloning the repository each time. With shallow cloning, it is not that big of a problem, but eventually, it will become a bottleneck. Let's implement some repository...
Currently, the `argo-watcher` is unable to run in more than one replica due to a lack of task synchronization between instances. We need to find the best approach to resolve...
As the default notification controller doesn't fully meet our needs, let's implement a simple notification from our side containing specific deployment details. The work was already started in `feat/notification-poc`, but...
As project usage increased, we started to get internal requests to restrict the projects that should be visible to a specific user. Currently, it is not possible. We want to...
Currently, if we do not receive a valid response, we fail the whole process. Let's implement some sort of retries to handle temporary networking problems etc.
Currently, `development.md` is missing some crucial parts of information, and some parts are outdated. It might be worth rewriting it from scratch with a focus on launching the development environment...