Fixes #6

since we don't expect admins to collect metrics for those services,
we provide them with some sensible defaults.

fixes potential security issue in swagger-ui

Root cause: Every readiness check triggered a cache warming
Solution: Move readiness to startup probe

Root cause: Due to the timeout, every request that was launched while fetching the previous manifests could led to another fetch and another cache.warmup
Solution: Be more synchronous

Root cause: Hash was computed on array.toString() which does not change much.
Solution: Use the actual content of the vite manifests to compute the version.

This prevents unnesessary large headers as the UI will filter for CSS files only

richard.petersen authored 3 years ago and julian.baeume committed 3 years ago

Includes the version string and dependencies

Signed-off-by: Julian Bäume <julian.baeume@open-xchange.com>

fixes readiness detection when using app-root different from /

this can not only be handled on ingress level, but we need to mount the
express app also on the app root level. This basically just works out of the
box.

describe a little better what this service is supposed to do and how
it plays along the other parts of the stack.

Promise.race resolve or rejects with the first promise that answers.
We now return the first promise that fulfills after all promises have
settled.

this is an implicit endpoint served by us.

this is going to be the next release, with incompatible API

just point manifest service instance to appsuite-stack release created for preview apps.
Needs some custom k8s resources (Ingress).

remove /api prefix and be more restful. This should help to simplify ingress objects

first hit wins. This allows us to transparently use sourcemaps and other files that
are not part of the sources according to the vite manifest