Releases: Financial-Times/upp-aggregate-healthcheck
Releases · Financial-Times/upp-aggregate-healthcheck
Helm migration
Helm integration (#10) * Helm and Jenkins integration * Some leftover config * Renamed UPP entry point for Jenkinsfile * Separated char name from version labels & Truncating labels to 63 chars
Remove ack endpoint - fix multiple remove ack
Fix remove ack (#9) * Removed path prefix for gtg endpoint, logged input params and updated documentation * Fixed log * Added more tests * Moved gtg handler a few lines up in the file to emphasize the fact that it uses the main router * Decreased timeout value for httpClient * Added parallel severity check * Fixed pod severity checking, decreased httpClient timeout * Fixed tests * Fixed tests and code format * Updated readme * Fixed remove ack endpoint * Moved cache-control header to services healthcheck handler * Added all cache control headers * Removed cache control headers from services health endpoint * Fixed circle CI
Removed path prefix for gtg endpoint, logged input params and added parallel severity levels checking
- removed path prefix for gtg endpoint, meaning that from now on the gtg endpoint will be accessible by __gtg instead of /__gtg (__health/__gtg)
- logged input params
- updated documentaion
- added some more tests
- added parallel severity checking
Used watchers on k8s resources to improve performance
Changes in this release:
- added watchers for deployments, services and acks map
- fixed severities on reload from cache
- added functionality to clear services that don't exist anymore from cache
- fixed functionality to disable a sticky category when a service is unhealthy
- added node name to individual service healthcheck for global services (i.e. system-healthcheck)