CAPI 1.62.0
Highlights
- Added ability to rotate database encryption key
Known Issues
- There is an issue that causes some apps to go into a crashing state following a deployment. Those apps can be brought back up with a restart. Apps that would be affected are apps that are not regularly deployed.
- When configured to make TLS connections to MySQL databases cloud controller ignores the configuration and makes a non-TLS connection.
CC API Version: 2.116.0 and 3.51.0
Service Broker API Version: 2.13
CAPI Release
- CC can mount nfs from a xenial stemcell details
- Operator can configure cf-deployment such that traffic between gorouter and cc is https details
- The 'cc.temporary_create_internal_domain' does not exist details
Cloud Controller
- 500 when updating processes with null memory details
- API client can observe that a deployment has completed details
- API client observes that non-web processes are updated by a deployment details
- API user with space.create permission can create a space using POST /v2/spaces details
- As a developer, if I try to create a service binding with accepts_incomplete=false and the broker starts an asynchronous operation, I see an error details
- Docs for fetching a route binding parameter do not show experimental details
- Operator can observe a metric for the number of DEPLOYING deployments details
- Operator can update existing CCDB data to be encrypted by a new encryption key. details
- Special characters in database passwords: 2018 edition details
- Stop using URI.encode/escape/decode/unescape details
- Work out why CC gets a GOAWAY from the Perm server when talking to it for the first time in a while details
Cloud Controller Database Migrations
- 20180628223056_rename_webish_process_guid_on_deployments.rb
- 20180703233121_set_missing_fields_for_backfilled_web_processes.rb
Pull Requests and Issues
- cloudfoundry/cloud_controller_ng #1171: Mark route binding parameters as experimental details
- cloudfoundry/capi-release #87: cloud_controller_api_health_check is flakey details
- cloudfoundry/cloud_controller_ng #1013:
pre-start
script innfs_server
job can take a long time details - cloudfoundry/cloud_controller_ng #1141: Error when using docker registry password longer than 1000 chars (for example Amazon Elasstic Container Registry tokens) details
- cloudfoundry/tps #13: tps crash request to include crashed cell id details