69cb269230
In K8S version 1.10, the proxy can sometimes get stuck believing that some services do not have any endpoints. This seems to be triggered by network instability, though the proxy doesn't seem to recover on its own, while bouncing the pod fixes the issue. This change adds a naive means of detecting and recoverying from this (`iptables-save | grep 'has no endpoints'` in the liveness probe) that may occasionally have false positives. As such, the liveness probe is configured very conservatively to avoid triggering CrashLoopBackoff in the event of a false positive. Finally, there is a whitelist feature to help avoid false positives for services that are known to legitimately have empty endpoints during the course of normal operation (e.g. Patroni might manage such an endpoint list). Change-Id: I29a770fab70b1fb79db59ef5408f40b2af1c01f9 |
||
---|---|---|
charts | ||
doc/source | ||
etc/promenade | ||
examples | ||
promenade | ||
tests | ||
tools | ||
.dockerignore | ||
.gitignore | ||
.gitreview | ||
.zuul.yaml | ||
Dockerfile | ||
LICENSE | ||
Makefile | ||
README.md | ||
entrypoint.sh | ||
requirements-direct.txt | ||
requirements-frozen.txt | ||
requirements.txt | ||
setup.py | ||
test-requirements.txt | ||
tox.ini |
README.md
Promenade
Promenade is a tool for bootstrapping a resilient Kubernetes cluster and managing its life-cycle via Helm charts.
Documentation can be found here.
Roadmap
The detailed Roadmap can be viewed on the OpenStack StoryBoard.
- Cluster bootstrapping
- Initial Genesis process results in a single node Kubernetes cluster with Under-cloud components deployed using Armada.
- Joining sufficient master nodes results in a resilient Kubernetes cluster.
- Destroy Genesis node after bootstrapping and re-provision as a normal node to ensure consistency.
- Life-cycle management
- Decommissioning of nodes.
- Updating Kubernetes version.
Getting Started
To get started, see getting started.
Configuration is documented here.
Bugs
Bugs are tracked in OpenStack StoryBoard.