Init-container that waits until only a certain amount of nodes are available until a new one starts
100K+
This init container checks the DNS entry for the service, to make sure that new nodes start after old nodes have been terminated. This image is used by the HiveMQ Operator (legacy).
To install:
In your Deployment add the following item to the spec.template.spec.initContainers field:
- name: dns-wait
image: hivemq/init-dns-wait:1.0.0
volumeMounts:
- mountPath: /mnt/misc
name: dns-wait-config
Also add the corresponding volume in the spec.template.spec.volumes field:
- name: dns-wait-config
configMap:
name: dns-wait-config
And finally, add the ConfigMap in the same namespace as the Deployment like such, replacing the parameters as required.
apiVersion: v1
kind: ConfigMap
data:
# Your HiveMQ DNS discovery address
address: your-discovery.namespace.svc.cluster.local.
# Time to wait for replication tasks after a previous node has left
coolDown: "40"
# Target count to wait for during the upgrade (this should be cluster size + 1)
count: "13"
# How long to wait in total while polling the DNS address. If this time is exceeded, the node will continue startup anyway.
timeOut: "600"
metadata:
name: dns-wait-config
WARNING: The legacy HiveMQ Operator and its Helm chart have been retired since April 2025 and no longer receive updates or support. To migrate to the current HiveMQ Platform Operator for Kubernetes, see HiveMQ Legacy Operator to Platform Operator Migration Guide.
Content type
Image
Digest
sha256:96f972c30…
Size
5.7 MB
Last updated
almost 4 years ago
docker pull hivemq/init-dns-wait:1.0.1