[EnMasse] Configuration options of messaging infrastructure

[Ulf Lilleengen]

Hi,

As part of this sprint, I'm working on the feature for automatically
(re)configure and upgrade messaging infrastructure based a new
'BrokeredInfraConfig' and 'StandardInfraConfig' resource type. See
https://github.com/EnMasseProject/enmasse/issues/1585 for more details.

The feature is now implemented, but I would like to gather some input on
what configuration options should be supported. At present, one can change:

* Admin pods memory usage
* Broker pods memory usage
* Broker component storage usage (only for 'new' broker deployments, not
applied to existing)
* Broker "address full" policy (BLOCK, FAIL, PAGE etc.)
* Router pods memory usage
* Router link capacity

A couple of more settings I think would be useful is:

* Storage class to use for volumes (otherwise use default)
* Node labels (too force pods to be deployed to nodes matching certain
hardware requirements for instance)

Are there other router, broker or openshift settings that would be useful
to change?

Best regards,

Ulf
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/enmasse/attachments/20181015/46dd1b22/attachment.htm>