Blue-Green deployment scenario

[christian.beikov]

Hello,

I have been reading the rolling upgrade chapter from the

documentation and I have some questions.

1. The documentation states that in the target cluster, every cache

  that should be migrated, should use a CLI cache loader pointing to

  the source cluster. I suppose that this can only be configured via

  XML but not via the CLI or JMX? That would be bad because after a

  node restart the cache loader would be enabled again.

2. How would the JMX URL look like if I wanted to connect to a secured

  Wildfly over HTTP? I was thinking of

  jmx:http-remoting-jmx://USER:PASSWORD@HOST:PORT/CACHEMANAGER/CACHE

3. What do I need to do to rollback to the source cluster after

  switching a few nodes to the target cluster?

Thanks in advance!

[nadirx]

Hi Christian,

1. The documentation states that in the target cluster, every cache

  that should be migrated, should use a CLI cache loader pointing to

  the source cluster. I suppose that this can only be configured via

  XML but not via the CLI or JMX? That would be bad because after a

  node restart the cache loader would be enabled again.

Unfortunately that's true. I'll create a Jira to track this so that we can target it for a future release (probably 8.2).

2. How would the JMX URL look like if I wanted to connect to a secured

  Wildfly over HTTP? I was thinking of

  jmx:http-remoting-jmx://USER:PASSWORD@HOST:PORT/CACHEMANAGER/CACHE

service:jmx:http-remoting-jmx://HOST:PORT

3. What do I need to do to rollback to the source cluster after

  switching a few nodes to the target cluster?

The original cluster should be unaffected, so this shouldn't be a problem, but please describe how you intend to proceed so I understand your process better.

Tristan

[christian.beikov]

Nevermind about the rollback. I think I was thinking about something like canary deployment where I add some new nodes to a cluster to test if the function correctly. I guess to do canary deployments, the node would just need the CLI cache loader and if something goes wrong, just turn that node off again?