Limit SSH validation for standby nodes in DR

Use Case

Ever wanted to maintain an active/passive environment where your standby nodes are shutdown? You may want to benefit from cost saving in cloud infrastructure or as a corporate requirement for disaster recovery purposes.

Myst supports standby nodes which are inaccessible via an SSH connection. Continue reading for more information as there are some limitations to be mindful of.

Enabling the feature

  1. Go to Platform Blueprint > Global Variables

  2. Create a new global variable:

    validation.ssh=false

  3. Run the update action making sure all your nodes are initially active.

Myst Action List

Product

Action

Standby Node State

Outcome

Notes

FMW

provision

Active

Success

Nodes are not required to be active

FMW

restart

Down

Success

Use the limit nodes feature to ignore standby nodes

FMW

patch

Down

Success

Use the limit nodes feature to ignore standby nodes. Best practice to start stand by and apply patches in a rolling fashion

FMW

update

Down

Success

Configuration changes are applied to the AdminServer. JCA Adapter configuration Distributed Mode should be ideally be set to shared. If set to distributed then only the active nodes are updated. When the standby nodes come online then the 'update' action needs to be run.

FMW

start

Down back to Active

Success

Code synchronizes from AdminServer allowing FMW to come up in a compatible state

OSB / SOA

deploy

Down

Success

Code synchronizes when standby nodes become active

Custom XPath

deploy-xpath

Down

Success

Requires shared disk otherwise custom XPath deployments will fail to deploy on standby nodes.

Jar Library

deploy-library

Down

Success

Requires shared disk otherwise Jar Library deployments will fail to deploy on standby nodes.

Last updated