have database credentials on local nodes
have a better handling of the different instances on my environment, such as Production versus Staging versus Business Continuity, etc
Customer has different environments with Crowd Data Center clusters for each scenario, such as Production, Business Continuity, Staging, Development, QA, etc.
To refresh these environments data from production storage (shared folder) is copied to those environments, so is also copied the crowd.cfg.xml file, which contain the database credentials. Those need to be updated on each refresh, to point to the correct database instance
In a situation, for example, when Production cluster is totally down and they want to activate Business Continuity one, they need to make sure that the database connection information and credentials are the correct ones. Whether, if that was local in each node, they could have that correctly setup during server build.
Have Crowd handle the crowd.cfg.xml file the same way Confluence does with confluence.cfg.xml or Jira with the dbconfig.xml one. Or let it be configurable.
Customer currently have scripts on crontab to handle that.