The best option here is likely to manually build a loop and call the each workflow for each server asynchronously. You then collect all the workflow tokens and save them for the end of the main workflow run. Let it block at the end in a loop checking the status of the child async workflows until they report completed. Then log whether they finished successfully or with error. Another benefit of the async approach is that the workflow tokens will be collected independently of the main workflow within the vRO client so you can inspect the status of the workflow run later just like any other workflow you kick off. It won't be hidden inside the main workflow run, which enables you to inspect them individually and right click the run and say "run again" if you want to isolate it and try again.
You can build the loop in the schema or use a scriptable task to do it.
The script would look something like this. It takes 2 inputs, an array called servers (in my example, just an array of strings) and a Workflow object which is the workflow you want to run for each iteration of the loop.
var wfTokens = new Array();
for(var s in servers) {
var props = new Properties();
//push workflow inputs
props.put("serverName", servers[s]);
wfTokens[s] = wf.execute(props, null, "");
}
var stillRunning = true;
//wait while all workflows complete
while(stillRunning) {
stillRunning = false;
for(var t in wfTokens) {
var state = Server.getWorkflowTokenState(wfTokens[t].id);
if(state == "running" || state == "waiting" || state == "active" || state == "suspended") stillRunning = true;
}
//sleep for 5 seconds before checking status again
if(stillRunning) System.sleep(5000);
}
//iterate through the collection of tokens and check the state. If it failed, output the exception that was thrown.
for(var t in wfTokens) {
var state = Server.getWorkflowTokenState(wfTokens[t].id);
if(state == "failed") {
System.log(wfTokens[t].id + " - failed - " + wfTokens[t].exception)
}
else {
System.log(wfTokens[t].id + " - " + state);
}
}
