<div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><div>We've looked at doing passive queue declares to get queue depths for alerting, reporting and auto-scaling of our consumers. Unfortunately passive queue declares appear to require configure access. I can see why queue.declare requires this but passive commands perhaps should have a different bit setting?</div>
</blockquote><div><br></div><div>We've moved to monitoring this with custom checks to the web API.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div>In addition, we are currently doing all of our monitoring via the Management Plugin's API. Unfortunately to get any data, the user calling the API to list information requires administration access. I'd love to be able to let Nagios/Your_Monitoring_Solution_Here poll the Rabbit node and get data without giving it access to change all of the configuration state and remove users.</div>
<div></div></blockquote></div><div><br></div><div>That's actually something I've been wanting to bring up on the list. Currently, we give Nagios it's own user with admin access so it can use the REST API. We also have a provisioning user that is used to create (via the REST API) the various users our apps need in the individual Chef recipes (we wrote a Chef "library" to abstract out the API). It would be nice to have a "read only" admin type that doesn't allow that type of admin to create/change permissions, because it looks like outside of being able to change permissions that standard RMQ permissions apply to the API.</div>
<div><br></div><div>-J</div>