Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

Maximum Heat Stack Create Requests


As I install my application I execute approx 100 "heat stack create" commands over HTTP api. Each stack creates approx 2-4 VMs. I can see that Openstack accepts all these requests and then (visible on UI) processes the requests in batches and as it completes one adds another to the executing threadpool.

Is there a max number of heat stack create commands I can send towards HEAT? I would assume that if, for example, only one worker thread was available for work that I could still send in a hundred or even a couple of hundred requests, all requests would be accepted via web interface but then downside being it would take a long time for the single worker thread to worker thru all requests.

Question summary

(Assuming sufficient CPU and RAM resources available for my instances)

  1. Is there a max number of requests (stack create) to send in over HTTP interface at any one time
  2. If the HTTP API interface accepts the request it is reasonable to assume that HEAT engine will get to my request eventually (it wont fall over?)