yahoo-eng-team team mailing list archive

[Launchpad Bug Tracker <1421155@xxxxxxxxxxxxxxxxxx>]

[Expired for OpenStack Compute (nova) because there has been no activity
for 60 days.]

** Changed in: nova
       Status: Incomplete => Expired

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1421155

Title:
  project instance limit also applicable for compute node

Status in OpenStack Compute (Nova):
  Expired

Bug description:
  I am running OpenStack Juno release, and am running it on high-capacity compute nodes. I created a few VMs of small sizes and after running 10 VMs on a compute node, new VM addition failed. It just gave a warning message in the logs which did not provide any indication of what the issue is. Following is a snippet of the nova-compute.log file on the compute node.
  -------------------------------------------
  2015-02-12 06:49:23.288 16846 WARNING oslo.messaging._drivers.amqpdriver [-] Number of call queues is greater than warning threshold: 10. There could be a leak.
  2015-02-12 06:49:46.622 16846 AUDIT nova.compute.resource_tracker [-] Auditing locally available compute resources
  2015-02-12 06:49:48.126 16846 AUDIT nova.compute.resource_tracker [-] Total physical ram (MB): 200913, total allocated virtual ram (MB): 16896
  -------------------------------------------

  After increasing the threshold limit of MaxInstances I was able to
  create new VMs.

  Ideal behavior should be:
  (1) Limit of number of VMs on compute node should be dependent on available resources rather than number of VMs quota for a project.
  (2) In worst case, there should be a different limiting parameter to govern number of VMs on a compute host.
  (3) Error logs should be more informative.

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1421155/+subscriptions