OpenStack Compute (nova)

  1. Bug #1299139
  2. Comment #3

Comment 3 for bug 1299139

Revision history for this message
Matt Riedemann (mriedem) wrote : Re: Instances stuck in deleting task_state never cleaned up

Tracing the first bad instance I have in that paste, 1d278e77-9900-4b2b-85eb-3a6aac4198c1, I see the delete request in the nova api log (attached):

2014-03-27 17:25:33.431 12851 INFO nova.osapi_compute.wsgi.server [req-06c80607-8eef-409e-a29c-152d693635ad 7bdd970e41c14774b4dd8f4e0e92d05c 5ab6bffec8ae44beb1b69750a0e8ea00] 9.5.48.212 "DELETE /v2/5ab6bffec8ae44beb1b69750a0e8ea00/servers/1d278e77-9900-4b2b-85eb-3a6aac4198c1 HTTP/1.1" status: 204 len: 198 time: 0.5162849

But I never see any terminate_instance happening in the compute log for that instance uuid (compute log attached).

The VM isn't in the hypervisor anymore either so the driver deleted the VM from the hypervisor but something blew up somewhere, not sure why I can't find that in the logs.