yahoo-eng-team team mailing list archive

[Jianghua Wang <wjh_fresh@xxxxxxx>]

Public bug reported:

Observed several failure in citrix xenserver CI for this test case:
tempest.api.compute.servers.test_server_rescue

See there are timeout waiting for vif:

$ grep 'Timeout waiting for vif plugging callbac' screen-n-cpu.txt.gz
2016-12-20 10:58:52.036 4528 WARNING nova.virt.xenapi.vmops [req-ff027cef-59be-4326-95e1-065f68077d63 tempest-ServerRescueTestJSON-1293983176 tempest-ServerRescueTestJSON-1293983176] [instance: 28b094ee-c571-4083-b72b-5ea78f1f4291] Timeout waiting for vif plugging callback

For rescue, it seems shouldn't wait for this event as this port should be active at the rescuing start.
But observed:
neutron service reported the 2nd vif-plugin event:


2016-12-20 10:52:31.689 712 DEBUG neutron.notifiers.nova [-] Sending events: [{'status': 'completed', 'tag': u'52d79a78-7205-4e69-8005-76a3cebbf267', 'name': 'network-vif-plugged', 'server_uuid': u'28b094ee-c571-4083-b72b-5ea78f1f4291'}] send_events /opt/stack/new/neutron/neutron/notifiers/nova.py:248

2016-12-20 10:53:45.179 712 DEBUG neutron.notifiers.nova [-] Sending
events: [{'status': 'completed', 'tag':
u'52d79a78-7205-4e69-8005-76a3cebbf267', 'name': 'network-vif-plugged',
'server_uuid': u'28b094ee-c571-4083-b72b-5ea78f1f4291'}] send_events
/opt/stack/new/neutron/neutron/notifiers/nova.py:248


And nova attempts to wait for this event after the 2nd event sent out; so it won't catch the 2nd event at all:
2016-12-20 10:53:46.326 4528 DEBUG nova.virt.xenapi.vmops [req-ff027cef-59be-4326-95e1-065f68077d63 tempest-ServerRescueTestJSON-1293983176 tempest-ServerRescueTestJSON-1293983176] wait for instance event:[('network-vif-plugged', u'52d79a78-7205-4e69-8005-76a3cebbf267')] _spawn /opt/stack/new/nova/nova/virt/xenapi/vmops.py:599

** Affects: nova
     Importance: Undecided
         Status: New

-- 
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/1651650

Title:
  XenAPI: server rescue test sometime failed with timeout waiting for
  vif plugging

Status in OpenStack Compute (nova):
  New

Bug description:
  Observed several failure in citrix xenserver CI for this test case:
  tempest.api.compute.servers.test_server_rescue

  See there are timeout waiting for vif:

  $ grep 'Timeout waiting for vif plugging callbac' screen-n-cpu.txt.gz
  2016-12-20 10:58:52.036 4528 WARNING nova.virt.xenapi.vmops [req-ff027cef-59be-4326-95e1-065f68077d63 tempest-ServerRescueTestJSON-1293983176 tempest-ServerRescueTestJSON-1293983176] [instance: 28b094ee-c571-4083-b72b-5ea78f1f4291] Timeout waiting for vif plugging callback

  For rescue, it seems shouldn't wait for this event as this port should be active at the rescuing start.
  But observed:
  neutron service reported the 2nd vif-plugin event:

  
  2016-12-20 10:52:31.689 712 DEBUG neutron.notifiers.nova [-] Sending events: [{'status': 'completed', 'tag': u'52d79a78-7205-4e69-8005-76a3cebbf267', 'name': 'network-vif-plugged', 'server_uuid': u'28b094ee-c571-4083-b72b-5ea78f1f4291'}] send_events /opt/stack/new/neutron/neutron/notifiers/nova.py:248

  2016-12-20 10:53:45.179 712 DEBUG neutron.notifiers.nova [-] Sending
  events: [{'status': 'completed', 'tag':
  u'52d79a78-7205-4e69-8005-76a3cebbf267', 'name': 'network-vif-
  plugged', 'server_uuid': u'28b094ee-c571-4083-b72b-5ea78f1f4291'}]
  send_events /opt/stack/new/neutron/neutron/notifiers/nova.py:248

  
  And nova attempts to wait for this event after the 2nd event sent out; so it won't catch the 2nd event at all:
  2016-12-20 10:53:46.326 4528 DEBUG nova.virt.xenapi.vmops [req-ff027cef-59be-4326-95e1-065f68077d63 tempest-ServerRescueTestJSON-1293983176 tempest-ServerRescueTestJSON-1293983176] wait for instance event:[('network-vif-plugged', u'52d79a78-7205-4e69-8005-76a3cebbf267')] _spawn /opt/stack/new/nova/nova/virt/xenapi/vmops.py:599

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1651650/+subscriptions