yahoo-eng-team team mailing list archive

[Thierry Carrez <thierry.carrez+lp@xxxxxxxxx>]

** Changed in: nova
       Status: Fix Committed => Fix Released

** Changed in: nova
    Milestone: None => grizzly-3

-- 
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/1068258

Title:
  _assert_compute_node_has_enough_memory doesn't support ram overcommit

Status in OpenStack Compute (Nova):
  Fix Released

Bug description:
  Cluster configuration

  compute-nodeA (4GB) of RAM reported in OS as 4010MB
  1 instance launched using 2048MB RAM

  compute-nodeB (4GB) of RAM reported in OS as 4010MB
  2 instances launched using 2048MB RAM each

  
  With this configuration it is not possible to migrate an instance from compute-nodeB to compute-nodeA and the scheduler reports the following:

  
  2012-10-18 17:34:51 WARNING nova.scheduler.manager [req-ce210164-9c16-4b3a-81ab-56dfd3c52803 f8f84ef1468540fc9dfa85fcf16c53a4 29715cb8427146aeb36841b4cbc8f728] Failed to schedule_live_migratio
  n: Migration error: Unable to migrate 15ca5cf7-649a-4f78-a554-9b83907eafc2 to rhel-compute1: Lack of memory(host:1962 <= instance:2048)
  2012-10-18 17:34:51 ERROR nova.rpc.amqp [req-ce210164-9c16-4b3a-81ab-56dfd3c52803 f8f84ef1468540fc9dfa85fcf16c53a4 29715cb8427146aeb36841b4cbc8f728] Exception during message handling
  2012-10-18 17:34:51 TRACE nova.rpc.amqp Traceback (most recent call last):
  2012-10-18 17:34:51 TRACE nova.rpc.amqp   File "/usr/lib/python2.6/site-packages/nova/rpc/amqp.py", line 253, in _process_data
  2012-10-18 17:34:51 TRACE nova.rpc.amqp     rval = node_func(context=ctxt, **node_args)
  2012-10-18 17:34:51 TRACE nova.rpc.amqp   File "/usr/lib/python2.6/site-packages/nova/scheduler/manager.py", line 97, in _schedule
  2012-10-18 17:34:51 TRACE nova.rpc.amqp     context, ex, *args, **kwargs)
  2012-10-18 17:34:51 TRACE nova.rpc.amqp   File "/usr/lib64/python2.6/contextlib.py", line 23, in __exit__
  2012-10-18 17:34:51 TRACE nova.rpc.amqp     self.gen.next()
  2012-10-18 17:34:51 TRACE nova.rpc.amqp   File "/usr/lib/python2.6/site-packages/nova/scheduler/manager.py", line 92, in _schedule
  2012-10-18 17:34:51 TRACE nova.rpc.amqp     return driver_method(*args, **kwargs)
  2012-10-18 17:34:51 TRACE nova.rpc.amqp   File "/usr/lib/python2.6/site-packages/nova/scheduler/driver.py", line 218, in schedule_live_migration
  2012-10-18 17:34:51 TRACE nova.rpc.amqp     disk_over_commit)
  2012-10-18 17:34:51 TRACE nova.rpc.amqp   File "/usr/lib/python2.6/site-packages/nova/scheduler/driver.py", line 301, in _live_migration_dest_check
  2012-10-18 17:34:51 TRACE nova.rpc.amqp     disk_over_commit)
  2012-10-18 17:34:51 TRACE nova.rpc.amqp   File "/usr/lib/python2.6/site-packages/nova/scheduler/driver.py", line 389, in assert_compute_node_has_enough_resources
  2012-10-18 17:34:51 TRACE nova.rpc.amqp     instance_ref, dest)
  2012-10-18 17:34:51 TRACE nova.rpc.amqp   File "/usr/lib/python2.6/site-packages/nova/scheduler/driver.py", line 422, in assert_compute_node_has_enough_memory
  2012-10-18 17:34:51 TRACE nova.rpc.amqp     raise exception.MigrationError(reason=reason % locals())
  2012-10-18 17:34:51 TRACE nova.rpc.amqp MigrationError: Migration error: Unable to migrate 15ca5cf7-649a-4f78-a554-9b83907eafc2 to rhel-compute1: Lack of memory(host:1962 <= instance:2048)
  2012-10-18 17:34:51 TRACE nova.rpc.amqp 

  
  /etc/nova/nova.conf:
  compute_scheduler_driver=nova.scheduler.filter_scheduler.FilterScheduler
  scheduler_available_filters=nova.scheduler.filters.standard_filters
  scheduler_default_filters=AvailabilityZoneFilter,RamFilter,ComputeFilter,CoreFilter,SameHostFilter,DifferentHostFilter
  ram_allocation_ratio=1.5

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1068258/+subscriptions