race condition when updating instance after it's deleted

[Chris Behrens]

Because we allow deletes during build again... There's a race condition in compute manager's _run_instance() where an instance can be deleted in the middle of an rpc.call to the network manager to allocate networks. rpc.call will raise a RemoteError and we will try to set the instance to ERROR state. The setting of instance to ERROR state fails with InstanceNotFound... and eats the original exception.

We shouldn't blow up when we try to set an instance to ERROR that has been deleted out from under us.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/5007

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/5007
Committed: http://github.com/openstack/nova/commit/d94f22dc7a24ec8f769a4fbbffeb1b34780bbf3d
Submitter: Jenkins
Branch: master

commit d94f22dc7a24ec8f769a4fbbffeb1b34780bbf3d
Author: Chris Behrens <email address hidden>
Date: Wed Mar 7 02:10:18 2012 +0000

    Ignore InstanceNotFound when trying to set instance to ERROR

    Fixes bug 948632

    There's a race condition where an instance can be deleted from an API
    call... immediately before some other error occurs during build.

    Also: compute tests should not test raising quantum exceptions, as
    those could never possibly make it back to us. rpc.call only raises
    RemoteError... so use that instead.

    Change-Id: Iacfe511a37e937eaec1d6213bf73153aa4d4e9bf