periodically seeing KeyError: 'old_instance_type_memory_mb' in compute.log
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack Compute (nova) |
Fix Released
|
High
|
Dan Smith |
Bug Description
When running multiple migrations concurrently, I see an error the repeats in the compute.log file. The trace can be seen here:
Here is the steps I followed to recreate this issue:
(1). Boot the instances
nova boot --image be8b6475-
nova boot --image be8b6475-
nova boot --image be8b6475-
[root@vs347 ~]# nova list
+------
| ID | Name | Status | Networks |
+------
| ccf6212b-
| 739a6d71-
| 11dabeb8-
+------
bvt-instance-
bvt-instance-
bvt-instance-
(2). migrate the instances simultaneously
[root@vs347 ~]# nova migrate MIGRATE1
[root@vs347 ~]# nova migrate MIGRATE2
[root@vs347 ~]# nova migrate MIGRATE3
[root@vs347 ~]#
* State as of 4:24 p.m.
+------
| ID | Name | Status | Networks |
+------
| ccf6212b-
| 739a6d71-
| 11dabeb8-
+------
* Now they are in resize-verify state so do confirm
[root@vs347 ~]# nova resize-confirm MIGRATE1
[root@vs347 ~]# nova resize-confirm MIGRATE2
[root@vs347 ~]# nova resize-confirm MIGRATE3
[root@vs347 ~]#
[root@vs347 ~]# nova list
+------
| ID | Name | Status | Networks |
+------
| ccf6212b-
| 739a6d71-
| 11dabeb8-
+------
[root@vs347 ~]# nova show MIGRATE1
+------
| Property | Value |
+------
| status | ERROR |
| updated | 2013-03-
| OS-EXT-
| OS-EXT-
| key_name | None |
| image | Rhel6MasterFile (be8b6475-
| hostId | a3883219ffb6890
| OS-EXT-STS:vm_state | error |
| OS-EXT-
| OS-EXT-
| flavor | m1.tiny (1) |
| demonet network | 172.0.0.2 |
| id | ccf6212b-
| security_groups | [{u'name': u'default'}] |
| user_id | 3ccf55fd609b453
| name | MIGRATE1 |
| created | 2013-03-
| tenant_id | 67b1c37f4ca6428
| OS-DCF:diskConfig | MANUAL |
| metadata | {} |
| accessIPv4 | |
| accessIPv6 | |
| fault | {u'message': u'KeyError', u'code': 500, u'details': u'\'old_
| | File "/usr/lib/
| | return function(self, context, *args, **kwargs) |
| | File "/usr/lib/
| | rt.confirm_
| | File "/usr/lib/
| | self.update_
| | File "/usr/lib/
| | retval = f(*args, **kwargs) |
| | File "/usr/lib/
| | self._update_
| | File "/usr/lib/
| | migration) |
| | File "/usr/lib/
| | itype = instance_
| | File "/usr/lib/
| | instance_type[key] = type_fn(
| | ', u'created': u'2013-
| OS-EXT-
| OS-EXT-
| config_drive | |
+------
Changed in nova: | |
milestone: | grizzly-rc2 → 2013.1 |
I am seeing this issue with periodic tasks as well - not just running multiple migrations