« back to all changes in this revision
Viewing changes to ChangeLog
- browse files at revision 1.1.74
- compare with another revision
- download diff
- download tarball
- view history from revision 1.1.74
- Committer: Package Import Robot
- Author(s): Adam Gandelman
- Date: 2013-08-09 10:12:27 UTC
- mto: This revision was merged to the branch mainline in revision 107.
- Revision ID: package-import@ubuntu.com-20130809101227-flqfubhwpot76pob
Tags: upstream-2013.1.3
Import upstream version 2013.1.3
- files added:
- nova/openstack/common/xmlutils.py
- files modified:
- AUTHORS
added
removed
ChangeLog
1
commit 89c7ee409564d3909dbe8e1fb171ae7273508339
2
Author: Thierry Carrez <thierry@openstack.org>
3
Date: Thu Aug 8 12:07:56 2013 +0200
4
5
Remove unsafe XML parsing
6
7
Move to using xmlutils for all XML parsing. Resolves bug 1190229.
8
Grizzly backport of Mikal Still's master patch.
9
10
Change-Id: I43afb2e188bbea99ea30fe6cb2eb1aeedc4ddfd4
11
12
commit 1934dbc8984fe02edb5c410995501cdb87b77647
13
Author: Yaguang Tang <yaguang.tang@canonical.com>
14
Date: Wed Apr 24 16:54:05 2013 +0800
15
16
Fix VMware Hyper console url parameter error.
17
18
fix bug #1172177
19
20
(cherry picked from commit 2e16432ae0337101c80d8ba6fbdf2e8a9c8d85e7)
21
22
Conflicts:
23
24
nova/tests/test_vmwareapi.py
25
nova/virt/vmwareapi/vmops.py
26
27
Change-Id: Iab858741cba61f218c639d76fa80c047196bcc64
28
29
commit d18ceb168f7ec41e736c9cd10d4538cb2ceffe31
30
Merge: 798984d aa081f3
31
Author: Jenkins <jenkins@review.openstack.org>
32
Date: Fri Aug 2 18:54:25 2013 +0000
33
34
Merge "Reset migrating task state for MigrationError exceptions" into stable/grizzly
35
36
commit 798984d6bafabb71f390bf253f61847a18d45630
37
Merge: b18818a 2af0a07
38
Author: Jenkins <jenkins@review.openstack.org>
39
Date: Fri Aug 2 08:00:38 2013 +0000
40
41
Merge "Config drive attached as cdrom" into stable/grizzly
42
43
commit b18818add1ed2f60d6dee7f6f826c1338e1b8335
44
Merge: 01edbaa 2c1061d
45
Author: Jenkins <jenkins@review.openstack.org>
46
Date: Fri Aug 2 05:31:25 2013 +0000
47
48
Merge "Add inst_type parameter" into stable/grizzly
49
50
commit 01edbaa471f2dfe848f9e47cf862a8ce0b38e246
51
Merge: c3fddb7 ef195a9
52
Author: Jenkins <jenkins@review.openstack.org>
53
Date: Fri Aug 2 03:23:22 2013 +0000
54
55
Merge "Fixes Hyper-V issues on versions prior to 2012" into stable/grizzly
56
57
commit c3fddb78dc38390f78c383efd33e224d1129dfef
58
Merge: bd7e467 eff629f
59
Author: Jenkins <jenkins@review.openstack.org>
60
Date: Fri Aug 2 02:25:02 2013 +0000
61
62
Merge "Fix resizes with attached file-based volumes" into stable/grizzly
63
64
commit bd7e467038a6f309cf76d4393e0a5f2b4b4b3339
65
Merge: 8247dc8 52ad911
66
Author: Jenkins <jenkins@review.openstack.org>
67
Date: Thu Aug 1 19:03:34 2013 +0000
68
69
Merge "Use cached nwinfo for secgroup rules" into stable/grizzly
70
71
commit 8247dc86dca8a4f6525e5d44128182c94fd4d17c
72
Merge: 340d337 6df1b7a
73
Author: Jenkins <jenkins@review.openstack.org>
74
Date: Thu Aug 1 00:39:31 2013 +0000
75
76
Merge "Make flavors is_public option actually work" into stable/grizzly
77
78
commit 52ad911963da4095b213952dee3a430fe0c4c30f
79
Author: Vishvananda Ishaya <vishvananda@gmail.com>
80
Date: Fri Jul 19 10:23:59 2013 -0700
81
82
Use cached nwinfo for secgroup rules
83
84
This stops a potential DOS with source security groups by using the
85
db cached version of the network info instead of calling out to
86
the network api multiple times.
87
88
Fixes bug 1184041
89
90
Change-Id: Id5f24ecf0e8cce60c27a9aecbc6e606c4c44d6b6
91
(cherry picked from commit 85aac04704350566d6b06aa7a3b99649946c672c)
92
93
commit 340d33763d0cdb6482cd4ae2c8c60a3fcd52b043
94
Merge: e0aa0f5 d1edaa9
95
Author: Jenkins <jenkins@review.openstack.org>
96
Date: Wed Jul 31 18:39:34 2013 +0000
97
98
Merge "Don't deallocate network if destroy time out" into stable/grizzly
99
100
commit e0aa0f5cc98e452a510119f355784d3300806c05
101
Merge: 60712a4 5f12da3
102
Author: Jenkins <jenkins@review.openstack.org>
103
Date: Wed Jul 31 08:00:24 2013 +0000
104
105
Merge "Verify that CONF.compute_driver is defined" into stable/grizzly
106
107
commit 60712a46528fa0733cc5597e06fc984f06eed3ce
108
Merge: c57fded 1ef0d96
109
Author: Jenkins <jenkins@review.openstack.org>
110
Date: Wed Jul 31 07:24:50 2013 +0000
111
112
Merge "VNC console does not work with VCDriver" into stable/grizzly
113
114
commit 2af0a078851a39ebad2cd1969ec3d52c49c15159
115
Author: Tony Yang <bjyangf@cn.ibm.com>
116
Date: Thu Jul 18 18:45:24 2013 +0800
117
118
Config drive attached as cdrom
119
120
Some guest OSes (e.g. Windows) require config drive to be a cdrom device
121
to access the iso9660 filesystem on it. Option config_drive_format is
122
therefore set to also control the device type of config drive. If it's
123
set to 'iso9660', config drive will be cdrom; if it's 'vfat', config
124
drive will be disk.
125
126
DocImpact
127
Fixes: bug #1155842
128
Change-Id: I6c08b1b8040a1fd0db8e2b3b1fc798060733001f
129
(cherry picked from commit 7f4b5771633f519a54aae985ae526418114b1a29)
130
131
commit c57fdedc250b03af8e85c60f70d1456508952ffe
132
Merge: c698dd3 d42fd15
133
Author: Jenkins <jenkins@review.openstack.org>
134
Date: Mon Jul 29 22:31:31 2013 +0000
135
136
Merge "Allow ::/0 for IPv6 security group rules" into stable/grizzly
137
138
commit c698dd32a47630c1d58b0f6bc9dfa34afdebec0e
139
Merge: e572faf 8e4a374
140
Author: Jenkins <jenkins@review.openstack.org>
141
Date: Mon Jul 29 22:07:54 2013 +0000
142
143
Merge "Fix VMwareVCdriver reporting incorrect stats" into stable/grizzly
144
145
commit ef195a97e8076fe30e75d91c2c734db589cba33f
146
Author: Alessandro Pilotti <ap@pilotti.it>
147
Date: Fri Jul 19 19:44:45 2013 +0300
148
149
Fixes Hyper-V issues on versions prior to 2012
150
151
Fixes bug: #1203059
152
153
Hyper-V 2012 introduced live migration support and a WMI namespace that
154
can be used to handle iSCSI targets.
155
156
The Hyper-V driver was not properly recognizing versions of the os prior to
157
2012 causing an exception during startup due to the usage of unsupported
158
features.
159
160
Unit tests have been added to test the behaviour of the driver with
161
different os versions.
162
163
Change-Id: I88056a2e89552b94513ba1ead046dfda57567815
164
165
commit d42fd15e624e1692925a9957775a28a380e4c581
166
Author: Rosario Di Somma <rosario.disomma@dreamhost.com>
167
Date: Wed Jul 10 16:32:48 2013 -0400
168
169
Allow ::/0 for IPv6 security group rules
170
171
Nova fails to create a security group rule if ::/0 is used as
172
value for the cidr field
173
174
Fixes: LP bug #1199943
175
Cherry picked from https://review.openstack.org/36556
176
177
Change-Id: Ifce53dbc9f0740337921491aa752900892f47052
178
Signed-off-by: Rosario Di Somma <rosario.disomma@dreamhost.com>
179
180
commit e572fafe747334c2e207b174d111eec3e2f74796
181
Merge: 34ddd27 11046d9
182
Author: Jenkins <jenkins@review.openstack.org>
183
Date: Mon Jul 29 18:35:12 2013 +0000
184
185
Merge "Force reopening eventlet's hub after fork" into stable/grizzly
186
187
commit 34ddd27a0107eda6a9362793c1b29eaf5aae08bf
188
Merge: 864837f 2580555
189
Author: Jenkins <jenkins@review.openstack.org>
190
Date: Mon Jul 29 18:34:52 2013 +0000
191
192
Merge "eventlet.monkey_patch() for spice and vnc proxy" into stable/grizzly
193
194
commit 864837fee6655f4c99dbdc2e4064fd390b5a95c2
195
Merge: 756a772 e301ab4
196
Author: Jenkins <jenkins@review.openstack.org>
197
Date: Mon Jul 29 18:24:03 2013 +0000
198
199
Merge "Update instance.node on evacuate" into stable/grizzly
200
201
commit 756a772b7f627e3077a72edad890d8e6cc47535c
202
Merge: 889b215 d34d4ca
203
Author: Jenkins <jenkins@review.openstack.org>
204
Date: Mon Jul 29 18:23:42 2013 +0000
205
206
Merge "Make resize/migrated shared storage aware" into stable/grizzly
207
208
commit 889b2157a39929db1d6edf50ddad23605ab2ceab
209
Merge: 91bcd15 a04b312
210
Author: Jenkins <jenkins@review.openstack.org>
211
Date: Mon Jul 29 18:23:27 2013 +0000
212
213
Merge "Don't pass empty image to filter on live migration" into stable/grizzly
214
215
commit 5f12da37dd229f7afe596d4b0e9741834bf8ddaa
216
Author: Simon Pasquier <pasquier.simon@gmail.com>
217
Date: Mon May 27 17:17:11 2013 +0200
218
219
Verify that CONF.compute_driver is defined
220
221
The compute_driver parameter is a mandatory parameter only for nodes
222
running the Nova compute service. However Other Nova services like
223
Nova API actually call the compute_driver_matches() function.
224
225
Change-Id: Iac7cd46c6bdb4268cfbc650dcde0c95e69545f04
226
Fixes: bug #1184641
227
(cherry picked from commit fa5079c257c00b8b56138233aca5755c54db9143)
228
229
commit 8e4a3746a063633bd2f2a86d7ad922621424dcc0
230
Author: kirankv <kiran-kumar.vaddi@hp.com>
231
Date: Fri May 17 07:14:44 2013 -0700
232
233
Fix VMwareVCdriver reporting incorrect stats
234
235
fix bug #1180779
236
237
VMware nova compute drivers VMwareESXDriver and VMwareVCDriver
238
currently report stats for the column local_gb in MB instead of in GB
239
The fix would be to modify vmwareapi/host.py to ensure the values
240
reported for disk space are calculated in GB instead of MB.
241
Fix is done by further dividing the existing MB value by 1024 to
242
convert to GB.
243
The unit tests to check for capacity and freespace have been modified
244
to check for GB.
245
246
Change-Id: I0610047ee2bddefe0ccb46568c0f48de4c0941af
247
248
commit 91bcd159d29421bc877959cbc601154c21e8705e
249
Merge: a61872b c17e0c1
250
Author: Jenkins <jenkins@review.openstack.org>
251
Date: Sun Jul 28 22:30:04 2013 +0000
252
253
Merge "Fix a race where a soft deleted instance might be removed by mistake" into stable/grizzly
254
255
commit a61872bd5dafcc40ae38e278ccfe07848a9770c1
256
Merge: 1cf30a3 fbde109
257
Author: Jenkins <jenkins@review.openstack.org>
258
Date: Sun Jul 28 06:50:24 2013 +0000
259
260
Merge "Remove junk file when ftp transfer failure" into stable/grizzly
261
262
commit 1cf30a3882a35e87dfcd1b8268a7487de802b2e4
263
Merge: 9cb9361 2f0f1c2
264
Author: Jenkins <jenkins@review.openstack.org>
265
Date: Sat Jul 27 15:57:15 2013 +0000
266
267
Merge "Checks if volume can be attached" into stable/grizzly
268
269
commit 9cb93616626e7834d3aa01fc26c1ab692ecd5311
270
Merge: 7f1c336 c30e03e
271
Author: Jenkins <jenkins@review.openstack.org>
272
Date: Sat Jul 27 15:56:53 2013 +0000
273
274
Merge "Fix filtering aggregate metadata by key" into stable/grizzly
275
276
commit 7f1c3366b68731f663eaf31fa720afcfde62b1e8
277
Merge: 109f3a9 d35e748
278
Author: Jenkins <jenkins@review.openstack.org>
279
Date: Sat Jul 27 14:57:10 2013 +0000
280
281
Merge "Avoid redefining host to none in get_instance_nw_info(...)" into stable/grizzly
282
283
commit 109f3a9d68ab412e0fec96883a84dfa0c8c671e8
284
Merge: 2563c44 9d783b6
285
Author: Jenkins <jenkins@review.openstack.org>
286
Date: Sat Jul 27 04:34:35 2013 +0000
287
288
Merge "Don't make nova-compute depend on iSCSI" into stable/grizzly
289
290
commit 2563c44a22ba778cfc86f5a1d0fb87c65d079fa1
291
Merge: cbcdfc5 5685ea8
292
Author: Jenkins <jenkins@review.openstack.org>
293
Date: Sat Jul 27 04:25:37 2013 +0000
294
295
Merge "Fix aggregate_get_by_host host filtering" into stable/grizzly
296
297
commit 2f0f1c25553dff0e8e9e3199331087151f5077e8
298
Author: Loganathan Parthipan <parthipan@hp.com>
299
Date: Tue Apr 30 13:13:33 2013 +0000
300
301
Checks if volume can be attached
302
303
This patches makes a boot fail fast if one of the volumes cannot
304
be attached at boot time. Currently the request is accepted but
305
spawning fails in addition to leaving the volume state as
306
available which is an inconsistent state
307
308
fixes: bug 1166770
309
(cherry picked from commit 8e4eec840d504d27a3a4c640dba67b3b47e81730)
310
311
Conflicts:
312
313
nova/tests/compute/test_compute.py
314
315
Change-Id: I35b2034f8aa9574fcb8a34b58ead9f5e9ef27664
316
317
commit cbcdfc5b256e9f85c46052c699dd31e4bf130dd4
318
Merge: 06cfad5 f89b19b
319
Author: Jenkins <jenkins@review.openstack.org>
320
Date: Thu Jul 25 22:04:17 2013 +0000
321
322
Merge "Allow confirm_resize if instance is in 'deleting' status" into stable/grizzly
323
324
commit 06cfad5b46ae5377172597c77fbd82df8583b478
325
Merge: 0b45996 7fae666
326
Author: Jenkins <jenkins@review.openstack.org>
327
Date: Thu Jul 25 21:39:17 2013 +0000
328
329
Merge "Disassociate fixed IPs not known to dnsmasq" into stable/grizzly
330
331
commit a04b312459f21594ced42302397767ba379f2491
332
Author: Nikola Dipanov <ndipanov@redhat.com>
333
Date: Thu Jul 11 15:12:58 2013 +0200
334
335
Don't pass empty image to filter on live migration
336
337
If live migration for a volume backed instance without an image supplied
338
is initiated with no destination supplied, image is set to None.
339
340
This value will cause one of the default filters - the
341
ImagePropertiesFilter to throw an exception in it's host_passes method,
342
as it expects an image to be a dictionary.
343
344
We solve this by leaving out the image entirely before engaging filters.
345
346
This patch is intended for stable/grizzly only, since live migration
347
functionality is being heavily refactored in Havana at the moment and
348
the code from the patch that introduces this bug
349
(Ie683a3ca5d6430c52ead77b41f98fbcb4114ea1e) was
350
accidentally dropped in I6de33ada6dc377e20f8df07da92244f2c150b9fe.
351
352
Fixes bug: 1199811
353
354
Change-Id: I8a5f1b176addbe09c8d8b5f1d1136d05af53ce3e
355
356
commit 1ef0d96be333688b6139168d196028f93bdc2381
357
Author: hartsocks <hartsocks@vmware.com>
358
Date: Tue May 21 20:38:28 2013 -0700
359
360
VNC console does not work with VCDriver
361
362
Introduces get_vnc_console_vcenter as a way to get
363
a modified vnc_console object that connects to the
364
ESXi host underneath vCenter control.
365
366
* adds new fake classes for testing
367
* documents classes, methods & API logic
368
369
fixes bug: #1178369
370
371
(cherry picked from commit 271fc68c1852e3764b7c64d71cd28ac3f803ecba)
372
373
Conflicts:
374
375
nova/virt/vmwareapi/fake.py
376
377
Change-Id: I48430cb9bc9615e02ca9af235f97853f3f0bdafd
378
379
commit c30e03e2aa1acc0207be47b3acc156ddcfd94947
380
Author: Kieran Spear <kispear@gmail.com>
381
Date: Mon Jul 15 22:00:19 2013 +1000
382
383
Fix filtering aggregate metadata by key
384
385
Filtering by metadata key has no effect in the following two functions:
386
387
aggregate_metadata_get_by_host
388
aggregate_host_get_by_metadata_key
389
390
Rely on the explicit join instead of joinedload to load the _metadata
391
relationship in these functions (indicated by the contains_eager
392
option), so that filtering by key still applies when the metadata is
393
collected.
394
395
Fixes bug 1201283.
396
397
Conflicts:
398
399
nova/tests/db/test_db_api.py
400
401
Change-Id: I3e44b5db8d67787bfa14f0e57d30a02a271a8833
402
(cherry picked from commit 7e8fa8f3e6acac0e510cd1fc5af591f9a321d03a)
403
404
commit 5685ea870508b9153179526e28daaf7a9328d373
405
Author: Kieran Spear <kispear@gmail.com>
406
Date: Sun Jul 14 01:08:52 2013 +1000
407
408
Fix aggregate_get_by_host host filtering
409
410
aggregate_get_by_host never actually joins the _hosts relationship,
411
so aggregates with non-matching hosts are still included in the output.
412
413
This change adds the join and improves the tests to cover the bug.
414
415
Fixes bug 1201277.
416
417
Conflicts:
418
419
nova/db/sqlalchemy/api.py
420
nova/tests/db/test_db_api.py
421
422
Change-Id: I87ebfe3de193b9ae47e3a2c8350a49d9cc863fb0
423
(cherry picked from commit 13513a182a54dffb62900a8464529459dd836136)
424
425
commit 0b459966b2c5b9a207ff652406dde78fca299d91
426
Merge: 767c850 7ad1db6
427
Author: Jenkins <jenkins@review.openstack.org>
428
Date: Wed Jul 24 18:23:23 2013 +0000
429
430
Merge "Naming instance directory by uuid in VMware Hyper." into stable/grizzly
431
432
commit 767c85016afe900c0b5d097d306d0129dba0b3f9
433
Merge: 78ebf1a 45fd6be
434
Author: Jenkins <jenkins@review.openstack.org>
435
Date: Wed Jul 24 16:25:51 2013 +0000
436
437
Merge "Fix nic order not correct after reboot" into stable/grizzly
438
439
commit 7ad1db6140c49c85ba643a6217878ef8f3d30463
440
Author: Yaguang Tang <yaguang.tang@canonical.com>
441
Date: Wed Jun 5 15:16:16 2013 +0800
442
443
Naming instance directory by uuid in VMware Hyper.
444
445
Like Libvirt Hyper, Using EC2 template for naming instance
446
directory may cause conflict. we follow what Libvirt does to use
447
instance uuid instead of EC2 template naming. To keep backwards
448
compability, when fails to find a instance by uuid, fall back to use
449
EC2 name.
450
451
Fix bug #1186944
452
453
(cherry picked from commit e517e5668ab20214432e06ef87f7b16776edaf7c)
454
455
Conflicts:
456
457
nova/tests/vmwareapi/db_fakes.py
458
nova/virt/vmwareapi/vmops.py
459
460
Change-Id: Iaba099225a9b3b0390cae4c5fbd82d7e6c11f0e0
461
462
commit 2c1061df8e4ac1a228d79b5ed9c331035fa08d72
463
Author: Yassine Lamgarchal <yassine.lamgarchal@enovance.com>
464
Date: Mon Jun 10 14:47:57 2013 +0200
465
466
Add inst_type parameter
467
468
Add inst_type parameter in order to generate the bandwidth tag into
469
the libvirt configuration.
470
471
Change-Id: I63c1fb51ea637d461755b309c7464f4b4d49a4b2
472
Fixes: bug #1185030
473
(cherry picked from commit 08f8773b9a4fe7fea248b4501230ebd6e6ef2fde)
474
475
commit 78ebf1a9efbf1cacbd1e993fe60b275dc7aff029
476
Merge: f427f01 26ae09c
477
Author: Jenkins <jenkins@review.openstack.org>
478
Date: Tue Jul 23 18:30:42 2013 +0000
479
480
Merge "Return proper error message when network conflicts" into stable/grizzly
481
482
commit f427f01367db031f5899c7a693725cf42b6faa26
483
Merge: 48ddffe 01fd5c2
484
Author: Jenkins <jenkins@review.openstack.org>
485
Date: Tue Jul 23 16:15:17 2013 +0000
486
487
Merge "Fix the filtered characters list from console-log" into stable/grizzly
488
489
commit 11046d94456c63fda51f626e1b816fd57612a4ec
490
Author: Nikola Dipanov <ndipanov@redhat.com>
491
Date: Wed Jul 17 18:34:36 2013 +0200
492
493
Force reopening eventlet's hub after fork
494
495
With this we reopen eventlet's hub after a fork (triggered from
496
websockify when a new client connects) to prevent sharing epoll's fd
497
with the parent, which may cause erratic behaviour.
498
499
This caused novncproxy/spicehtml5proxy to stop working when it had more
500
than two clients connected.
501
502
Fixes bug #1193031
503
504
Change-Id: I3ff9001543b84b1037597da243422490bb611657
505
506
commit 25805554a7d57ff1f40d33104f481845490a787f
507
Author: Nikola Dipanov <ndipanov@redhat.com>
508
Date: Wed Jul 17 18:31:27 2013 +0200
509
510
eventlet.monkey_patch() for spice and vnc proxy
511
512
The VNC proxy and spice apps did not run eventlet.monkey_patch(). These
513
services uses rpc to talk to nova-consoleauth, so eventlet is required.
514
515
Note that this is a stable/grizzly patch only. The issue does not
516
affect master (havana) anymore as it was resolved as a side effect of
517
switching everything in bin/ over to using entry points. The monkey
518
patching is done in nova/cmd/__init__.py now.
519
520
Change-Id: I61cb1212d283e41e6fda88a9521a9e54143af1fb
521
522
commit 6df1b7a2a1413a98bffc8b8e0b947f3c90e3bbf5
523
Author: Russell Bryant <rbryant@redhat.com>
524
Date: Thu Jun 27 21:00:05 2013 +0000
525
526
Make flavors is_public option actually work
527
528
When you create a flavor, you can set an is_public flag to be True or
529
False. It is True by default. When False, the intention is that the
530
flavor is only accessible by an admin, unless you use the flavor_access
531
API extension to grant access to specific tenants.
532
533
Unfortunately, the only place in the code where this was being enforced
534
was when listing flavors through the API. It would filter out the
535
non-public ones for a non-admin. Otherwise, the flavor was accessible.
536
You could get the details, and you could boot an instance with it, if
537
you figured out a valid flavor ID.
538
539
This patch adds enforcement down in the db layer. It also fixes one
540
place in the API where the context wasn't passed down to enable the
541
enforcement to happen.
542
543
Fix bug 1194093.
544
545
(cherry picked from commit b65d506a5f9d9b2b20777a9aceb44a8ffed6a5de)
546
547
Conflicts:
548
nova/api/openstack/compute/contrib/flavor_access.py
549
nova/api/openstack/compute/contrib/flavormanage.py
550
nova/api/openstack/compute/flavors.py
551
nova/compute/api.py
552
nova/db/sqlalchemy/api.py
553
nova/tests/api/openstack/compute/contrib/test_flavor_access.py
554
nova/tests/api/openstack/compute/contrib/test_flavor_disabled.py
555
nova/tests/api/openstack/compute/contrib/test_flavor_manage.py
556
nova/tests/api/openstack/compute/contrib/test_flavor_rxtx.py
557
nova/tests/api/openstack/compute/contrib/test_flavor_swap.py
558
nova/tests/api/openstack/compute/contrib/test_flavorextradata.py
559
nova/tests/api/openstack/compute/test_flavors.py
560
nova/tests/db/test_db_api.py
561
562
Change-Id: I5b37fa0bb19683fe1642fd81222547d4a317054e
563
564
commit 48ddffe52c4e0f38420f39482b11a9b3921388eb
565
Merge: 378c045 ce789f7
566
Author: Jenkins <jenkins@review.openstack.org>
567
Date: Sun Jul 21 10:26:02 2013 +0000
568
569
Merge "Refresh volume connections when starting instances" into stable/grizzly
570
571
commit e301ab416f68db0ad05a7f975a43731454b892ed
572
Author: Matt Riedemann <mriedem@us.ibm.com>
573
Date: Fri Jul 5 11:18:04 2013 -0700
574
575
Update instance.node on evacuate
576
577
The evacuate API works like a migration from a dead node. As part of
578
migration, the instance.node property is updated to reflect the target
579
node that the instance is migrating to. This is later used when the
580
migration is confirmed/reverted. The problem is when an instance is
581
being evacuated, the node property isn't getting updated and the
582
instance will have the old (dead) node value set on it after it's
583
evacuated. This patch updates instance.node when the instance is
584
evacuated.
585
586
Fixes bug 1198284
587
588
Change-Id: I4901fb937160e955df7c97c37265783e844fdbef
589
(cherry picked from commit 3da3dfb733237ed2c908d8770961d8d6656ea1a3)
590
591
commit 45fd6be08a04bc49dbc5c7fd67cdc7defb889284
592
Author: Aaron Rosen <arosen@nicira.com>
593
Date: Mon Jul 15 17:32:43 2013 -0700
594
595
Fix nic order not correct after reboot
596
597
Previously if an instance was rebooted or migrated the original nic order
598
would not be maintained. This patch retrieves the correct nic order from
599
info_cache so that the order is mantained.
600
601
Fixes bug 1187092
602
603
Change-Id: I3f5488cc14c692fde7d5bd3376389e2f2f5550ff
604
(cherry picked from commit bcd12a776e8cde8c4b64e7c1a21a7387ece55b31)
605
606
commit 378c045b1d40a4af8330eb7ece3dcda1620a578c
607
Merge: 008a68a fa6c2d4
608
Author: Jenkins <jenkins@review.openstack.org>
609
Date: Mon Jul 15 21:54:51 2013 +0000
610
611
Merge "Call virt.driver.destroy before deallocating network." into stable/grizzly
612
613
commit 008a68a77db5655b0a164b79ceb7b821eb1d084b
614
Merge: 265b868 3546cde
615
Author: Jenkins <jenkins@review.openstack.org>
616
Date: Fri Jul 12 06:13:24 2013 +0000
617
618
Merge "Support FlatDHCP network for VMware Hyper." into stable/grizzly
619
620
commit 7fae6661b894596b45ec31e70627b76a643276d8
621
Author: Yufang Zhang <yufang521247@gmail.com>
622
Date: Thu May 16 19:42:36 2013 +0800
623
624
Disassociate fixed IPs not known to dnsmasq
625
626
Currently, fix_ip disassociation only happens in release_fixed_ip(),
627
which is called by dhcp-bridge script when ip is released. But there
628
are circumstances when instancs fail to get lease from dnsmasq(instances
629
fail to boot or instances fail to spawn). In such case, DHCPRELEASE
630
packet sent to dnsmasq would never trigger dhcp-bridge to run. Thus the
631
fixed ip would not be disassociated from instance until
632
_disassociate_stale_fixed_ips() begins to run.
633
634
Actually, this patch borrows the idea from Vishvananda's
635
3f6739e2a8acdf20d01c0b5039a33f26d879587b commit: explicitly disassociate
636
fixed ip which is not known to dnsmasq, instead of via the dhcp-release
637
script.
638
639
Steps to reproduce:
640
1. Create an instance which doesn't send dhcp request to dnsmasq.
641
2. Delete the instance.
642
3. The fixed ip associated with the instance is not returned to pool.
643
644
Bug 1180766
645
646
Change-Id: Icc022ea2fb7af08fb7354e3a49c77ad5ae58f34c
647
(cherry picked from commit 383ca6cf5ff9172a43339dbf845245fbdf2f8210)
648
649
commit d1edaa9551247ae4ccac0b5a9dfb27d5adecff8d
650
Author: Wangpan <hzwangpan@corp.netease.com>
651
Date: Thu Jun 13 15:19:21 2013 +0800
652
653
Don't deallocate network if destroy time out
654
655
Fixes bug #1169057
656
Sames as bug 1180178, if libvirt returns time out error during destroy
657
an instance, we also can't deallocate the network of the instance,
658
otherwise the fixed ip of it may be re-used by another new instance,
659
and resulted in the new instance can't be accessed by the fixed ip.
660
661
Change-Id: I04dfd21bb1b7168abc71f39e836126a11dad6d8d
662
(cherry picked from commit d925044e09509495e7474d5f625bbd45d8985afa)
663
664
commit d35e7488aaf9b74b38c07639e2031fc95e3acb55
665
Author: David Scannell <dscannell@gridcentric.com>
666
Date: Mon Jun 10 13:04:55 2013 -0400
667
668
Avoid redefining host to none in get_instance_nw_info(...)
669
670
The host parameter is required and a part of the method signature.
671
When it is redefined from kwargs.get('host') the supplied value
672
gets redefined to None. This removes the redefinition and keeps the
673
parameter value intact.
674
675
Fixes bug #1194178
676
677
Change-Id: I779cbce909935a4b794c16cefb0128f16f8563fb
678
(cherry picked from commit b0a2a2d29f24cea15d866f352d4beae5da965605)
679
680
commit 265b8681c0f1145797d517d47e5bb0654776b2d9
681
Merge: 27960ef d21c3e6
682
Author: Jenkins <jenkins@review.openstack.org>
683
Date: Wed Jul 10 18:40:19 2013 +0000
684
685
Merge "Handle instance directories correctly for migrates." into stable/grizzly
686
687
commit 01fd5c212ae80318aacd60d4e1a36f91c78867f7
688
Author: Xavier Queralt <xqueralt@redhat.com>
689
Date: Sun Jun 23 21:39:32 2013 +0200
690
691
Fix the filtered characters list from console-log
692
693
In fffcaea1 ^M was added to the list of characters to filter from the
694
console-log output but, because of a typo in the regexp, the
695
character '-' was added to that list too and has been removed since
696
then.
697
698
Fixes bug #1194032
699
700
Change-Id: Ia02b0b4c7027cbd09a51042ca1b5d28890413626
701
(cherry picked from commit c651f4928c65407cff86a9f9efc046a0f76a9519)
702
703
commit 27960efb0618bdb19eb32834fb1aaae26f2c01f8
704
Merge: 3ba31f2 d2cb4b2
705
Author: Jenkins <jenkins@review.openstack.org>
706
Date: Tue Jul 9 13:01:57 2013 +0000
707
708
Merge "Sync RPC module from Oslo" into stable/grizzly
709
710
commit eff629fa4cfdc68e670d70a4a38810c4935f81af
711
Author: Rafi Khardalian <rafi@metacloud.com>
712
Date: Thu Jun 13 05:09:45 2013 +0000
713
714
Fix resizes with attached file-based volumes
715
716
Bug 1190364
717
718
get_instance_disk_info() now takes an additional, optional argument
719
of block_device_info. When passed in, we use it to filter out
720
volumes, such that they are not returned. Previously, any devices
721
classified as "file" as parsed from the instance XML were considered
722
local disks. This assumption is incorrect, since some volumes are
723
simply files, rather than block devices.
724
725
Added a new test, based almost entirely off the existing test case
726
for this function, which represents an XML file with two fake
727
file-based (NFS-like) volumes and associated block_device_info to
728
claim those volumes.
729
730
Change-Id: I4809a8fd68eb0f709347aa7cad0fa85389ade41c
731
(cherry picked from commit e71081b856b06513cc1e56a9adb2a619967f0eae)
732
733
commit fbde109390f32204f6355d937449f0f54f07ffd7
734
Author: Chang Bo Guo <guochbo@cn.ibm.com>
735
Date: Tue Jul 2 19:47:56 2013 -0700
736
737
Remove junk file when ftp transfer failure
738
739
During spawn with powervm driver, the image transfer to the IVM host
740
may fail in some conditions, eg. IVM host doesn't have enough disk
741
space. It leaves the junk file in IVM host, and this will disturb
742
next spawn after resolving environment issue, need clean it.
743
744
Fixes bug #1188884
745
746
Change-Id: I6347dec4cfda61bed0960159e8a3402fcffa296b
747
748
commit 3ba31f29df0c153a041e4f5ac8a5dbd45a9f941d
749
Merge: 89150b2 d367ab6
750
Author: Jenkins <jenkins@review.openstack.org>
751
Date: Tue Jul 2 19:29:42 2013 +0000
752
753
Merge "Skip security group code when there is no network." into stable/grizzly
754
755
commit 89150b221ec4f3e2ba175207fe98d65be26d3b43
756
Merge: 0e0d615 19fbc29
757
Author: Jenkins <jenkins@review.openstack.org>
758
Date: Tue Jul 2 19:29:15 2013 +0000
759
760
Merge "Silence exceptions from qpid connection.close() (from oslo)" into stable/grizzly
761
762
commit 0e0d615bc6db993d0f63e01bda2bb209dcf9e709
763
Merge: b92075c 69dd7c5
764
Author: Jenkins <jenkins@review.openstack.org>
765
Date: Tue Jul 2 19:10:35 2013 +0000
766
767
Merge "Regenerate missing resized backing files" into stable/grizzly
768
769
commit d367ab62720988231524accab33488ed1d6c8555
770
Author: Brent Eagles <beagles@redhat.com>
771
Date: Fri Jun 14 13:21:54 2013 -0230
772
773
Skip security group code when there is no network.
774
775
This patch adds a simple check to the nova/quantum API that immediately
776
returns an empty list if there are no configured networks available.
777
778
Fixes bug 1191044
779
780
Change-Id: I34ac4dd56c71f99c7ed9ff632d6dc7036d91008d
781
(cherry picked from commit 8f3b3db9b04bb5ff7b9486dc6fc211745b983855)
782
783
commit b92075c9b73b568e99b61544286da506bfb991a4
784
Author: Édouard Thuleau <edouard.thuleau@cloudwatt.com>
785
Date: Tue Jun 11 22:32:49 2013 +0200
786
787
Quantumapi returns an empty network list
788
789
When nova retries network list and if the list is empty, quantumapi
790
should return a empty list instead of an empty dictionnary.
791
792
Fixes bug 1189976
793
794
Conflicts:
795
796
nova/tests/network/test_quantumv2.py
797
798
Change-Id: I5b8c498a6c86e92d0de14d8de9e55722408c5669
799
(cherry picked from commit 14d3ae58c1ec39edd1b0b2a2a1947c5f8e96bab5)
800
801
commit ce789f7f62ad932fb3833ec6c69bfb3fdaadcbc6
802
Author: Brent Eagles <beagles@redhat.com>
803
Date: Thu Jun 6 15:33:42 2013 -0230
804
805
Refresh volume connections when starting instances
806
807
This patches adds network and block device information to the signature
808
of the power_on method on the virtualization driver. The relevant call
809
sites are also modified to provide the required information. The libvirt
810
implementation of power_on has been altered to re-establish network and
811
volume related connections.
812
813
Fixes bug: 1188326
814
815
(cherry picked from commit db3989586a8d5bbbcf857b9294a124ecc5fc57e8)
816
817
Conflicts:
818
nova/tests/test_vmwareapi.py
819
nova/virt/baremetal/driver.py
820
nova/virt/libvirt/driver.py
821
nova/virt/vmwareapi/driver.py
822
nova/virt/vmwareapi/vmops.py
823
824
(Includes some collateral additions for console tests in the vmware
825
tests.)
826
827
Change-Id: If617b570e082e3aa321414a2680a3aa0754f6153
828
829
commit 1066418fcbc0adfbf17687e90973ac39529c69e3
830
Merge: fc98015 4bb6263
831
Author: Jenkins <jenkins@review.openstack.org>
832
Date: Wed Jun 26 03:46:50 2013 +0000
833
834
Merge "correctly set iface-id in vmware driver" into stable/grizzly
835
836
commit fc98015be44e1557e6208a8b27651e7f8ca052a4
837
Merge: a8dab9f 7825e39
838
Author: Jenkins <jenkins@review.openstack.org>
839
Date: Tue Jun 25 01:41:07 2013 +0000
840
841
Merge "Fix bug where consoleauth depended on remote conductor service" into stable/grizzly
842
843
commit f89b19b96bb6dab266c741d39f637dd66eb1c8be
844
Author: Aditi Raveesh <aditi.raveesh@gmail.com>
845
Date: Wed Jun 12 11:30:47 2013 +0530
846
847
Allow confirm_resize if instance is in 'deleting' status
848
849
Deleting an instance puts it into 'deleting' task_state.
850
If the instance is in 'resized' vm_status, this tries a
851
confirm_resize on the instance. confirm_resize currently expects
852
the task_state to be None. It should instead expect
853
None/deleting task_state.
854
855
Fixes bug: 1189391
856
857
Conflicts:
858
nova/tests/compute/test_compute.py
859
860
Change-Id: I6a3f82eab9ed2242bfd38bc1ac760dbb4b3d8c4a
861
(cherry picked from commit 0fed513b695cd15aaca24d24926a1c7d13e63509)
862
863
commit 4bb6263e91e27f6023ff05cf718a648d340a4143
864
Author: Dan Wendlandt <dan@nicira.com>
865
Date: Sat May 25 14:22:23 2013 -0700
866
867
correctly set iface-id in vmware driver
868
869
bug 1183452
870
871
Commit 70c659059b8fbef811ce79700aecb01c60242ebd updated the vmware
872
driver to use the new vif model, but was incorrect in how it tried to
873
grab the iface-id and set it in the vmx file spec. This patch fixes
874
that issue and adds a check for this in the existing unit tests.
875
876
Change-Id: I756539871d1844a828f9a0a295fc9fa3e59610f7
877
(cherry picked from commit c028e54f6be75d1cd335550cf5291ddf94f973e3)
878
879
commit 19fbc29754f72de6c178ba008f2b330ac24ef1c7
880
Author: David Ripton <dripton@redhat.com>
881
Date: Mon Jun 3 09:23:13 2013 -0400
882
883
Silence exceptions from qpid connection.close() (from oslo)
884
885
This is commit 15d8d698b7 in oslo-incubator.
886
887
Fixes bug 1185465
888
889
Don't log exceptions that happen during connection.close(). They're
890
not surprising (a closed socket can't be read or written), don't
891
require sysadmin attention, and provide red herrings in the logs
892
to distract from finding real root causes of problems.
893
894
Change-Id: Ia0e25fbc11094b5ad5d6d2cf7d0e786acb3df1f6
895
(cherry picked from commit 361d116d820939207a0b01399c3ce50c7bc30dcd)
896
897
commit a8dab9fe6912f1a355b8cb4486960dd8b6eace7a
898
Merge: 19c443b ee75c29
899
Author: Jenkins <jenkins@review.openstack.org>
900
Date: Fri Jun 21 02:54:56 2013 +0000
901
902
Merge "Clean up failed image transfers in instance spawn" into stable/grizzly
903
904
commit c17e0c1d8f15854c2ef5cf3aadb1a89e8db55ecf
905
Author: Hans Lindgren <hanlind@kth.se>
906
Date: Sun Jun 2 00:29:59 2013 +0200
907
908
Fix a race where a soft deleted instance might be removed by mistake
909
910
When a soft deleted instance is restored there is a tiny risk that it
911
might be deleted if the reclaim periodic task runs at the same time.
912
This happens because the restore operation sets deleted_at to None
913
which makes the reclaim job think it is ok to delete the instance.
914
915
This can be resolved by restricting reclaims to only those instances
916
whose task_state is None.
917
918
Resolves bug 1186243.
919
920
Conflicts:
921
nova/compute/manager.py
922
nova/db/sqlalchemy/api.py
923
nova/tests/compute/test_compute.py
924
925
Change-Id: Ia138e2d504b7482c46900583f019a32c70513245
926
(cherry picked from commit cdf10381caad217ae9eb99c69a4226e40ba81db4)
927
928
commit 9d783b648878e386265cef5b82a931ed677ded43
929
Author: Xavier Queralt <xqueralt@redhat.com>
930
Date: Mon Jun 10 16:06:40 2013 +0200
931
932
Don't make nova-compute depend on iSCSI
933
934
Because not all volume drivers require iSCSI it makes sense to not
935
set the initiator in the connector and let the volume drivers
936
return an error if iSCSI is required.
937
938
bug #1007084
939
940
Change-Id: I5da472e5013fdc0c733ecb12b919c1068873bdc6
941
(cherry picked from commit 3ba0c171c0b6f3516662bc0ab660b14cd6d34d8a)
942
943
commit 19c443bc58f5cdc9623bc6f7e41d9c4cf588a4aa
944
Author: Yaguang Tang <yaguang.tang@canonical.com>
945
Date: Fri May 24 11:50:43 2013 +0800
946
947
Fix VMware Hyper can't honor hw_vif_model image property.
948
949
Support setting instance vif model through image property,
950
change default vif model to e1000. Available vif model options
951
for VMware Hyperv is "VirtualE1000, VirtualPCNet32, VirtualVmxnet".
952
Also this patch removes the unused list_interface method and tests.
953
954
DocImpact
955
Fix bug #1183192
956
Fix bug #1187251
957
958
Change-Id: I3a98bc26df7e2252bf043a71acbba2a158e213b8
959
(cherry picked from commit 088995fc2c792d6d029abd0e45d5c0cf09a5125a)
960
961
commit d2cb4b25f7226cd2051d38905942e515d257931c
962
Author: Ben Nemec <bnemec@us.ibm.com>
963
Date: Tue Jun 18 16:59:04 2013 +0000
964
965
Sync RPC module from Oslo
966
967
Includes backport of change for bug 1175808
968
Fixes bug 1172922 and bug 1122763
969
970
Change-Id: Ib52e9458a9db01e2c8fd2a6d84010733f85980f8
971
972
commit 3f6060693691661ca54cfc30297433b9cffcd612
973
Merge: 8849413 03c2429
974
Author: Jenkins <jenkins@review.openstack.org>
975
Date: Tue Jun 18 21:10:31 2013 +0000
976
977
Merge "Volume IO usage gets reset to 0 after a reboot / crash" into stable/grizzly
978
979
commit 884941368c5e8045905af38fb1f8477363d98b13
980
Merge: ef10eee 7726dae
981
Author: Jenkins <jenkins@review.openstack.org>
982
Date: Tue Jun 18 20:18:44 2013 +0000
983
984
Merge "set ERROR state when scheduler hits max attempts" into stable/grizzly
985
986
commit ef10eee92fdb7d4c584e1382bb1a3cbfaa3aed28
987
Merge: 7422d83 5c4edd2
988
Author: Jenkins <jenkins@review.openstack.org>
989
Date: Tue Jun 18 20:18:23 2013 +0000
990
991
Merge "Don't update API cell on get_nwinfo" into stable/grizzly
992
993
commit 7422d832ce38ff73f80d845cabf7e6d063a5609f
994
Merge: 49ab0e8 8cd3ed2
995
Author: Jenkins <jenkins@review.openstack.org>
996
Date: Tue Jun 18 20:09:13 2013 +0000
997
998
Merge "Fix powervm driver resize instance error" into stable/grizzly
999
1000
commit 49ab0e80cac7885eee6c02c148432a4d248e57a4
1001
Merge: e71247c a16db70
1002
Author: Jenkins <jenkins@review.openstack.org>
1003
Date: Tue Jun 18 20:08:31 2013 +0000
1004
1005
Merge "Catch InstanceNotFound in instance_actions GET" into stable/grizzly
1006
1007
commit e71247ca9f0ebd66968e0a116357b8ce383442ed
1008
Merge: 8e6d4b6 b7ff9e7
1009
Author: Jenkins <jenkins@review.openstack.org>
1010
Date: Tue Jun 18 20:07:53 2013 +0000
1011
1012
Merge "Make nova-api use servicegroup.API.service_is_up()." into stable/grizzly
1013
1014
commit 8e6d4b62b97f33ad3c57a87e26b94bea8ac9f73c
1015
Merge: 60774bd 078bbac
1016
Author: Jenkins <jenkins@review.openstack.org>
1017
Date: Tue Jun 18 09:57:30 2013 +0000
1018
1019
Merge "Don't snat all traffic when force_snat_range set" into stable/grizzly
1020
1021
commit 69dd7c50a95518788018dd1cb8aad60029fdac42
1022
Author: Rafi Khardalian <rafi@metacloud.com>
1023
Date: Wed May 29 21:26:19 2013 +0000
1024
1025
Regenerate missing resized backing files
1026
1027
Fixes bug 1185588 (for stable/grizzly)
1028
1029
The only way to reliably determine whether we are using a legacy
1030
(pre-Grizzly) backing structure is to look at the backing file
1031
reference for the instance/disk itself. Thus, if we determine the
1032
backing file contains an underscore, we use the value following it
1033
as our legacy_backing_size, which is then used to create the
1034
necessary backing file.
1035
1036
Change-Id: I145f71068c972a604e5361ae79d6315286136786
1037
Cherry-Pick: 6e768363450774cfee90e41aa5c40af780d3e04a
1038
1039
commit 60774bd58b7ab9e95e0ca71acb2f473dea9e1a67
1040
Author: Aaron Rosen <arosen@nicira.com>
1041
Date: Fri May 31 16:17:58 2013 -0700
1042
1043
Fix nova-compute fails to start if quantum is down
1044
1045
When nova-compute starts it needs to check the state of instances that
1046
were previously running and needs to query quantum. Previously, if
1047
it queried quantum and quantum raised an error or was unreachable
1048
nova-compute would terminate. This patch adds a retry loop to prevent
1049
this from happening.
1050
1051
Fixes bug 1186357
1052
1053
Change-Id: I0d865e464f7bf60a0b6dc4ca8a42436d16e29790
1054
(cherry picked from commit 6fda12bb14ab91592910f9ebbde9ef9f8d588b08)
1055
1056
commit d34d4cacf7b20f72c67f7873dcf2c372abc60ecd
1057
Author: Rafi Khardalian <rafi@metacloud.com>
1058
Date: Sun May 5 22:18:33 2013 +0000
1059
1060
Make resize/migrated shared storage aware
1061
1062
Fixes bug 1177247 (for stable/grizzly)
1063
1064
Added some logic to check for whether or not we are on a shared
1065
filesystem and set shared_storage accordingly. We perform similar
1066
checks in other areas of the code, typically through RPC calls.
1067
However, all the resize/migrate code is slated to be refactored for
1068
Hava, so the idea was to keep this patch as minimally intrusive as
1069
possible.
1070
1071
When shared_storage is true, we pass that on to the cleanup call
1072
so that it no longer executes an rm via SSH, which was ultimately
1073
destroying the original instance directory.
1074
1075
Change-Id: Ie9decedd373c000211c171df64e1e96fe78e5081
1076
Cherry-Pick: 9290bddd9f270d8ea4fbd6d953a8634473979cd5
1077
1078
commit fa6c2d4def22f0ddd296572d9ce5b3b6f7267e2f
1079
Author: Melanie Witt <melwitt@yahoo-inc.com>
1080
Date: Thu May 23 00:23:16 2013 +0000
1081
1082
Call virt.driver.destroy before deallocating network.
1083
1084
Call virt driver destroy before releasing the instance's
1085
ip address, so that certain failures can be handled.
1086
For instance, if destroy raises InstancePowerOffFailure,
1087
do not deallocate network resources.
1088
1089
Fixes bug 1180178
1090
1091
Change-Id: I9bde984d2393bd73a73eae6065defb2ed7c01492
1092
(cherry picked from commit dacb187fcd685c15fd084fdfce8c5e165130ff54)
1093
1094
commit 61fc52948f26656c4bec4a09ee3daa16153f991b
1095
Author: Aaron Rosen <arosen@nicira.com>
1096
Date: Sat Jun 8 19:58:58 2013 -0700
1097
1098
Fix assumed port has port_security_enabled
1099
1100
Previously if adding a security group to an instance it would fail
1101
if the port did not contain port_security_enabled=True. This patch
1102
also changes the unit tests so that networks by default are not
1103
associated with port_security_enabled=True because this is only true
1104
if a plugin is running the port_security extension.
1105
1106
Fixes bug 1189059
1107
1108
Change-Id: Ieebb7aafba01c18164b86a01a3c289a6957321e4
1109
(cherry picked from commit eddab7ba0e45c47fa8503b29484ce296f46c0d27)
1110
1111
commit d21c3e67ae0965304f2179be14fc16319eb2da9f
1112
Author: Michael Still <mikal@stillhq.com>
1113
Date: Wed May 8 15:28:40 2013 +1000
1114
1115
Handle instance directories correctly for migrates.
1116
1117
In Grizzly we changed how we name instance directories. Unfortunately
1118
the decision as to what directory name to use is based on whether
1119
there is a pre-existing directory on the compute node with the old
1120
name.
1121
1122
In the case of live migration, that doesn't work because we have
1123
libvirt copy the directory across for us. So, we need to pass
1124
around the directory name in the migration request. This is better
1125
anyway, as it means that ops types don't need to handle an instance
1126
directory magically renaming itself at some point.
1127
1128
Resolves bug 1175286.
1129
1130
Change-Id: Ib102e644a8742726fa2049303354d466b8109e0b
1131
(cherry picked from commit 5c11a8c537eb4349e02e8f42468b441c259fbfc2)
1132
1133
commit 7825e39bdcf3d1477e97443bdb1c987f841cd303
1134
Author: Sam Morrison <sorrison@gmail.com>
1135
Date: Fri May 31 15:03:25 2013 +1000
1136
1137
Fix bug where consoleauth depended on remote conductor service
1138
1139
Fixes bug 1186123
1140
1141
Conductor API was being imported directly so ignoring use_local FLAG
1142
1143
Change-Id: I055a5301d4c18d201d67f114047f503114751603
1144
(cherry picked from commit 6c6e43e466a5383b3ef3c852c9b2749f0fad64dc)
1145
1146
commit b7ff9e7e50a53950daffe24ac94721f4bb7676ed
1147
Author: Akira Yoshiyama <a-yoshiyama@bu.jp.nec.com>
1148
Date: Sun Apr 14 11:39:55 2013 +0000
1149
1150
Make nova-api use servicegroup.API.service_is_up().
1151
1152
Fix bug 1168861
1153
1154
Change-Id: I38e06c8832ec054755daa70898a303383557041b
1155
(cherry picked from commit 616e5d965bd8a2923f00ba4ff194e6a6ee2468cd)
1156
1157
commit 8e6edd2fe9845ee0d93600ebfcf90ce029bd1f70
1158
Author: Adam Gandelman <adamg@canonical.com>
1159
Date: Thu Jun 6 12:09:02 2013 -0700
1160
1161
Bump stable/grizzly next version to 2013.1.3
1162
1163
Change-Id: Ie9b5fb202effaa2d42270d64642369adf7b11840
1164
1165
commit 078bbac53aab5a71e9901c71f86f3065f2570ccf
1166
Author: Vishvananda Ishaya <vishvananda@gmail.com>
1167
Date: Mon May 20 09:39:59 2013 -0700
1168
1169
Don't snat all traffic when force_snat_range set
1170
1171
When using /proc/sys/net/bridge/bridge-nf-call-iptables=1, bridged
1172
traffic gets snatted even though it is being bridged out to an
1173
external gateway. This can lead to asymmetric routes where traffic
1174
goes out the fixed network and comes back on the float network
1175
and can lead to packets being blocked by firewalls. Work around
1176
this problem by only fallback snatting when the traffic is
1177
going to one of the force_snat_ranges. If force_snat_range is not
1178
set it defaults to the existing behavior of snatting all traffic
1179
that doesn't hit an earlier rule.
1180
1181
Fixes bug 1182143
1182
1183
Change-Id: If496dad2308a68a6a4a9fd43d695943bc5034ac4
1184
(cherry picked from commit 769782be844e4c74fdc0aad0ff704b06c874c3ad)
1185
1
1186
commit 8c136a8ee4733d8d3af61406fc9eeccb4e9b7a68
2
1187
Author: Chris Behrens <cbehrens@codestud.com>
3
1188
Date: Wed May 29 14:21:00 2013 +0000
118
1303
119
1304
Merge "The vm_state should not be modified until the task is complete." into stable/grizzly
120
1305
1306
commit 5c4edd24d4cd2883ad7619aa632e7556b82a8dab
1307
Author: Chris Behrens <cbehrens@codestud.com>
1308
Date: Wed May 15 09:23:26 2013 +0000
1309
1310
Don't update API cell on get_nwinfo
1311
1312
Every call to get_instance_nw_info() updates an instance's info_cache.
1313
The purpose of this is to make sure we're constantly healing the
1314
info_cache table. However, this causes a lot of unneeded updates
1315
traveling up to the API cell. Cells has its own periodic task that
1316
heals all instance data.
1317
1318
Fixes bug 1180304
1319
1320
Conflicts:
1321
nova/network/api.py
1322
nova/tests/cells/test_cells_messaging.py
1323
1324
Change-Id: I07774712a6358999fd1fdd0f5a90c6f05e3d8d6e
1325
(cherry picked from commit 4251bba3bd456f7ba1c7481bb84d839a5fafc757)
1326
1327
commit ee75c2970808e53bb849802d6ade2f9df2d76e0b
1328
Author: Chang Bo Guo <guochbo@cn.ibm.com>
1329
Date: Wed May 29 06:18:39 2013 -0700
1330
1331
Clean up failed image transfers in instance spawn
1332
1333
Boot one instance with powervm driver, will transfer the image
1334
file to host. Need clean up failed image transfers that may occur
1335
during instance spawn.
1336
1337
Fixes bug #1173017
1338
(cherry picked from commit 68cc1cd511298208dde59878bfbef474a3625c9f)
1339
1340
Change-Id: Id06fa84bee63adc4af96a339526d75540978263f
1341
1342
commit 7726dae1c3720477191cfdb239fdd2c5d0952285
1343
Author: Ryan Moore <ryan.moore@hp.com>
1344
Date: Mon May 20 14:56:55 2013 +0100
1345
1346
set ERROR state when scheduler hits max attempts
1347
1348
Presently when scheduler raises NoValidHost due to max attempts
1349
being reached, the instance remains in a build state.
1350
Exception handler for NoValidHost in manager.run_instance() needs
1351
request_spec[instance_uuids] to know which host to put into an
1352
error state in _set_vm_state_and_notify().
1353
schedule_run_instances() was popping instance_uuids from the
1354
request_spec prior to a call to _schedule().
1355
Changed pop of instance_uuids prior to call to _schedule() to be a get.
1356
Added pop of instance_uuids to beneath call to _schedule() as
1357
individual creates do not need them.
1358
1359
Conflicts:
1360
nova/scheduler/filter_scheduler.py
1361
1362
Change-Id: I9654820e01d5611763e9e673f15f46b947d09e6d
1363
Fixes: bug #1182056
1364
(cherry picked from commit aefc28dd481354edd0f3b5aec18db006680b2ffe)
1365
1366
commit a16db707c1722c87f29eacd20d5d359c37968c3f
1367
Author: Chris Yeoh <cyeoh@au1.ibm.com>
1368
Date: Wed May 22 21:52:25 2013 +0930
1369
1370
Catch InstanceNotFound in instance_actions GET
1371
1372
Catch the InstanceNotFound exception in the instance_actions
1373
index function and handle the error gracefully so a stacktrace is
1374
not generated in the log files. This patch does not change the
1375
API
1376
1377
Fixes bug 1182867
1378
1379
Change-Id: I0764f8abda7176deefe0d103bd86acd0af78780d
1380
(cherry picked from commit 8eede5e9814f3305e0bb66eeb93623ba25b16e9a)
1381
1382
commit 26ae09c6b49c202d65df401b1f6353329d006c63
1383
Author: Haiwei Xu <xu-haiwei@mxw.nes.nec.co.jp>
1384
Date: Mon Apr 1 16:48:56 2013 +0900
1385
1386
Return proper error message when network conflicts
1387
1388
Fixes bug 1158075
1389
When creating a network which is already existing, we got
1390
an internal server error which can't show the error reason.
1391
This patch returns conflict error(HTTP 409) instead of
1392
internal server error(HTTP 500), so that users can know
1393
the error reason from the error message.
1394
1395
Change-Id: I9e39a7152e37756b34b1c8d77de057012743f0f7
1396
(cherry picked from commit 8a80c4c0968a9deceaa6d99629c6d7f7889769e5)
1397
121
1398
commit ec335d00c37421c0ae4397e017b7872d62184723
122
1399
Author: Aaron Rosen <arosen@nicira.com>
123
1400
Date: Tue May 21 21:15:30 2013 -0700
205
1482
Change-Id: I67d63a7aef699b555897582c74b04e857db7aafb
206
1483
(cherry picked from commit 4465b705fc7cb7a150a7f8d6f70ccf6c20ff3284)
207
1484
1485
commit 03c2429c3cf35c09941888c74d1a8dd47f37ad4e
1486
Author: Michael Kerrin <michael.kerrin@hp.com>
1487
Date: Tue Apr 16 09:43:28 2013 +0000
1488
1489
Volume IO usage gets reset to 0 after a reboot / crash
1490
1491
Check for reset of block device stats before updating volume usage cache.
1492
1493
If an instance is reboot or crashes, the results of block device stats for
1494
that domain get reset. Add check that if block device stats are lower than
1495
the current stats in the database, then add current values to totals and
1496
update.
1497
1498
Change-Id: I3011872f8d80a1519771f8423001650dcbcc7f5b
1499
Fixes bug: 1169097
1500
(cherry picked from commit a4cb456631f0e70685fbc361eced9baa477a8c56)
1501
208
1502
commit 605ea6fc5fd311fac26b7260c118b71ad1dfd6cb
209
1503
Author: Dongdong Zhou <dzhou121@gmail.com>
210
1504
Date: Wed Apr 17 12:54:14 2013 +0100
323
1617
Change-Id: I4b42a9a764f6e15bb94a383e8a725d52d7436bd7
324
1618
(cherry picked from commit dc887267572c821ce97bec9536a699b72bdae443)
325
1619
1620
commit 3546cdeb0548e03e71004914b2ab651a316fe047
1621
Author: Yaguang Tang <yaguang.tang@canonical.com>
1622
Date: Tue Apr 23 18:55:19 2013 +0800
1623
1624
Support FlatDHCP network for VMware Hyper.
1625
1626
Auto create bridge when using FlatDHCP network, Add unittests for vif.
1627
1628
Fix bug #1171783
1629
Fix bug #1177635
1630
1631
Change-Id: Ibb75c8e3174283f0f5e94bacfd61ee93e8bd6227
1632
(cherry picked from commit bc66b8aa968a0c0b74e2f21ed0fc856f8933c6d3)
1633
326
1634
commit 159fdd2dc3feb35d331ecf3d2899df1b481c76dc
327
1635
Merge: b8073a2 3c84539
328
1636
Author: Jenkins <jenkins@review.openstack.org>
507
1815
508
1816
Change-Id: Ie683a3ca5d6430c52ead77b41f98fbcb4114ea1e
509
1817
1818
commit 8cd3ed2adb531d2a8db4d281fd3f876130ef483c
1819
Author: guohliu <guohliu@cn.ibm.com>
1820
Date: Wed Apr 17 18:22:57 2013 +0800
1821
1822
Fix powervm driver resize instance error
1823
1824
During a resize operation in powervm, the logical volume
1825
of the instance will be copied and compressed to a file.
1826
Current logic decompresses it twice before copying this file
1827
to a new volume which will cause the new instance to not
1828
boot properly.
1829
1830
Fixes bug #1169848
1831
1832
Change-Id: I91f536919511aafbf7dbe14abbee48dbceb189b2
1833
(cherry picked from commit 5643bd5d9bf17335e109d4b2d93e161f4d12c6f9)
1834
510
1835
commit 0f6deca9dc548c6c76373bf0c062b88f1ff68260
511
1836
Merge: b330a07 1ffab28
512
1837
Author: Jenkins <jenkins@review.openstack.org>
1033
2358
Change-Id: Ie991c9452a21cfb036d3955a094ec345805d63d1
1034
2359
(cherry picked from commit 782bab8dd2930a4479d84dcb906321313d06b25b)
1035
2360
2361
commit aa081f36a846bbfb697725a1ebcb06683307e25a
2362
Author: Hans Lindgren <hanlind@kth.se>
2363
Date: Mon Apr 29 23:17:41 2013 +0200
2364
2365
Reset migrating task state for MigrationError exceptions
2366
2367
Create an exception type for pre-check errors and add it to the list
2368
of exceptions for which instance task state is reset to ACTIVE to
2369
avoid cases where an instance get stuck in MIGRATING state.
2370
2371
Convert existing instances of raise MigrationError in scheduler and
2372
driver pre-checks to have them raise the new MigrationPreCheckError
2373
exception instead.
2374
2375
In addition, also make sure that expected exceptions are passed over
2376
RPC silently.
2377
2378
Resolves bug 1171526.
2379
2380
Conflicts:
2381
nova/scheduler/filter_scheduler.py
2382
nova/scheduler/manager.py
2383
nova/virt/xenapi/vmops.py
2384
2385
Change-Id: I00fa0962e555f895dd98861684ec7767fed3b37b
2386
(cherry picked from commit 9eb3dc7ed44d99652a17238724842e3067d46120)
2387
1036
2388
commit 79ad6cbc448d1e596b565aaccdc9dbd6895d231c
1037
2389
Author: Devananda van der Veen <devananda.vdv@gmail.com>
1038
2390
Date: Tue Apr 30 14:34:56 2013 -0700
Loggerhead is a web-based interface for Breezy
Version: 2.0.1