Statistics
| Branch: | Tag: | Revision:

root / NEWS @ f39d39b6

History | View | Annotate | Download (68.8 kB)

1
News
2
====
3

    
4
Version 2.5.0 rc3
5
-----------------
6

    
7
*(Released Wed, 26 Oct 2011)*
8

    
9
Incompatible/important changes and bugfixes
10
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
11

    
12
- The default of the ``/2/instances/[instance_name]/rename`` RAPI
13
  resource's ``ip_check`` parameter changed from ``True`` to ``False``
14
  to match the underlying LUXI interface.
15
- The ``/2/nodes/[node_name]/evacuate`` RAPI resource was changed to use
16
  body parameters, see :doc:`RAPI documentation <rapi>`. The server does
17
  not maintain backwards-compatibility as the underlying operation
18
  changed in an incompatible way. The RAPI client can talk to old
19
  servers, but it needs to be told so as the return value changed.
20
- When creating file-based instances via RAPI, the ``file_driver``
21
  parameter no longer defaults to ``loop`` and must be specified.
22
- The deprecated ``bridge`` NIC parameter is no longer supported. Use
23
  ``link`` instead.
24
- Support for the undocumented and deprecated RAPI instance creation
25
  request format version 0 has been dropped. Use version 1, supported
26
  since Ganeti 2.1.3 and :doc:`documented <rapi>`, instead.
27
- Pyparsing 1.4.6 or above is required, see :doc:`installation
28
  documentation <install>`.
29
- The "cluster-verify" hooks are now executed per group by the
30
  ``OP_CLUSTER_VERIFY_GROUP`` opcode. This maintains the same behavior
31
  if you just run ``gnt-cluster verify``, which generates one opcode per
32
  group.
33
- The environment as passed to the OS scripts is cleared, and thus no
34
  environment variables defined in the node daemon's environment will be
35
  inherited by the scripts.
36
- The :doc:`iallocator <iallocator>` mode ``multi-evacuate`` has been
37
  deprecated.
38
- :doc:`New iallocator modes <design-multi-reloc>` have been added to
39
  support operations involving multiple node groups.
40
- Offline nodes are ignored when failing over an instance.
41

    
42
New features
43
~~~~~~~~~~~~
44

    
45
- The ganeti-htools project has been merged into the ganeti-core source
46
  tree and will be built as part of Ganeti (see :doc:`install-quick`).
47
- Implemented support for :doc:`shared storage <design-shared-storage>`.
48
- Add support for disks larger than 2 TB in ``lvmstrap`` by supporting
49
  GPT-style partition tables (requires `parted
50
  <http://www.gnu.org/s/parted/>`_).
51
- Added support for floppy drive and 2nd CD-ROM drive in KVM hypervisor.
52
- Allowed adding tags on instance creation.
53
- Export instance tags to hooks (``INSTANCE_TAGS``, see :doc:`hooks`)
54
- Allow instances to be started in a paused state, enabling the user to
55
  see the complete console output on boot using the console.
56
- Added new hypervisor flag to control default reboot behaviour
57
  (``reboot_behavior``).
58
- Added support for KVM keymaps (hypervisor parameter ``keymap``).
59
- Improved out-of-band management support:
60

    
61
  - Added ``gnt-node health`` command reporting the health status of
62
    nodes.
63
  - Added ``gnt-node power`` command to manage power status of nodes.
64
  - Added command for emergency power-off (EPO), ``gnt-cluster epo``.
65

    
66
- Instance migration can fall back to failover if instance is not
67
  running.
68
- Filters can be used when listing nodes, instances, groups and locks;
69
  see *ganeti(7)* manpage.
70
- Added post-execution status as variables to :doc:`hooks <hooks>`
71
  environment.
72
- Instance tags are exported/imported together with the instance.
73
- When given an explicit job ID, ``gnt-job info`` will work for archived
74
  jobs.
75
- Jobs can define dependencies on other jobs (not yet supported via
76
  RAPI or command line, but used by internal commands and usable via
77
  LUXI).
78

    
79
  - Lock monitor (``gnt-debug locks``) shows jobs waiting for
80
    dependencies.
81

    
82
- Instance failover is now available as a RAPI resource
83
  (``/2/instances/[instance_name]/failover``).
84
- ``gnt-instance info`` defaults to static information if primary node
85
  is offline.
86
- Opcodes have a new ``comment`` attribute.
87
- Added basic SPICE support to KVM hypervisor.
88
- ``tools/ganeti-listrunner`` allows passing of arguments to executable.
89

    
90
Node group improvements
91
~~~~~~~~~~~~~~~~~~~~~~~
92

    
93
- ``gnt-cluster verify`` has been modified to check groups separately,
94
  thereby improving performance.
95
- Node group support has been added to ``gnt-cluster verify-disks``,
96
  which now operates per node group.
97
- Watcher has been changed to work better with node groups.
98

    
99
  - One process and state file per node group.
100
  - Slow watcher in one group doesn't block other group's watcher.
101

    
102
- Added new command, ``gnt-group evacuate``, to move all instances in a
103
  node group to other groups.
104
- Added ``gnt-instance change-group`` to move an instance to another
105
  node group.
106
- ``gnt-cluster command`` and ``gnt-cluster copyfile`` now support
107
  per-group operations.
108
- Node groups can be tagged.
109
- Some operations switch from an exclusive to a shared lock as soon as
110
  possible.
111
- Instance's primary and secondary nodes' groups are now available as
112
  query fields (``pnode.group``, ``pnode.group.uuid``, ``snodes.group``
113
  and ``snodes.group.uuid``).
114

    
115
Misc
116
~~~~
117

    
118
- Numerous updates to documentation and manpages.
119

    
120
  - :doc:`RAPI <rapi>` documentation now has detailed parameter
121
    descriptions.
122

    
123
- A lockset's internal lock is now also visible in lock monitor.
124
- Log messages from job queue workers now contain information about the
125
  opcode they're processing.
126
- ``gnt-instance console`` no longer requires the instance lock.
127
- A short delay when waiting for job changes reduces the number of LUXI
128
  requests significantly.
129
- DRBD metadata volumes are overwritten with zeros during disk creation.
130

    
131

    
132
Version 2.5.0 rc2
133
-----------------
134

    
135
*(Released Tue, 18 Oct 2011)*
136

    
137
This was the second release candidate of the 2.5 series.
138

    
139

    
140
Version 2.5.0 rc1
141
-----------------
142

    
143
*(Released Tue, 4 Oct 2011)*
144

    
145
This was the first release candidate of the 2.5 series.
146

    
147

    
148
Version 2.5.0 beta3
149
-------------------
150

    
151
*(Released Wed, 31 Aug 2011)*
152

    
153
This was the third beta release of the 2.5 series.
154

    
155

    
156
Version 2.5.0 beta2
157
-------------------
158

    
159
*(Released Mon, 22 Aug 2011)*
160

    
161
This was the second beta release of the 2.5 series.
162

    
163

    
164
Version 2.5.0 beta1
165
-------------------
166

    
167
*(Released Fri, 12 Aug 2011)*
168

    
169
This was the first beta release of the 2.5 series.
170

    
171

    
172
Version 2.4.5
173
-------------
174

    
175
*(unreleased)*
176

    
177
- Fixed bug when parsing command line parameter values ending in
178
  backslash
179
- Fixed assertion error after unclean master shutdown
180
- Disable HTTP client pool for RPC, significantly reducing memory usage
181
  of master daemon
182

    
183

    
184
Version 2.4.4
185
-------------
186

    
187
*(Released Tue, 23 Aug 2011)*
188

    
189
Small bug-fixes:
190

    
191
- Fixed documentation for importing with ``--src-dir`` option
192
- Fixed a bug in ``ensure-dirs`` with queue/archive permissions
193
- Fixed a parsing issue with DRBD 8.3.11 in the Linux kernel
194

    
195

    
196
Version 2.4.3
197
-------------
198

    
199
*(Released Fri, 5 Aug 2011)*
200

    
201
Many bug-fixes and a few small features:
202

    
203
- Fixed argument order in ``ReserveLV`` and ``ReserveMAC`` which caused
204
  issues when you tried to add an instance with two MAC addresses in one
205
  request
206
- KVM: fixed per-instance stored UID value
207
- KVM: configure bridged NICs at migration start
208
- KVM: Fix a bug where instance will not start with never KVM versions
209
  (>= 0.14)
210
- Added OS search path to ``gnt-cluster info``
211
- Fixed an issue with ``file_storage_dir`` where you were forced to
212
  provide an absolute path, but the documentation states it is a
213
  relative path, the documentation was right
214
- Added a new parameter to instance stop/start called ``--no-remember``
215
  that will make the state change to not be remembered
216
- Implemented ``no_remember`` at RAPI level
217
- Improved the documentation
218
- Node evacuation: don't call IAllocator if node is already empty
219
- Fixed bug in DRBD8 replace disks on current nodes
220
- Fixed bug in recreate-disks for DRBD instances
221
- Moved assertion checking locks in ``gnt-instance replace-disks``
222
  causing it to abort with not owning the right locks for some situation
223
- Job queue: Fixed potential race condition when cancelling queued jobs
224
- Fixed off-by-one bug in job serial generation
225
- ``gnt-node volumes``: Fix instance names
226
- Fixed aliases in bash completion
227
- Fixed a bug in reopening log files after being sent a SIGHUP
228
- Added a flag to burnin to allow specifying VCPU count
229
- Bugfixes to non-root Ganeti configuration
230

    
231

    
232
Version 2.4.2
233
-------------
234

    
235
*(Released Thu, 12 May 2011)*
236

    
237
Many bug-fixes and a few new small features:
238

    
239
- Fixed a bug related to log opening failures
240
- Fixed a bug in instance listing with orphan instances
241
- Fixed a bug which prevented resetting the cluster-level node parameter
242
  ``oob_program`` to the default
243
- Many fixes related to the ``cluster-merge`` tool
244
- Fixed a race condition in the lock monitor, which caused failures
245
  during (at least) creation of many instances in parallel
246
- Improved output for gnt-job info
247
- Removed the quiet flag on some ssh calls which prevented debugging
248
  failures
249
- Improved the N+1 failure messages in cluster verify by actually
250
  showing the memory values (needed and available)
251
- Increased lock attempt timeouts so that when executing long operations
252
  (e.g. DRBD replace-disks) other jobs do not enter 'blocking acquire'
253
  too early and thus prevent the use of the 'fair' mechanism
254
- Changed instance query data (``gnt-instance info``) to not acquire
255
  locks unless needed, thus allowing its use on locked instance if only
256
  static information is asked for
257
- Improved behaviour with filesystems that do not support rename on an
258
  opened file
259
- Fixed the behaviour of ``prealloc_wipe_disks`` cluster parameter which
260
  kept locks on all nodes during the wipe, which is unneeded
261
- Fixed ``gnt-watcher`` handling of errors during hooks execution
262
- Fixed bug in ``prealloc_wipe_disks`` with small disk sizes (less than
263
  10GiB) which caused the wipe to fail right at the end in some cases
264
- Fixed master IP activation when doing master failover with no-voting
265
- Fixed bug in ``gnt-node add --readd`` which allowed the re-adding of
266
  the master node itself
267
- Fixed potential data-loss in under disk full conditions, where Ganeti
268
  wouldn't check correctly the return code and would consider
269
  partially-written files 'correct'
270
- Fixed bug related to multiple VGs and DRBD disk replacing
271
- Added new disk parameter ``metavg`` that allows placement of the meta
272
  device for DRBD in a different volume group
273
- Fixed error handling in the node daemon when the system libc doesn't
274
  have major number 6 (i.e. if ``libc.so.6`` is not the actual libc)
275
- Fixed lock release during replace-disks, which kept cluster-wide locks
276
  when doing disk replaces with an iallocator script
277
- Added check for missing bridges in cluster verify
278
- Handle EPIPE errors while writing to the terminal better, so that
279
  piping the output to e.g. ``less`` doesn't cause a backtrace
280
- Fixed rare case where a ^C during Luxi calls could have been
281
  interpreted as server errors, instead of simply terminating
282
- Fixed a race condition in LUGroupAssignNodes (``gnt-group
283
  assign-nodes``)
284
- Added a few more parameters to the KVM hypervisor, allowing a second
285
  CDROM, custom disk type for CDROMs and a floppy image
286
- Removed redundant message in instance rename when the name is given
287
  already as a FQDN
288
- Added option to ``gnt-instance recreate-disks`` to allow creating the
289
  disks on new nodes, allowing recreation when the original instance
290
  nodes are completely gone
291
- Added option when converting disk templates to DRBD to skip waiting
292
  for the resync, in order to make the instance available sooner
293
- Added two new variables to the OS scripts environment (containing the
294
  instance's nodes)
295
- Made the root_path and optional parameter for the xen-pvm hypervisor,
296
  to allow use of ``pvgrub`` as bootloader
297
- Changed the instance memory modifications to only check out-of-memory
298
  conditions on memory increases, and turned the secondary node warnings
299
  into errors (they can still be overridden via ``--force``)
300
- Fixed the handling of a corner case when the Python installation gets
301
  corrupted (e.g. a bad disk) while ganeti-noded is running and we try
302
  to execute a command that doesn't exist
303
- Fixed a bug in ``gnt-instance move`` (LUInstanceMove) when the primary
304
  node of the instance returned failures during instance shutdown; this
305
  adds the option ``--ignore-consistency`` to gnt-instance move
306

    
307
And as usual, various improvements to the error messages, documentation
308
and man pages.
309

    
310
Version 2.4.1
311
-------------
312

    
313
*(Released Wed, 09 Mar 2011)*
314

    
315
Emergency bug-fix release. ``tools/cfgupgrade`` was broken and overwrote
316
the RAPI users file if run twice (even with ``--dry-run``).
317

    
318
The release fixes that bug (nothing else changed).
319

    
320

    
321
Version 2.4.0
322
-------------
323

    
324
*(Released Mon, 07 Mar 2011)*
325

    
326
Final 2.4.0 release. Just a few small fixes:
327

    
328
- Fixed RAPI node evacuate
329
- Fixed the kvm-ifup script
330
- Fixed internal error handling for special job cases
331
- Updated man page to specify the escaping feature for options
332

    
333

    
334
Version 2.4.0 rc3
335
-----------------
336

    
337
*(Released Mon, 28 Feb 2011)*
338

    
339
A critical fix for the ``prealloc_wipe_disks`` feature: it is possible
340
that this feature wiped the disks of the wrong instance, leading to loss
341
of data.
342

    
343
Other changes:
344

    
345
- Fixed title of query field containing instance name
346
- Expanded the glossary in the documentation
347
- Fixed one unittest (internal issue)
348

    
349

    
350
Version 2.4.0 rc2
351
-----------------
352

    
353
*(Released Mon, 21 Feb 2011)*
354

    
355
A number of bug fixes plus just a couple functionality changes.
356

    
357
On the user-visible side, the ``gnt-* list`` command output has changed
358
with respect to "special" field states. The current rc1 style of display
359
can be re-enabled by passing a new ``--verbose`` (``-v``) flag, but in
360
the default output mode special fields states are displayed as follows:
361

    
362
- Offline resource: ``*``
363
- Unavailable/not applicable: ``-``
364
- Data missing (RPC failure): ``?``
365
- Unknown field: ``??``
366

    
367
Another user-visible change is the addition of ``--force-join`` to
368
``gnt-node add``.
369

    
370
As for bug fixes:
371

    
372
- ``tools/cluster-merge`` has seen many fixes and is now enabled again
373
- Fixed regression in RAPI/instance reinstall where all parameters were
374
  required (instead of optional)
375
- Fixed ``gnt-cluster repair-disk-sizes``, was broken since Ganeti 2.2
376
- Fixed iallocator usage (offline nodes were not considered offline)
377
- Fixed ``gnt-node list`` with respect to non-vm_capable nodes
378
- Fixed hypervisor and OS parameter validation with respect to
379
  non-vm_capable nodes
380
- Fixed ``gnt-cluster verify`` with respect to offline nodes (mostly
381
  cosmetic)
382
- Fixed ``tools/listrunner`` with respect to agent-based usage
383

    
384

    
385
Version 2.4.0 rc1
386
-----------------
387

    
388
*(Released Fri,  4 Feb 2011)*
389

    
390
Many changes and fixes since the beta1 release. While there were some
391
internal changes, the code has been mostly stabilised for the RC
392
release.
393

    
394
Note: the dumb allocator was removed in this release, as it was not kept
395
up-to-date with the IAllocator protocol changes. It is recommended to
396
use the ``hail`` command from the ganeti-htools package.
397

    
398
Note: the 2.4 and up versions of Ganeti are not compatible with the
399
0.2.x branch of ganeti-htools. You need to upgrade to
400
ganeti-htools-0.3.0 (or later).
401

    
402
Regressions fixed from 2.3
403
~~~~~~~~~~~~~~~~~~~~~~~~~~
404

    
405
- Fixed the ``gnt-cluster verify-disks`` command
406
- Made ``gnt-cluster verify-disks`` work in parallel (as opposed to
407
  serially on nodes)
408
- Fixed disk adoption breakage
409
- Fixed wrong headers in instance listing for field aliases
410

    
411
Other bugs fixed
412
~~~~~~~~~~~~~~~~
413

    
414
- Fixed corner case in KVM handling of NICs
415
- Fixed many cases of wrong handling of non-vm_capable nodes
416
- Fixed a bug where a missing instance symlink was not possible to
417
  recreate with any ``gnt-*`` command (now ``gnt-instance
418
  activate-disks`` does it)
419
- Fixed the volume group name as reported by ``gnt-cluster
420
  verify-disks``
421
- Increased timeouts for the import-export code, hopefully leading to
422
  fewer aborts due network or instance timeouts
423
- Fixed bug in ``gnt-node list-storage``
424
- Fixed bug where not all daemons were started on cluster
425
  initialisation, but only at the first watcher run
426
- Fixed many bugs in the OOB implementation
427
- Fixed watcher behaviour in presence of instances with offline
428
  secondaries
429
- Fixed instance list output for instances running on the wrong node
430
- a few fixes to the cluster-merge tool, but it still cannot merge
431
  multi-node groups (currently it is not recommended to use this tool)
432

    
433

    
434
Improvements
435
~~~~~~~~~~~~
436

    
437
- Improved network configuration for the KVM hypervisor
438
- Added e1000 as a supported NIC for Xen-HVM
439
- Improved the lvmstrap tool to also be able to use partitions, as
440
  opposed to full disks
441
- Improved speed of disk wiping (the cluster parameter
442
  ``prealloc_wipe_disks``, so that it has a low impact on the total time
443
  of instance creations
444
- Added documentation for the OS parameters
445
- Changed ``gnt-instance deactivate-disks`` so that it can work if the
446
  hypervisor is not responding
447
- Added display of blacklisted and hidden OS information in
448
  ``gnt-cluster info``
449
- Extended ``gnt-cluster verify`` to also validate hypervisor, backend,
450
  NIC and node parameters, which might create problems with currently
451
  invalid (but undetected) configuration files, but prevents validation
452
  failures when unrelated parameters are modified
453
- Changed cluster initialisation to wait for the master daemon to become
454
  available
455
- Expanded the RAPI interface:
456

    
457
  - Added config redistribution resource
458
  - Added activation/deactivation of instance disks
459
  - Added export of console information
460

    
461
- Implemented log file reopening on SIGHUP, which allows using
462
  logrotate(8) for the Ganeti log files
463
- Added a basic OOB helper script as an example
464

    
465

    
466
Version 2.4.0 beta1
467
-------------------
468

    
469
*(Released Fri, 14 Jan 2011)*
470

    
471
User-visible
472
~~~~~~~~~~~~
473

    
474
- Fixed timezone issues when formatting timestamps
475
- Added support for node groups, available via ``gnt-group`` and other
476
  commands
477
- Added out-of-band framework and management, see :doc:`design
478
  document <design-oob>`
479
- Removed support for roman numbers from ``gnt-node list`` and
480
  ``gnt-instance list``.
481
- Allowed modification of master network interface via ``gnt-cluster
482
  modify --master-netdev``
483
- Accept offline secondaries while shutting down instance disks
484
- Added ``blockdev_prefix`` parameter to Xen PVM and HVM hypervisors
485
- Added support for multiple LVM volume groups
486
- Avoid sorting nodes for ``gnt-node list`` if specific nodes are
487
  requested
488
- Added commands to list available fields:
489

    
490
  - ``gnt-node list-fields``
491
  - ``gnt-group list-fields``
492
  - ``gnt-instance list-fields``
493

    
494
- Updated documentation and man pages
495

    
496
Integration
497
~~~~~~~~~~~
498

    
499
- Moved ``rapi_users`` file into separate directory, now named
500
  ``.../ganeti/rapi/users``, ``cfgupgrade`` moves the file and creates a
501
  symlink
502
- Added new tool for running commands on many machines,
503
  ``tools/ganeti-listrunner``
504
- Implemented more verbose result in ``OpInstanceConsole`` opcode, also
505
  improving the ``gnt-instance console`` output
506
- Allowed customisation of disk index separator at ``configure`` time
507
- Export node group allocation policy to :doc:`iallocator <iallocator>`
508
- Added support for non-partitioned md disks in ``lvmstrap``
509
- Added script to gracefully power off KVM instances
510
- Split ``utils`` module into smaller parts
511
- Changed query operations to return more detailed information, e.g.
512
  whether an information is unavailable due to an offline node. To use
513
  this new functionality, the LUXI call ``Query`` must be used. Field
514
  information is now stored by the master daemon and can be retrieved
515
  using ``QueryFields``. Instances, nodes and groups can also be queried
516
  using the new opcodes ``OpQuery`` and ``OpQueryFields`` (not yet
517
  exposed via RAPI). The following commands make use of this
518
  infrastructure change:
519

    
520
  - ``gnt-group list``
521
  - ``gnt-group list-fields``
522
  - ``gnt-node list``
523
  - ``gnt-node list-fields``
524
  - ``gnt-instance list``
525
  - ``gnt-instance list-fields``
526
  - ``gnt-debug locks``
527

    
528
Remote API
529
~~~~~~~~~~
530

    
531
- New RAPI resources (see :doc:`rapi`):
532

    
533
  - ``/2/modify``
534
  - ``/2/groups``
535
  - ``/2/groups/[group_name]``
536
  - ``/2/groups/[group_name]/assign-nodes``
537
  - ``/2/groups/[group_name]/modify``
538
  - ``/2/groups/[group_name]/rename``
539
  - ``/2/instances/[instance_name]/disk/[disk_index]/grow``
540

    
541
- RAPI changes:
542

    
543
  - Implemented ``no_install`` for instance creation
544
  - Implemented OS parameters for instance reinstallation, allowing
545
    use of special settings on reinstallation (e.g. for preserving data)
546

    
547
Misc
548
~~~~
549

    
550
- Added IPv6 support in import/export
551
- Pause DRBD synchronization while wiping disks on instance creation
552
- Updated unittests and QA scripts
553
- Improved network parameters passed to KVM
554
- Converted man pages from docbook to reStructuredText
555

    
556

    
557
Version 2.3.1
558
-------------
559

    
560
*(Released Mon, 20 Dec 2010)*
561

    
562
Released version 2.3.1~rc1 without any changes.
563

    
564

    
565
Version 2.3.1 rc1
566
-----------------
567

    
568
*(Released Wed, 1 Dec 2010)*
569

    
570
- impexpd: Disable OpenSSL compression in socat if possible (backport
571
  from master, commit e90739d625b, see :doc:`installation guide
572
  <install-quick>` for details)
573
- Changed unittest coverage report to exclude test scripts
574
- Added script to check version format
575

    
576

    
577
Version 2.3.0
578
-------------
579

    
580
*(Released Wed, 1 Dec 2010)*
581

    
582
Released version 2.3.0~rc1 without any changes.
583

    
584

    
585
Version 2.3.0 rc1
586
-----------------
587

    
588
*(Released Fri, 19 Nov 2010)*
589

    
590
A number of bugfixes and documentation updates:
591

    
592
- Update ganeti-os-interface documentation
593
- Fixed a bug related to duplicate MACs or similar items which should be
594
  unique
595
- Fix breakage in OS state modify
596
- Reinstall instance: disallow offline secondaries (fixes bug related to
597
  OS changing but reinstall failing)
598
- plus all the other fixes between 2.2.1 and 2.2.2
599

    
600

    
601
Version 2.3.0 rc0
602
-----------------
603

    
604
*(Released Tue, 2 Nov 2010)*
605

    
606
- Fixed clearing of the default iallocator using ``gnt-cluster modify``
607
- Fixed master failover race with watcher
608
- Fixed a bug in ``gnt-node modify`` which could lead to an inconsistent
609
  configuration
610
- Accept previously stopped instance for export with instance removal
611
- Simplify and extend the environment variables for instance OS scripts
612
- Added new node flags, ``master_capable`` and ``vm_capable``
613
- Added optional instance disk wiping prior during allocation. This is a
614
  cluster-wide option and can be set/modified using
615
  ``gnt-cluster {init,modify} --prealloc-wipe-disks``.
616
- Added IPv6 support, see :doc:`design document <design-2.3>` and
617
  :doc:`install-quick`
618
- Added a new watcher option (``--ignore-pause``)
619
- Added option to ignore offline node on instance start/stop
620
  (``--ignore-offline``)
621
- Allow overriding OS parameters with ``gnt-instance reinstall``
622
- Added ability to change node's secondary IP address using ``gnt-node
623
  modify``
624
- Implemented privilege separation for all daemons except
625
  ``ganeti-noded``, see ``configure`` options
626
- Complain if an instance's disk is marked faulty in ``gnt-cluster
627
  verify``
628
- Implemented job priorities (see ``ganeti(7)`` manpage)
629
- Ignore failures while shutting down instances during failover from
630
  offline node
631
- Exit daemon's bootstrap process only once daemon is ready
632
- Export more information via ``LUInstanceQuery``/remote API
633
- Improved documentation, QA and unittests
634
- RAPI daemon now watches ``rapi_users`` all the time and doesn't need a
635
  restart if the file was created or changed
636
- Added LUXI protocol version sent with each request and response,
637
  allowing detection of server/client mismatches
638
- Moved the Python scripts among gnt-* and ganeti-* into modules
639
- Moved all code related to setting up SSH to an external script,
640
  ``setup-ssh``
641
- Infrastructure changes for node group support in future versions
642

    
643

    
644
Version 2.2.2
645
-------------
646

    
647
*(Released Fri, 19 Nov 2010)*
648

    
649
A few small bugs fixed, and some improvements to the build system:
650

    
651
- Fix documentation regarding conversion to drbd
652
- Fix validation of parameters in cluster modify (``gnt-cluster modify
653
  -B``)
654
- Fix error handling in node modify with multiple changes
655
- Allow remote imports without checked names
656

    
657

    
658
Version 2.2.1
659
-------------
660

    
661
*(Released Tue, 19 Oct 2010)*
662

    
663
- Disable SSL session ID cache in RPC client
664

    
665

    
666
Version 2.2.1 rc1
667
-----------------
668

    
669
*(Released Thu, 14 Oct 2010)*
670

    
671
- Fix interaction between Curl/GnuTLS and the Python's HTTP server
672
  (thanks Apollon Oikonomopoulos!), finally allowing the use of Curl
673
  with GnuTLS
674
- Fix problems with interaction between Curl and Python's HTTP server,
675
  resulting in increased speed in many RPC calls
676
- Improve our release script to prevent breakage with older aclocal and
677
  Python 2.6
678

    
679

    
680
Version 2.2.1 rc0
681
-----------------
682

    
683
*(Released Thu, 7 Oct 2010)*
684

    
685
- Fixed issue 125, replace hardcoded "xenvg" in ``gnt-cluster`` with
686
  value retrieved from master
687
- Added support for blacklisted or hidden OS definitions
688
- Added simple lock monitor (accessible via (``gnt-debug locks``)
689
- Added support for -mem-path in KVM hypervisor abstraction layer
690
- Allow overriding instance parameters in tool for inter-cluster
691
  instance moves (``tools/move-instance``)
692
- Improved opcode summaries (e.g. in ``gnt-job list``)
693
- Improve consistency of OS listing by sorting it
694
- Documentation updates
695

    
696

    
697
Version 2.2.0.1
698
---------------
699

    
700
*(Released Fri, 8 Oct 2010)*
701

    
702
- Rebuild with a newer autotools version, to fix python 2.6 compatibility
703

    
704

    
705
Version 2.2.0
706
-------------
707

    
708
*(Released Mon, 4 Oct 2010)*
709

    
710
- Fixed regression in ``gnt-instance rename``
711

    
712

    
713
Version 2.2.0 rc2
714
-----------------
715

    
716
*(Released Wed, 22 Sep 2010)*
717

    
718
- Fixed OS_VARIANT variable for OS scripts
719
- Fixed cluster tag operations via RAPI
720
- Made ``setup-ssh`` exit with non-zero code if an error occurred
721
- Disabled RAPI CA checks in watcher
722

    
723

    
724
Version 2.2.0 rc1
725
-----------------
726

    
727
*(Released Mon, 23 Aug 2010)*
728

    
729
- Support DRBD versions of the format "a.b.c.d"
730
- Updated manpages
731
- Re-introduce support for usage from multiple threads in RAPI client
732
- Instance renames and modify via RAPI
733
- Work around race condition between processing and archival in job
734
  queue
735
- Mark opcodes following failed one as failed, too
736
- Job field ``lock_status`` was removed due to difficulties making it
737
  work with the changed job queue in Ganeti 2.2; a better way to monitor
738
  locks is expected for a later 2.2.x release
739
- Fixed dry-run behaviour with many commands
740
- Support ``ssh-agent`` again when adding nodes
741
- Many additional bugfixes
742

    
743

    
744
Version 2.2.0 rc0
745
-----------------
746

    
747
*(Released Fri, 30 Jul 2010)*
748

    
749
Important change: the internal RPC mechanism between Ganeti nodes has
750
changed from using a home-grown http library (based on the Python base
751
libraries) to use the PycURL library. This requires that PycURL is
752
installed on nodes. Please note that on Debian/Ubuntu, PycURL is linked
753
against GnuTLS by default. cURL's support for GnuTLS had known issues
754
before cURL 7.21.0 and we recommend using the latest cURL release or
755
linking against OpenSSL. Most other distributions already link PycURL
756
and cURL against OpenSSL. The command::
757

    
758
  python -c 'import pycurl; print pycurl.version'
759

    
760
can be used to determine the libraries PycURL and cURL are linked
761
against.
762

    
763
Other significant changes:
764

    
765
- Rewrote much of the internals of the job queue, in order to achieve
766
  better parallelism; this decouples job query operations from the job
767
  processing, and it should allow much nicer behaviour of the master
768
  daemon under load, and it also has uncovered some long-standing bugs
769
  related to the job serialisation (now fixed)
770
- Added a default iallocator setting to the cluster parameters,
771
  eliminating the need to always pass nodes or an iallocator for
772
  operations that require selection of new node(s)
773
- Added experimental support for the LXC virtualization method
774
- Added support for OS parameters, which allows the installation of
775
  instances to pass parameter to OS scripts in order to customise the
776
  instance
777
- Added a hypervisor parameter controlling the migration type (live or
778
  non-live), since hypervisors have various levels of reliability; this
779
  has renamed the 'live' parameter to 'mode'
780
- Added a cluster parameter ``reserved_lvs`` that denotes reserved
781
  logical volumes, meaning that cluster verify will ignore them and not
782
  flag their presence as errors
783
- The watcher will now reset the error count for failed instances after
784
  8 hours, thus allowing self-healing if the problem that caused the
785
  instances to be down/fail to start has cleared in the meantime
786
- Added a cluster parameter ``drbd_usermode_helper`` that makes Ganeti
787
  check for, and warn, if the drbd module parameter ``usermode_helper``
788
  is not consistent with the cluster-wide setting; this is needed to
789
  make diagnose easier of failed drbd creations
790
- Started adding base IPv6 support, but this is not yet
791
  enabled/available for use
792
- Rename operations (cluster, instance) will now return the new name,
793
  which is especially useful if a short name was passed in
794
- Added support for instance migration in RAPI
795
- Added a tool to pre-configure nodes for the SSH setup, before joining
796
  them to the cluster; this will allow in the future a simplified model
797
  for node joining (but not yet fully enabled in 2.2); this needs the
798
  paramiko python library
799
- Fixed handling of name-resolving errors
800
- Fixed consistency of job results on the error path
801
- Fixed master-failover race condition when executed multiple times in
802
  sequence
803
- Fixed many bugs related to the job queue (mostly introduced during the
804
  2.2 development cycle, so not all are impacting 2.1)
805
- Fixed instance migration with missing disk symlinks
806
- Fixed handling of unknown jobs in ``gnt-job archive``
807
- And many other small fixes/improvements
808

    
809
Internal changes:
810

    
811
- Enhanced both the unittest and the QA coverage
812
- Switched the opcode validation to a generic model, and extended the
813
  validation to all opcode parameters
814
- Changed more parts of the code that write shell scripts to use the
815
  same class for this
816
- Switched the master daemon to use the asyncore library for the Luxi
817
  server endpoint
818

    
819

    
820
Version 2.2.0 beta 0
821
--------------------
822

    
823
*(Released Thu, 17 Jun 2010)*
824

    
825
- Added tool (``move-instance``) and infrastructure to move instances
826
  between separate clusters (see :doc:`separate documentation
827
  <move-instance>` and :doc:`design document <design-2.2>`)
828
- Added per-request RPC timeout
829
- RAPI now requires a Content-Type header for requests with a body (e.g.
830
  ``PUT`` or ``POST``) which must be set to ``application/json`` (see
831
  :rfc:`2616` (HTTP/1.1), section 7.2.1)
832
- ``ganeti-watcher`` attempts to restart ``ganeti-rapi`` if RAPI is not
833
  reachable
834
- Implemented initial support for running Ganeti daemons as separate
835
  users, see configure-time flags ``--with-user-prefix`` and
836
  ``--with-group-prefix`` (only ``ganeti-rapi`` is supported at this
837
  time)
838
- Instances can be removed after export (``gnt-backup export
839
  --remove-instance``)
840
- Self-signed certificates generated by Ganeti now use a 2048 bit RSA
841
  key (instead of 1024 bit)
842
- Added new cluster configuration file for cluster domain secret
843
- Import/export now use SSL instead of SSH
844
- Added support for showing estimated time when exporting an instance,
845
  see the ``ganeti-os-interface(7)`` manpage and look for
846
  ``EXP_SIZE_FD``
847

    
848

    
849
Version 2.1.8
850
-------------
851

    
852
*(Released Tue, 16 Nov 2010)*
853

    
854
Some more bugfixes. Unless critical bugs occur, this will be the last
855
2.1 release:
856

    
857
- Fix case of MAC special-values
858
- Fix mac checker regex
859
- backend: Fix typo causing "out of range" error
860
- Add missing --units in gnt-instance list man page
861

    
862

    
863
Version 2.1.7
864
-------------
865

    
866
*(Released Tue, 24 Aug 2010)*
867

    
868
Bugfixes only:
869
  - Don't ignore secondary node silently on non-mirrored disk templates
870
    (issue 113)
871
  - Fix --master-netdev arg name in gnt-cluster(8) (issue 114)
872
  - Fix usb_mouse parameter breaking with vnc_console (issue 109)
873
  - Properly document the usb_mouse parameter
874
  - Fix path in ganeti-rapi(8) (issue 116)
875
  - Adjust error message when the ganeti user's .ssh directory is
876
    missing
877
  - Add same-node-check when changing the disk template to drbd
878

    
879

    
880
Version 2.1.6
881
-------------
882

    
883
*(Released Fri, 16 Jul 2010)*
884

    
885
Bugfixes only:
886
  - Add an option to only select some reboot types during qa/burnin.
887
    (on some hypervisors consequent reboots are not supported)
888
  - Fix infrequent race condition in master failover. Sometimes the old
889
    master ip address would be still detected as up for a short time
890
    after it was removed, causing failover to fail.
891
  - Decrease mlockall warnings when the ctypes module is missing. On
892
    Python 2.4 we support running even if no ctypes module is installed,
893
    but we were too verbose about this issue.
894
  - Fix building on old distributions, on which man doesn't have a
895
    --warnings option.
896
  - Fix RAPI not to ignore the MAC address on instance creation
897
  - Implement the old instance creation format in the RAPI client.
898

    
899

    
900
Version 2.1.5
901
-------------
902

    
903
*(Released Thu, 01 Jul 2010)*
904

    
905
A small bugfix release:
906
  - Fix disk adoption: broken by strict --disk option checking in 2.1.4
907
  - Fix batch-create: broken in the whole 2.1 series due to a lookup on
908
    a non-existing option
909
  - Fix instance create: the --force-variant option was ignored
910
  - Improve pylint 0.21 compatibility and warnings with Python 2.6
911
  - Fix modify node storage with non-FQDN arguments
912
  - Fix RAPI client to authenticate under Python 2.6 when used
913
    for more than 5 requests needing authentication
914
  - Fix gnt-instance modify -t (storage) giving a wrong error message
915
    when converting a non-shutdown drbd instance to plain
916

    
917

    
918
Version 2.1.4
919
-------------
920

    
921
*(Released Fri, 18 Jun 2010)*
922

    
923
A small bugfix release:
924

    
925
  - Fix live migration of KVM instances started with older Ganeti
926
    versions which had fewer hypervisor parameters
927
  - Fix gnt-instance grow-disk on down instances
928
  - Fix an error-reporting bug during instance migration
929
  - Better checking of the ``--net`` and ``--disk`` values, to avoid
930
    silently ignoring broken ones
931
  - Fix an RPC error reporting bug affecting, for example, RAPI client
932
    users
933
  - Fix bug triggered by different API version os-es on different nodes
934
  - Fix a bug in instance startup with custom hvparams: OS level
935
    parameters would fail to be applied.
936
  - Fix the RAPI client under Python 2.6 (but more work is needed to
937
    make it work completely well with OpenSSL)
938
  - Fix handling of errors when resolving names from DNS
939

    
940

    
941
Version 2.1.3
942
-------------
943

    
944
*(Released Thu, 3 Jun 2010)*
945

    
946
A medium sized development cycle. Some new features, and some
947
fixes/small improvements/cleanups.
948

    
949
Significant features
950
~~~~~~~~~~~~~~~~~~~~
951

    
952
The node deamon now tries to mlock itself into memory, unless the
953
``--no-mlock`` flag is passed. It also doesn't fail if it can't write
954
its logs, and falls back to console logging. This allows emergency
955
features such as ``gnt-node powercycle`` to work even in the event of a
956
broken node disk (tested offlining the disk hosting the node's
957
filesystem and dropping its memory caches; don't try this at home)
958

    
959
KVM: add vhost-net acceleration support. It can be tested with a new
960
enough version of the kernel and of qemu-kvm.
961

    
962
KVM: Add instance chrooting feature. If you use privilege dropping for
963
your VMs you can also now force them to chroot to an empty directory,
964
before starting the emulated guest.
965

    
966
KVM: Add maximum migration bandwith and maximum downtime tweaking
967
support (requires a new-enough version of qemu-kvm).
968

    
969
Cluster verify will now warn if the master node doesn't have the master
970
ip configured on it.
971

    
972
Add a new (incompatible) instance creation request format to RAPI which
973
supports all parameters (previously only a subset was supported, and it
974
wasn't possible to extend the old format to accomodate all the new
975
features. The old format is still supported, and a client can check for
976
this feature, before using it, by checking for its presence in the
977
``features`` RAPI resource.
978

    
979
Now with ancient latin support. Try it passing the ``--roman`` option to
980
``gnt-instance info``, ``gnt-cluster info`` or ``gnt-node list``
981
(requires the python-roman module to be installed, in order to work).
982

    
983
Other changes
984
~~~~~~~~~~~~~
985

    
986
As usual many internal code refactorings, documentation updates, and
987
such. Among others:
988

    
989
  - Lots of improvements and cleanups to the experimental Remote API
990
    (RAPI) client library.
991
  - A new unit test suite for the core daemon libraries.
992
  - A fix to creating missing directories makes sure the umask is not
993
    applied anymore. This enforces the same directory permissions
994
    everywhere.
995
  - Better handling terminating daemons with ctrl+c (used when running
996
    them in debugging mode).
997
  - Fix a race condition in live migrating a KVM instance, when stat()
998
    on the old proc status file returned EINVAL, which is an unexpected
999
    value.
1000
  - Fixed manpage checking with newer man and utf-8 charachters. But now
1001
    you need the en_US.UTF-8 locale enabled to build Ganeti from git.
1002

    
1003

    
1004
Version 2.1.2.1
1005
---------------
1006

    
1007
*(Released Fri, 7 May 2010)*
1008

    
1009
Fix a bug which prevented untagged KVM instances from starting.
1010

    
1011

    
1012
Version 2.1.2
1013
-------------
1014

    
1015
*(Released Fri, 7 May 2010)*
1016

    
1017
Another release with a long development cycle, during which many
1018
different features were added.
1019

    
1020
Significant features
1021
~~~~~~~~~~~~~~~~~~~~
1022

    
1023
The KVM hypervisor now can run the individual instances as non-root, to
1024
reduce the impact of a VM being hijacked due to bugs in the
1025
hypervisor. It is possible to run all instances as a single (non-root)
1026
user, to manually specify a user for each instance, or to dynamically
1027
allocate a user out of a cluster-wide pool to each instance, with the
1028
guarantee that no two instances will run under the same user ID on any
1029
given node.
1030

    
1031
An experimental RAPI client library, that can be used standalone
1032
(without the other Ganeti libraries), is provided in the source tree as
1033
``lib/rapi/client.py``. Note this client might change its interface in
1034
the future, as we iterate on its capabilities.
1035

    
1036
A new command, ``gnt-cluster renew-crypto`` has been added to easily
1037
replace the cluster's certificates and crypto keys. This might help in
1038
case they have been compromised, or have simply expired.
1039

    
1040
A new disk option for instance creation has been added that allows one
1041
to "adopt" currently existing logical volumes, with data
1042
preservation. This should allow easier migration to Ganeti from
1043
unmanaged (or managed via other software) instances.
1044

    
1045
Another disk improvement is the possibility to convert between redundant
1046
(DRBD) and plain (LVM) disk configuration for an instance. This should
1047
allow better scalability (starting with one node and growing the
1048
cluster, or shrinking a two-node cluster to one node).
1049

    
1050
A new feature that could help with automated node failovers has been
1051
implemented: if a node sees itself as offline (by querying the master
1052
candidates), it will try to shutdown (hard) all instances and any active
1053
DRBD devices. This reduces the risk of duplicate instances if an
1054
external script automatically failovers the instances on such nodes. To
1055
enable this, the cluster parameter ``maintain_node_health`` should be
1056
enabled; in the future this option (per the name) will enable other
1057
automatic maintenance features.
1058

    
1059
Instance export/import now will reuse the original instance
1060
specifications for all parameters; that means exporting an instance,
1061
deleting it and the importing it back should give an almost identical
1062
instance. Note that the default import behaviour has changed from
1063
before, where it created only one NIC; now it recreates the original
1064
number of NICs.
1065

    
1066
Cluster verify has added a few new checks: SSL certificates validity,
1067
/etc/hosts consistency across the cluster, etc.
1068

    
1069
Other changes
1070
~~~~~~~~~~~~~
1071

    
1072
As usual, many internal changes were done, documentation fixes,
1073
etc. Among others:
1074

    
1075
- Fixed cluster initialization with disabled cluster storage (regression
1076
  introduced in 2.1.1)
1077
- File-based storage supports growing the disks
1078
- Fixed behaviour of node role changes
1079
- Fixed cluster verify for some corner cases, plus a general rewrite of
1080
  cluster verify to allow future extension with more checks
1081
- Fixed log spamming by watcher and node daemon (regression introduced
1082
  in 2.1.1)
1083
- Fixed possible validation issues when changing the list of enabled
1084
  hypervisors
1085
- Fixed cleanup of /etc/hosts during node removal
1086
- Fixed RAPI response for invalid methods
1087
- Fixed bug with hashed passwords in ``ganeti-rapi`` daemon
1088
- Multiple small improvements to the KVM hypervisor (VNC usage, booting
1089
  from ide disks, etc.)
1090
- Allow OS changes without re-installation (to record a changed OS
1091
  outside of Ganeti, or to allow OS renames)
1092
- Allow instance creation without OS installation (useful for example if
1093
  the OS will be installed manually, or restored from a backup not in
1094
  Ganeti format)
1095
- Implemented option to make cluster ``copyfile`` use the replication
1096
  network
1097
- Added list of enabled hypervisors to ssconf (possibly useful for
1098
  external scripts)
1099
- Added a new tool (``tools/cfgupgrade12``) that allows upgrading from
1100
  1.2 clusters
1101
- A partial form of node re-IP is possible via node readd, which now
1102
  allows changed node primary IP
1103
- Command line utilities now show an informational message if the job is
1104
  waiting for a lock
1105
- The logs of the master daemon now show the PID/UID/GID of the
1106
  connected client
1107

    
1108

    
1109
Version 2.1.1
1110
-------------
1111

    
1112
*(Released Fri, 12 Mar 2010)*
1113

    
1114
During the 2.1.0 long release candidate cycle, a lot of improvements and
1115
changes have accumulated with were released later as 2.1.1.
1116

    
1117
Major changes
1118
~~~~~~~~~~~~~
1119

    
1120
The node evacuate command (``gnt-node evacuate``) was significantly
1121
rewritten, and as such the IAllocator protocol was changed - a new
1122
request type has been added. This unfortunate change during a stable
1123
series is designed to improve performance of node evacuations; on
1124
clusters with more than about five nodes and which are well-balanced,
1125
evacuation should proceed in parallel for all instances of the node
1126
being evacuated. As such, any existing IAllocator scripts need to be
1127
updated, otherwise the above command will fail due to the unknown
1128
request. The provided "dumb" allocator has not been updated; but the
1129
ganeti-htools package supports the new protocol since version 0.2.4.
1130

    
1131
Another important change is increased validation of node and instance
1132
names. This might create problems in special cases, if invalid host
1133
names are being used.
1134

    
1135
Also, a new layer of hypervisor parameters has been added, that sits at
1136
OS level between the cluster defaults and the instance ones. This allows
1137
customisation of virtualization parameters depending on the installed
1138
OS. For example instances with OS 'X' may have a different KVM kernel
1139
(or any other parameter) than the cluster defaults. This is intended to
1140
help managing a multiple OSes on the same cluster, without manual
1141
modification of each instance's parameters.
1142

    
1143
A tool for merging clusters, ``cluster-merge``, has been added in the
1144
tools sub-directory.
1145

    
1146
Bug fixes
1147
~~~~~~~~~
1148

    
1149
- Improved the int/float conversions that should make the code more
1150
  robust in face of errors from the node daemons
1151
- Fixed the remove node code in case of internal configuration errors
1152
- Fixed the node daemon behaviour in face of inconsistent queue
1153
  directory (e.g. read-only file-system where we can't open the files
1154
  read-write, etc.)
1155
- Fixed the behaviour of gnt-node modify for master candidate demotion;
1156
  now it either aborts cleanly or, if given the new "auto_promote"
1157
  parameter, will automatically promote other nodes as needed
1158
- Fixed compatibility with (unreleased yet) Python 2.6.5 that would
1159
  completely prevent Ganeti from working
1160
- Fixed bug for instance export when not all disks were successfully
1161
  exported
1162
- Fixed behaviour of node add when the new node is slow in starting up
1163
  the node daemon
1164
- Fixed handling of signals in the LUXI client, which should improve
1165
  behaviour of command-line scripts
1166
- Added checks for invalid node/instance names in the configuration (now
1167
  flagged during cluster verify)
1168
- Fixed watcher behaviour for disk activation errors
1169
- Fixed two potentially endless loops in http library, which led to the
1170
  RAPI daemon hanging and consuming 100% CPU in some cases
1171
- Fixed bug in RAPI daemon related to hashed passwords
1172
- Fixed bug for unintended qemu-level bridging of multi-NIC KVM
1173
  instances
1174
- Enhanced compatibility with non-Debian OSes, but not using absolute
1175
  path in some commands and allowing customisation of the ssh
1176
  configuration directory
1177
- Fixed possible future issue with new Python versions by abiding to the
1178
  proper use of ``__slots__`` attribute on classes
1179
- Added checks that should prevent directory traversal attacks
1180
- Many documentation fixes based on feedback from users
1181

    
1182
New features
1183
~~~~~~~~~~~~
1184

    
1185
- Added an "early_release" more for instance replace disks and node
1186
  evacuate, where we release locks earlier and thus allow higher
1187
  parallelism within the cluster
1188
- Added watcher hooks, intended to allow the watcher to restart other
1189
  daemons (e.g. from the ganeti-nbma project), but they can be used of
1190
  course for any other purpose
1191
- Added a compile-time disable for DRBD barriers, to increase
1192
  performance if the administrator trusts the power supply or the
1193
  storage system to not lose writes
1194
- Added the option of using syslog for logging instead of, or in
1195
  addition to, Ganeti's own log files
1196
- Removed boot restriction for paravirtual NICs for KVM, recent versions
1197
  can indeed boot from a paravirtual NIC
1198
- Added a generic debug level for many operations; while this is not
1199
  used widely yet, it allows one to pass the debug value all the way to
1200
  the OS scripts
1201
- Enhanced the hooks environment for instance moves (failovers,
1202
  migrations) where the primary/secondary nodes changed during the
1203
  operation, by adding {NEW,OLD}_{PRIMARY,SECONDARY} vars
1204
- Enhanced data validations for many user-supplied values; one important
1205
  item is the restrictions imposed on instance and node names, which
1206
  might reject some (invalid) host names
1207
- Add a configure-time option to disable file-based storage, if it's not
1208
  needed; this allows greater security separation between the master
1209
  node and the other nodes from the point of view of the inter-node RPC
1210
  protocol
1211
- Added user notification in interactive tools if job is waiting in the
1212
  job queue or trying to acquire locks
1213
- Added log messages when a job is waiting for locks
1214
- Added filtering by node tags in instance operations which admit
1215
  multiple instances (start, stop, reboot, reinstall)
1216
- Added a new tool for cluster mergers, ``cluster-merge``
1217
- Parameters from command line which are of the form ``a=b,c=d`` can now
1218
  use backslash escapes to pass in values which contain commas,
1219
  e.g. ``a=b\\c,d=e`` where the 'a' parameter would get the value
1220
  ``b,c``
1221
- For KVM, the instance name is the first parameter passed to KVM, so
1222
  that it's more visible in the process list
1223

    
1224

    
1225
Version 2.1.0
1226
-------------
1227

    
1228
*(Released Tue, 2 Mar 2010)*
1229

    
1230
Ganeti 2.1 brings many improvements with it. Major changes:
1231

    
1232
- Added infrastructure to ease automated disk repairs
1233
- Added new daemon to export configuration data in a cheaper way than
1234
  using the remote API
1235
- Instance NICs can now be routed instead of being associated with a
1236
  networking bridge
1237
- Improved job locking logic to reduce impact of jobs acquiring multiple
1238
  locks waiting for other long-running jobs
1239

    
1240
In-depth implementation details can be found in the Ganeti 2.1 design
1241
document.
1242

    
1243
Details
1244
~~~~~~~
1245

    
1246
- Added chroot hypervisor
1247
- Added more options to xen-hvm hypervisor (``kernel_path`` and
1248
  ``device_model``)
1249
- Added more options to xen-pvm hypervisor (``use_bootloader``,
1250
  ``bootloader_path`` and ``bootloader_args``)
1251
- Added the ``use_localtime`` option for the xen-hvm and kvm
1252
  hypervisors, and the default value for this has changed to false (in
1253
  2.0 xen-hvm always enabled it)
1254
- Added luxi call to submit multiple jobs in one go
1255
- Added cluster initialization option to not modify ``/etc/hosts``
1256
  file on nodes
1257
- Added network interface parameters
1258
- Added dry run mode to some LUs
1259
- Added RAPI resources:
1260

    
1261
  - ``/2/instances/[instance_name]/info``
1262
  - ``/2/instances/[instance_name]/replace-disks``
1263
  - ``/2/nodes/[node_name]/evacuate``
1264
  - ``/2/nodes/[node_name]/migrate``
1265
  - ``/2/nodes/[node_name]/role``
1266
  - ``/2/nodes/[node_name]/storage``
1267
  - ``/2/nodes/[node_name]/storage/modify``
1268
  - ``/2/nodes/[node_name]/storage/repair``
1269

    
1270
- Added OpCodes to evacuate or migrate all instances on a node
1271
- Added new command to list storage elements on nodes (``gnt-node
1272
  list-storage``) and modify them (``gnt-node modify-storage``)
1273
- Added new ssconf files with master candidate IP address
1274
  (``ssconf_master_candidates_ips``), node primary IP address
1275
  (``ssconf_node_primary_ips``) and node secondary IP address
1276
  (``ssconf_node_secondary_ips``)
1277
- Added ``ganeti-confd`` and a client library to query the Ganeti
1278
  configuration via UDP
1279
- Added ability to run hooks after cluster initialization and before
1280
  cluster destruction
1281
- Added automatic mode for disk replace (``gnt-instance replace-disks
1282
  --auto``)
1283
- Added ``gnt-instance recreate-disks`` to re-create (empty) disks
1284
  after catastrophic data-loss
1285
- Added ``gnt-node repair-storage`` command to repair damaged LVM volume
1286
  groups
1287
- Added ``gnt-instance move`` command to move instances
1288
- Added ``gnt-cluster watcher`` command to control watcher
1289
- Added ``gnt-node powercycle`` command to powercycle nodes
1290
- Added new job status field ``lock_status``
1291
- Added parseable error codes to cluster verification (``gnt-cluster
1292
  verify --error-codes``) and made output less verbose (use
1293
  ``--verbose`` to restore previous behaviour)
1294
- Added UUIDs to the main config entities (cluster, nodes, instances)
1295
- Added support for OS variants
1296
- Added support for hashed passwords in the Ganeti remote API users file
1297
  (``rapi_users``)
1298
- Added option to specify maximum timeout on instance shutdown
1299
- Added ``--no-ssh-init`` option to ``gnt-cluster init``
1300
- Added new helper script to start and stop Ganeti daemons
1301
  (``daemon-util``), with the intent to reduce the work necessary to
1302
  adjust Ganeti for non-Debian distributions and to start/stop daemons
1303
  from one place
1304
- Added more unittests
1305
- Fixed critical bug in ganeti-masterd startup
1306
- Removed the configure-time ``kvm-migration-port`` parameter, this is
1307
  now customisable at the cluster level for both the KVM and Xen
1308
  hypervisors using the new ``migration_port`` parameter
1309
- Pass ``INSTANCE_REINSTALL`` variable to OS installation script when
1310
  reinstalling an instance
1311
- Allowed ``@`` in tag names
1312
- Migrated to Sphinx (http://sphinx.pocoo.org/) for documentation
1313
- Many documentation updates
1314
- Distribute hypervisor files on ``gnt-cluster redist-conf``
1315
- ``gnt-instance reinstall`` can now reinstall multiple instances
1316
- Updated many command line parameters
1317
- Introduced new OS API version 15
1318
- No longer support a default hypervisor
1319
- Treat virtual LVs as inexistent
1320
- Improved job locking logic to reduce lock contention
1321
- Match instance and node names case insensitively
1322
- Reimplemented bash completion script to be more complete
1323
- Improved burnin
1324

    
1325

    
1326
Version 2.0.6
1327
-------------
1328

    
1329
*(Released Thu, 4 Feb 2010)*
1330

    
1331
- Fix cleaner behaviour on nodes not in a cluster (Debian bug 568105)
1332
- Fix a string formatting bug
1333
- Improve safety of the code in some error paths
1334
- Improve data validation in the master of values returned from nodes
1335

    
1336

    
1337
Version 2.0.5
1338
-------------
1339

    
1340
*(Released Thu, 17 Dec 2009)*
1341

    
1342
- Fix security issue due to missing validation of iallocator names; this
1343
  allows local and remote execution of arbitrary executables
1344
- Fix failure of gnt-node list during instance removal
1345
- Ship the RAPI documentation in the archive
1346

    
1347

    
1348
Version 2.0.4
1349
-------------
1350

    
1351
*(Released Wed, 30 Sep 2009)*
1352

    
1353
- Fixed many wrong messages
1354
- Fixed a few bugs related to the locking library
1355
- Fixed MAC checking at instance creation time
1356
- Fixed a DRBD parsing bug related to gaps in /proc/drbd
1357
- Fixed a few issues related to signal handling in both daemons and
1358
  scripts
1359
- Fixed the example startup script provided
1360
- Fixed insserv dependencies in the example startup script (patch from
1361
  Debian)
1362
- Fixed handling of drained nodes in the iallocator framework
1363
- Fixed handling of KERNEL_PATH parameter for xen-hvm (Debian bug
1364
  #528618)
1365
- Fixed error related to invalid job IDs in job polling
1366
- Fixed job/opcode persistence on unclean master shutdown
1367
- Fixed handling of partial job processing after unclean master
1368
  shutdown
1369
- Fixed error reporting from LUs, previously all errors were converted
1370
  into execution errors
1371
- Fixed error reporting from burnin
1372
- Decreased significantly the memory usage of the job queue
1373
- Optimised slightly multi-job submission
1374
- Optimised slightly opcode loading
1375
- Backported the multi-job submit framework from the development
1376
  branch; multi-instance start and stop should be faster
1377
- Added script to clean archived jobs after 21 days; this will reduce
1378
  the size of the queue directory
1379
- Added some extra checks in disk size tracking
1380
- Added an example ethers hook script
1381
- Added a cluster parameter that prevents Ganeti from modifying of
1382
  /etc/hosts
1383
- Added more node information to RAPI responses
1384
- Added a ``gnt-job watch`` command that allows following the ouput of a
1385
  job
1386
- Added a bind-address option to ganeti-rapi
1387
- Added more checks to the configuration verify
1388
- Enhanced the burnin script such that some operations can be retried
1389
  automatically
1390
- Converted instance reinstall to multi-instance model
1391

    
1392

    
1393
Version 2.0.3
1394
-------------
1395

    
1396
*(Released Fri, 7 Aug 2009)*
1397

    
1398
- Added ``--ignore-size`` to the ``gnt-instance activate-disks`` command
1399
  to allow using the pre-2.0.2 behaviour in activation, if any existing
1400
  instances have mismatched disk sizes in the configuration
1401
- Added ``gnt-cluster repair-disk-sizes`` command to check and update
1402
  any configuration mismatches for disk sizes
1403
- Added ``gnt-master cluste-failover --no-voting`` to allow master
1404
  failover to work on two-node clusters
1405
- Fixed the ``--net`` option of ``gnt-backup import``, which was
1406
  unusable
1407
- Fixed detection of OS script errors in ``gnt-backup export``
1408
- Fixed exit code of ``gnt-backup export``
1409

    
1410

    
1411
Version 2.0.2
1412
-------------
1413

    
1414
*(Released Fri, 17 Jul 2009)*
1415

    
1416
- Added experimental support for stripped logical volumes; this should
1417
  enhance performance but comes with a higher complexity in the block
1418
  device handling; stripping is only enabled when passing
1419
  ``--with-lvm-stripecount=N`` to ``configure``, but codepaths are
1420
  affected even in the non-stripped mode
1421
- Improved resiliency against transient failures at the end of DRBD
1422
  resyncs, and in general of DRBD resync checks
1423
- Fixed a couple of issues with exports and snapshot errors
1424
- Fixed a couple of issues in instance listing
1425
- Added display of the disk size in ``gnt-instance info``
1426
- Fixed checking for valid OSes in instance creation
1427
- Fixed handling of the "vcpus" parameter in instance listing and in
1428
  general of invalid parameters
1429
- Fixed http server library, and thus RAPI, to handle invalid
1430
  username/password combinations correctly; this means that now they
1431
  report unauthorized for queries too, not only for modifications,
1432
  allowing earlier detect of configuration problems
1433
- Added a new "role" node list field, equivalent to the master/master
1434
  candidate/drained/offline flags combinations
1435
- Fixed cluster modify and changes of candidate pool size
1436
- Fixed cluster verify error messages for wrong files on regular nodes
1437
- Fixed a couple of issues with node demotion from master candidate role
1438
- Fixed node readd issues
1439
- Added non-interactive mode for ``ganeti-masterd --no-voting`` startup
1440
- Added a new ``--no-voting`` option for masterfailover to fix failover
1441
  on two-nodes clusters when the former master node is unreachable
1442
- Added instance reinstall over RAPI
1443

    
1444

    
1445
Version 2.0.1
1446
-------------
1447

    
1448
*(Released Tue, 16 Jun 2009)*
1449

    
1450
- added ``-H``/``-B`` startup parameters to ``gnt-instance``, which will
1451
  allow re-adding the start in single-user option (regression from 1.2)
1452
- the watcher writes the instance status to a file, to allow monitoring
1453
  to report the instance status (from the master) based on cached
1454
  results of the watcher's queries; while this can get stale if the
1455
  watcher is being locked due to other work on the cluster, this is
1456
  still an improvement
1457
- the watcher now also restarts the node daemon and the rapi daemon if
1458
  they died
1459
- fixed the watcher to handle full and drained queue cases
1460
- hooks export more instance data in the environment, which helps if
1461
  hook scripts need to take action based on the instance's properties
1462
  (no longer need to query back into ganeti)
1463
- instance failovers when the instance is stopped do not check for free
1464
  RAM, so that failing over a stopped instance is possible in low memory
1465
  situations
1466
- rapi uses queries for tags instead of jobs (for less job traffic), and
1467
  for cluster tags it won't talk to masterd at all but read them from
1468
  ssconf
1469
- a couple of error handling fixes in RAPI
1470
- drbd handling: improved the error handling of inconsistent disks after
1471
  resync to reduce the frequency of "there are some degraded disks for
1472
  this instance" messages
1473
- fixed a bug in live migration when DRBD doesn't want to reconnect (the
1474
  error handling path called a wrong function name)
1475

    
1476

    
1477
Version 2.0.0 final
1478
-------------------
1479

    
1480
*(Released Wed, 27 May 2009)*
1481

    
1482
- no changes from rc5
1483

    
1484

    
1485
Version 2.0 release candidate 5
1486
-------------------------------
1487

    
1488
*(Released Wed, 20 May 2009)*
1489

    
1490
- fix a couple of bugs (validation, argument checks)
1491
- fix ``gnt-cluster getmaster`` on non-master nodes (regression)
1492
- some small improvements to RAPI and IAllocator
1493
- make watcher automatically start the master daemon if down
1494

    
1495

    
1496
Version 2.0 release candidate 4
1497
-------------------------------
1498

    
1499
*(Released Mon, 27 Apr 2009)*
1500

    
1501
- change the OS list to not require locks; this helps with big clusters
1502
- fix ``gnt-cluster verify`` and ``gnt-cluster verify-disks`` when the
1503
  volume group is broken
1504
- ``gnt-instance info``, without any arguments, doesn't run for all
1505
  instances anymore; either pass ``--all`` or pass the desired
1506
  instances; this helps against mistakes on big clusters where listing
1507
  the information for all instances takes a long time
1508
- miscellaneous doc and man pages fixes
1509

    
1510

    
1511
Version 2.0 release candidate 3
1512
-------------------------------
1513

    
1514
*(Released Wed, 8 Apr 2009)*
1515

    
1516
- Change the internal locking model of some ``gnt-node`` commands, in
1517
  order to reduce contention (and blocking of master daemon) when
1518
  batching many creation/reinstall jobs
1519
- Fixes to Xen soft reboot
1520
- No longer build documentation at build time, instead distribute it in
1521
  the archive, in order to reduce the need for the whole docbook/rst
1522
  toolchains
1523

    
1524

    
1525
Version 2.0 release candidate 2
1526
-------------------------------
1527

    
1528
*(Released Fri, 27 Mar 2009)*
1529

    
1530
- Now the cfgupgrade scripts works and can upgrade 1.2.7 clusters to 2.0
1531
- Fix watcher startup sequence, improves the behaviour of busy clusters
1532
- Some other fixes in ``gnt-cluster verify``, ``gnt-instance
1533
  replace-disks``, ``gnt-instance add``, ``gnt-cluster queue``, KVM VNC
1534
  bind address and other places
1535
- Some documentation fixes and updates
1536

    
1537

    
1538
Version 2.0 release candidate 1
1539
-------------------------------
1540

    
1541
*(Released Mon, 2 Mar 2009)*
1542

    
1543
- More documentation updates, now all docs should be more-or-less
1544
  up-to-date
1545
- A couple of small fixes (mixed hypervisor clusters, offline nodes,
1546
  etc.)
1547
- Added a customizable HV_KERNEL_ARGS hypervisor parameter (for Xen PVM
1548
  and KVM)
1549
- Fix an issue related to $libdir/run/ganeti and cluster creation
1550

    
1551

    
1552
Version 2.0 beta 2
1553
------------------
1554

    
1555
*(Released Thu, 19 Feb 2009)*
1556

    
1557
- Xen PVM and KVM have switched the default value for the instance root
1558
  disk to the first partition on the first drive, instead of the whole
1559
  drive; this means that the OS installation scripts must be changed
1560
  accordingly
1561
- Man pages have been updated
1562
- RAPI has been switched by default to HTTPS, and the exported functions
1563
  should all work correctly
1564
- RAPI v1 has been removed
1565
- Many improvements to the KVM hypervisor
1566
- Block device errors are now better reported
1567
- Many other bugfixes and small improvements
1568

    
1569

    
1570
Version 2.0 beta 1
1571
------------------
1572

    
1573
*(Released Mon, 26 Jan 2009)*
1574

    
1575
- Version 2 is a general rewrite of the code and therefore the
1576
  differences are too many to list, see the design document for 2.0 in
1577
  the ``doc/`` subdirectory for more details
1578
- In this beta version there is not yet a migration path from 1.2 (there
1579
  will be one in the final 2.0 release)
1580
- A few significant changes are:
1581

    
1582
  - all commands are executed by a daemon (``ganeti-masterd``) and the
1583
    various ``gnt-*`` commands are just front-ends to it
1584
  - all the commands are entered into, and executed from a job queue,
1585
    see the ``gnt-job(8)`` manpage
1586
  - the RAPI daemon supports read-write operations, secured by basic
1587
    HTTP authentication on top of HTTPS
1588
  - DRBD version 0.7 support has been removed, DRBD 8 is the only
1589
    supported version (when migrating from Ganeti 1.2 to 2.0, you need
1590
    to migrate to DRBD 8 first while still running Ganeti 1.2)
1591
  - DRBD devices are using statically allocated minor numbers, which
1592
    will be assigned to existing instances during the migration process
1593
  - there is support for both Xen PVM and Xen HVM instances running on
1594
    the same cluster
1595
  - KVM virtualization is supported too
1596
  - file-based storage has been implemented, which means that it is
1597
    possible to run the cluster without LVM and DRBD storage, for
1598
    example using a shared filesystem exported from shared storage (and
1599
    still have live migration)
1600

    
1601

    
1602
Version 1.2.7
1603
-------------
1604

    
1605
*(Released Tue, 13 Jan 2009)*
1606

    
1607
- Change the default reboot type in ``gnt-instance reboot`` to "hard"
1608
- Reuse the old instance mac address by default on instance import, if
1609
  the instance name is the same.
1610
- Handle situations in which the node info rpc returns incomplete
1611
  results (issue 46)
1612
- Add checks for tcp/udp ports collisions in ``gnt-cluster verify``
1613
- Improved version of batcher:
1614

    
1615
  - state file support
1616
  - instance mac address support
1617
  - support for HVM clusters/instances
1618

    
1619
- Add an option to show the number of cpu sockets and nodes in
1620
  ``gnt-node list``
1621
- Support OSes that handle more than one version of the OS api (but do
1622
  not change the current API in any other way)
1623
- Fix ``gnt-node migrate``
1624
- ``gnt-debug`` man page
1625
- Fixes various more typos and small issues
1626
- Increase disk resync maximum speed to 60MB/s (from 30MB/s)
1627

    
1628

    
1629
Version 1.2.6
1630
-------------
1631

    
1632
*(Released Wed, 24 Sep 2008)*
1633

    
1634
- new ``--hvm-nic-type`` and ``--hvm-disk-type`` flags to control the
1635
  type of disk exported to fully virtualized instances.
1636
- provide access to the serial console of HVM instances
1637
- instance auto_balance flag, set by default. If turned off it will
1638
  avoid warnings on cluster verify if there is not enough memory to fail
1639
  over an instance. in the future it will prevent automatically failing
1640
  it over when we will support that.
1641
- batcher tool for instance creation, see ``tools/README.batcher``
1642
- ``gnt-instance reinstall --select-os`` to interactively select a new
1643
  operating system when reinstalling an instance.
1644
- when changing the memory amount on instance modify a check has been
1645
  added that the instance will be able to start. also warnings are
1646
  emitted if the instance will not be able to fail over, if auto_balance
1647
  is true.
1648
- documentation fixes
1649
- sync fields between ``gnt-instance list/modify/add/import``
1650
- fix a race condition in drbd when the sync speed was set after giving
1651
  the device a remote peer.
1652

    
1653

    
1654
Version 1.2.5
1655
-------------
1656

    
1657
*(Released Tue, 22 Jul 2008)*
1658

    
1659
- note: the allowed size and number of tags per object were reduced
1660
- fix a bug in ``gnt-cluster verify`` with inconsistent volume groups
1661
- fixed twisted 8.x compatibility
1662
- fixed ``gnt-instance replace-disks`` with iallocator
1663
- add TCP keepalives on twisted connections to detect restarted nodes
1664
- disk increase support, see ``gnt-instance grow-disk``
1665
- implement bulk node/instance query for RAPI
1666
- add tags in node/instance listing (optional)
1667
- experimental migration (and live migration) support, read the man page
1668
  for ``gnt-instance migrate``
1669
- the ``ganeti-watcher`` logs are now timestamped, and the watcher also
1670
  has some small improvements in handling its state file
1671

    
1672

    
1673
Version 1.2.4
1674
-------------
1675

    
1676
*(Released Fri, 13 Jun 2008)*
1677

    
1678
- Experimental readonly, REST-based remote API implementation;
1679
  automatically started on master node, TCP port 5080, if enabled by
1680
  ``--enable-rapi`` parameter to configure script.
1681
- Instance allocator support. Add and import instance accept a
1682
  ``--iallocator`` parameter, and call that instance allocator to decide
1683
  which node to use for the instance. The iallocator document describes
1684
  what's expected from an allocator script.
1685
- ``gnt-cluster verify`` N+1 memory redundancy checks: Unless passed the
1686
  ``--no-nplus1-mem`` option ``gnt-cluster verify`` now checks that if a
1687
  node is lost there is still enough memory to fail over the instances
1688
  that reside on it.
1689
- ``gnt-cluster verify`` hooks: it is now possible to add post-hooks to
1690
  ``gnt-cluster verify``, to check for site-specific compliance. All the
1691
  hooks will run, and their output, if any, will be displayed. Any
1692
  failing hook will make the verification return an error value.
1693
- ``gnt-cluster verify`` now checks that its peers are reachable on the
1694
  primary and secondary interfaces
1695
- ``gnt-node add`` now supports the ``--readd`` option, to readd a node
1696
  that is still declared as part of the cluster and has failed.
1697
- ``gnt-* list`` commands now accept a new ``-o +field`` way of
1698
  specifying output fields, that just adds the chosen fields to the
1699
  default ones.
1700
- ``gnt-backup`` now has a new ``remove`` command to delete an existing
1701
  export from the filesystem.
1702
- New per-instance parameters hvm_acpi, hvm_pae and hvm_cdrom_image_path
1703
  have been added. Using them you can enable/disable acpi and pae
1704
  support, and specify a path for a cd image to be exported to the
1705
  instance. These parameters as the name suggest only work on HVM
1706
  clusters.
1707
- When upgrading an HVM cluster to Ganeti 1.2.4, the values for ACPI and
1708
  PAE support will be set to the previously hardcoded values, but the
1709
  (previously hardcoded) path to the CDROM ISO image will be unset and
1710
  if required, needs to be set manually with ``gnt-instance modify``
1711
  after the upgrade.
1712
- The address to which an instance's VNC console is bound is now
1713
  selectable per-instance, rather than being cluster wide. Of course
1714
  this only applies to instances controlled via VNC, so currently just
1715
  applies to HVM clusters.
1716

    
1717

    
1718
Version 1.2.3
1719
-------------
1720

    
1721
*(Released Mon, 18 Feb 2008)*
1722

    
1723
- more tweaks to the disk activation code (especially helpful for DRBD)
1724
- change the default ``gnt-instance list`` output format, now there is
1725
  one combined status field (see the manpage for the exact values this
1726
  field will have)
1727
- some more fixes for the mac export to hooks change
1728
- make Ganeti not break with DRBD 8.2.x (which changed the version
1729
  format in ``/proc/drbd``) (issue 24)
1730
- add an upgrade tool from "remote_raid1" disk template to "drbd" disk
1731
  template, allowing migration from DRBD0.7+MD to DRBD8
1732

    
1733

    
1734
Version 1.2.2
1735
-------------
1736

    
1737
*(Released Wed, 30 Jan 2008)*
1738

    
1739
- fix ``gnt-instance modify`` breakage introduced in 1.2.1 with the HVM
1740
  support (issue 23)
1741
- add command aliases infrastructure and a few aliases
1742
- allow listing of VCPUs in the ``gnt-instance list`` and improve the
1743
  man pages and the ``--help`` option of ``gnt-node
1744
  list``/``gnt-instance list``
1745
- fix ``gnt-backup list`` with down nodes (issue 21)
1746
- change the tools location (move from $pkgdatadir to $pkglibdir/tools)
1747
- fix the dist archive and add a check for including svn/git files in
1748
  the future
1749
- some developer-related changes: improve the burnin and the QA suite,
1750
  add an upload script for testing during development
1751

    
1752

    
1753
Version 1.2.1
1754
-------------
1755

    
1756
*(Released Wed, 16 Jan 2008)*
1757

    
1758
- experimental HVM support, read the install document, section
1759
  "Initializing the cluster"
1760
- allow for the PVM hypervisor per-instance kernel and initrd paths
1761
- add a new command ``gnt-cluster verify-disks`` which uses a new
1762
  algorithm to improve the reconnection of the DRBD pairs if the device
1763
  on the secondary node has gone away
1764
- make logical volume code auto-activate LVs at disk activation time
1765
- slightly improve the speed of activating disks
1766
- allow specification of the MAC address at instance creation time, and
1767
  changing it later via ``gnt-instance modify``
1768
- fix handling of external commands that generate lots of output on
1769
  stderr
1770
- update documentation with regard to minimum version of DRBD8 supported
1771

    
1772

    
1773
Version 1.2.0
1774
-------------
1775

    
1776
*(Released Tue, 4 Dec 2007)*
1777

    
1778
- Log the ``xm create`` output to the node daemon log on failure (to
1779
  help diagnosing the error)
1780
- In debug mode, log all external commands output if failed to the logs
1781
- Change parsing of lvm commands to ignore stderr
1782

    
1783

    
1784
Version 1.2b3
1785
-------------
1786

    
1787
*(Released Wed, 28 Nov 2007)*
1788

    
1789
- Another round of updates to the DRBD 8 code to deal with more failures
1790
  in the replace secondary node operation
1791
- Some more logging of failures in disk operations (lvm, drbd)
1792
- A few documentation updates
1793
- QA updates
1794

    
1795

    
1796
Version 1.2b2
1797
-------------
1798

    
1799
*(Released Tue, 13 Nov 2007)*
1800

    
1801
- Change configuration file format from Python's Pickle to JSON.
1802
  Upgrading is possible using the cfgupgrade utility.
1803
- Add support for DRBD 8.0 (new disk template ``drbd``) which allows for
1804
  faster replace disks and is more stable (DRBD 8 has many improvements
1805
  compared to DRBD 0.7)
1806
- Added command line tags support (see man pages for ``gnt-instance``,
1807
  ``gnt-node``, ``gnt-cluster``)
1808
- Added instance rename support
1809
- Added multi-instance startup/shutdown
1810
- Added cluster rename support
1811
- Added ``gnt-node evacuate`` to simplify some node operations
1812
- Added instance reboot operation that can speedup reboot as compared to
1813
  stop and start
1814
- Soften the requirement that hostnames are in FQDN format
1815
- The ``ganeti-watcher`` now activates drbd pairs after secondary node
1816
  reboots
1817
- Removed dependency on debian's patched fping that uses the
1818
  non-standard ``-S`` option
1819
- Now the OS definitions are searched for in multiple, configurable
1820
  paths (easier for distros to package)
1821
- Some changes to the hooks infrastructure (especially the new
1822
  post-configuration update hook)
1823
- Other small bugfixes
1824

    
1825
.. vim: set textwidth=72 syntax=rst :
1826
.. Local Variables:
1827
.. mode: rst
1828
.. fill-column: 72
1829
.. End: