Statistics
| Branch: | Tag: | Revision:

root / NEWS @ b989b9d9

History | View | Annotate | Download (21.1 kB)

1
News
2
====
3

    
4

    
5
Version 2.1.0
6
-------------
7

    
8
Ganeti 2.1 brings many improvements with it. Major changes:
9

    
10
- Added infrastructure to ease automated disk repairs
11
- Added new daemon to export configuration data in a cheaper way than
12
  using the remote API
13
- Instance NICs can now be routed instead of being associated with a
14
  networking bridge
15
- Improved job locking logic to reduce impact of jobs acquiring multiple
16
  locks waiting for other long-running jobs
17

    
18
Detailed implementation details can be found in the Ganeti 2.1 design
19
document.
20

    
21
Details
22
~~~~~~~
23

    
24
- Added chroot hypervisor
25
- Added more options to xen-hvm hypervisor (``kernel_path`` and
26
  ``device_model``)
27
- Added more options to xen-pvm hypervisor (``use_bootloader``,
28
  ``bootloader_path`` and ``bootloader_args``)
29
- Added luxi call to submit multiple jobs in one go
30
- Added cluster initialization time option to not modify ``/etc/hosts``
31
  file on nodes
32
- Added network interface parameters
33
- Added dry run mode to some LUs
34
- Added RAPI resources:
35

    
36
  - ``/2/instances/[instance_name]/info``
37
  - ``/2/instances/[instance_name]/replace-disks``
38
  - ``/2/nodes/[node_name]/evacuate``
39
  - ``/2/nodes/[node_name]/migrate``
40
  - ``/2/nodes/[node_name]/role``
41
  - ``/2/nodes/[node_name]/storage``
42
  - ``/2/nodes/[node_name]/storage/modify``
43
  - ``/2/nodes/[node_name]/storage/repair``
44

    
45
- Added OpCodes to evacuate or migrate all instances on a node
46
- Added new command to list physical volumes on nodes (``gnt-node
47
  physical-volumes``) and to modify them (``gnt-node modify-volume``)
48
- Added new ssconf files with master candidate IP address
49
  (``ssconf_master_candidates_ips``), node primary IP address
50
  (``ssconf_node_primary_ips``) and node secondary IP address
51
  (``ssconf_node_secondary_ips``)
52
- Added ``ganeti-confd`` and a client library to query the Ganeti
53
  configuration via UDP
54
- Added ability to run hooks after cluster initialization and before
55
  cluster destruction
56
- Added automatic mode for disk replace (``gnt-instance replace-disks
57
  --auto``)
58
- Added ``gnt-instance recreate-disks`` to re-create (empty) disks
59
  after catastrophic data-loss
60
- Added ``gnt-node repair-volume`` command to repair damanged LVM volume
61
  groups
62
- Added ``gnt-instance move`` command to move instances
63
- Added ``gnt-cluster watcher`` command to control watcher
64
- Added ``gnt-node powercycle`` command to powercycle nodes
65
- Added new job status field ``lock_status``
66
- Added parseable error codes to cluster verification (``gnt-cluster
67
  verify --error-codes``) and made output less verbose (use
68
  ``--verbose`` to restore previous behaviour)
69
- Added UUIDs to the main config entities (cluster, nodes, instances)
70
- Added support for OS variants
71
- Added more unittests
72
- Fixed critical bug in ganeti-masterd startup
73
- Pass ``INSTANCE_REINSTALL`` variable to OS installation script when
74
  reinstalling an instance
75
- Converted to Sphinx (http://sphinx.pocoo.org/) for documentation
76
- Many documentation updates
77
- Distribute hypervisor files on ``gnt-cluster redist-conf``
78
- ``gnt-instance reinstall`` can now reinstall multiple instances
79
- Updated many command line parameters
80
- Introduced new OS API version 15
81
- No longer support a default hypervisor
82
- Treat virtual LVs as inexistent
83
- Improved job locking logic to reduce lock contention
84
- Match instance and node names case insensitively
85
- Reimplemented bash completion script to be more complete
86
- Improved burnin
87
- Added option to specify maximum timeout on instance shutdown
88
- Added ``--no-ssh-init`` option to ``gnt-cluster init``
89

    
90

    
91
Version 2.0.4
92
-------------
93

    
94
- Fixed many wrong messages
95
- Fixed a few bugs related to the locking library
96
- Fixed MAC checking at instance creation time
97
- Fixed a DRBD parsing bug related to gaps in /proc/drbd
98
- Fixed a few issues related to signal handling in both daemons and
99
  scripts
100
- Fixed the example startup script provided
101
- Fixed insserv dependencies in the example startup script (patch from
102
  Debian)
103
- Fixed handling of drained nodes in the iallocator framework
104
- Fixed handling of KERNEL_PATH parameter for xen-hvm (Debian bug
105
  #528618)
106
- Fixed error related to invalid job IDs in job polling
107
- Fixed job/opcode persistence on unclean master shutdown
108
- Fixed handling of partial job processing after unclean master
109
  shutdown
110
- Fixed error reporting from LUs, previously all errors were converted
111
  into execution errors
112
- Fixed error reporting from burnin
113
- Decreased significantly the memory usage of the job queue
114
- Optimised slightly multi-job submission
115
- Optimised slightly opcode loading
116
- Backported the multi-job submit framework from the development
117
  branch; multi-instance start and stop should be faster
118
- Added script to clean archived jobs after 21 days; this will reduce
119
  the size of the queue directory
120
- Added some extra checks in disk size tracking
121
- Added an example ethers hook script
122
- Added a cluster parameter that prevents Ganeti from modifying of
123
  /etc/hosts
124
- Added more node information to RAPI responses
125
- Added a ``gnt-job watch`` command that allows following the ouput of a
126
  job
127
- Added a bind-address option to ganeti-rapi
128
- Added more checks to the configuration verify
129
- Enhanced the burnin script such that some operations can be retried
130
  automatically
131
- Converted instance reinstall to multi-instance model
132

    
133

    
134
Version 2.0.3
135
-------------
136

    
137
- Added ``--ignore-size`` to the ``gnt-instance activate-disks`` command
138
  to allow using the pre-2.0.2 behaviour in activation, if any existing
139
  instances have mismatched disk sizes in the configuration
140
- Added ``gnt-cluster repair-disk-sizes`` command to check and update
141
  any configuration mismatches for disk sizes
142
- Added ``gnt-master cluste-failover --no-voting`` to allow master
143
  failover to work on two-node clusters
144
- Fixed the ``--net`` option of ``gnt-backup import``, which was
145
  unusable
146
- Fixed detection of OS script errors in ``gnt-backup export``
147
- Fixed exit code of ``gnt-backup export``
148

    
149

    
150
Version 2.0.2
151
-------------
152

    
153
- Added experimental support for stripped logical volumes; this should
154
  enhance performance but comes with a higher complexity in the block
155
  device handling; stripping is only enabled when passing
156
  ``--with-lvm-stripecount=N`` to ``configure``, but codepaths are
157
  affected even in the non-stripped mode
158
- Improved resiliency against transient failures at the end of DRBD
159
  resyncs, and in general of DRBD resync checks
160
- Fixed a couple of issues with exports and snapshot errors
161
- Fixed a couple of issues in instance listing
162
- Added display of the disk size in ``gnt-instance info``
163
- Fixed checking for valid OSes in instance creation
164
- Fixed handling of the "vcpus" parameter in instance listing and in
165
  general of invalid parameters
166
- Fixed http server library, and thus RAPI, to handle invalid
167
  username/password combinations correctly; this means that now they
168
  report unauthorized for queries too, not only for modifications,
169
  allowing earlier detect of configuration problems
170
- Added a new "role" node list field, equivalent to the master/master
171
  candidate/drained/offline flags combinations
172
- Fixed cluster modify and changes of candidate pool size
173
- Fixed cluster verify error messages for wrong files on regular nodes
174
- Fixed a couple of issues with node demotion from master candidate role
175
- Fixed node readd issues
176
- Added non-interactive mode for ``ganeti-masterd --no-voting`` startup
177
- Added a new ``--no-voting`` option for masterfailover to fix failover
178
  on two-nodes clusters when the former master node is unreachable
179
- Added instance reinstall over RAPI
180

    
181

    
182
Version 2.0.1
183
-------------
184

    
185
- added ``-H``/``-B`` startup parameters to ``gnt-instance``, which will
186
  allow re-adding the start in single-user option (regression from 1.2)
187
- the watcher writes the instance status to a file, to allow monitoring
188
  to report the instance status (from the master) based on cached
189
  results of the watcher's queries; while this can get stale if the
190
  watcher is being locked due to other work on the cluster, this is
191
  still an improvement
192
- the watcher now also restarts the node daemon and the rapi daemon if
193
  they died
194
- fixed the watcher to handle full and drained queue cases
195
- hooks export more instance data in the environment, which helps if
196
  hook scripts need to take action based on the instance's properties
197
  (no longer need to query back into ganeti)
198
- instance failovers when the instance is stopped do not check for free
199
  RAM, so that failing over a stopped instance is possible in low memory
200
  situations
201
- rapi uses queries for tags instead of jobs (for less job traffic), and
202
  for cluster tags it won't talk to masterd at all but read them from
203
  ssconf
204
- a couple of error handling fixes in RAPI
205
- drbd handling: improved the error handling of inconsistent disks after
206
  resync to reduce the frequency of "there are some degraded disks for
207
  this instance" messages
208
- fixed a bug in live migration when DRBD doesn't want to reconnect (the
209
  error handling path called a wrong function name)
210

    
211

    
212
Version 2.0.0 final
213
-------------------
214

    
215
- no changes from rc5
216

    
217

    
218
Version 2.0 release candidate 5
219
-------------------------------
220

    
221
- fix a couple of bugs (validation, argument checks)
222
- fix ``gnt-cluster getmaster`` on non-master nodes (regression)
223
- some small improvements to RAPI and IAllocator
224
- make watcher automatically start the master daemon if down
225

    
226

    
227
Version 2.0 release candidate 4
228
-------------------------------
229

    
230
- change the OS list to not require locks; this helps with big clusters
231
- fix ``gnt-cluster verify`` and ``gnt-cluster verify-disks`` when the
232
  volume group is broken
233
- ``gnt-instance info``, without any arguments, doesn't run for all
234
  instances anymore; either pass ``--all`` or pass the desired
235
  instances; this helps against mistakes on big clusters where listing
236
  the information for all instances takes a long time
237
- miscellaneous doc and man pages fixes
238

    
239

    
240
Version 2.0 release candidate 3
241
-------------------------------
242

    
243
- Change the internal locking model of some ``gnt-node`` commands, in
244
  order to reduce contention (and blocking of master daemon) when
245
  batching many creation/reinstall jobs
246
- Fixes to Xen soft reboot
247
- No longer build documentation at build time, instead distribute it in
248
  the archive, in order to reduce the need for the whole docbook/rst
249
  toolchains
250

    
251

    
252
Version 2.0 release candidate 2
253
-------------------------------
254

    
255
- Now the cfgupgrade scripts works and can upgrade 1.2.7 clusters to 2.0
256
- Fix watcher startup sequence, improves the behaviour of busy clusters
257
- Some other fixes in ``gnt-cluster verify``, ``gnt-instance
258
  replace-disks``, ``gnt-instance add``, ``gnt-cluster queue``, KVM VNC
259
  bind address and other places
260
- Some documentation fixes and updates
261

    
262

    
263
Version 2.0 release candidate 1
264
-------------------------------
265

    
266
- More documentation updates, now all docs should be more-or-less
267
  up-to-date
268
- A couple of small fixes (mixed hypervisor clusters, offline nodes,
269
  etc.)
270
- Added a customizable HV_KERNEL_ARGS hypervisor parameter (for Xen PVM
271
  and KVM)
272
- Fix an issue related to $libdir/run/ganeti and cluster creation
273

    
274

    
275
Version 2.0 beta 2
276
------------------
277

    
278
- Xen PVM and KVM have switched the default value for the instance root
279
  disk to the first partition on the first drive, instead of the whole
280
  drive; this means that the OS installation scripts must be changed
281
  accordingly
282
- Man pages have been updated
283
- RAPI has been switched by default to HTTPS, and the exported functions
284
  should all work correctly
285
- RAPI v1 has been removed
286
- Many improvements to the KVM hypervisor
287
- Block device errors are now better reported
288
- Many other bugfixes and small improvements
289

    
290

    
291
Version 2.0 beta 1
292
------------------
293

    
294
- Version 2 is a general rewrite of the code and therefore the
295
  differences are too many to list, see the design document for 2.0 in
296
  the ``doc/`` subdirectory for more details
297
- In this beta version there is not yet a migration path from 1.2 (there
298
  will be one in the final 2.0 release)
299
- A few significant changes are:
300

    
301
  - all commands are executed by a daemon (``ganeti-masterd``) and the
302
    various ``gnt-*`` commands are just front-ends to it
303
  - all the commands are entered into, and executed from a job queue,
304
    see the ``gnt-job(8)`` manpage
305
  - the RAPI daemon supports read-write operations, secured by basic
306
    HTTP authentication on top of HTTPS
307
  - DRBD version 0.7 support has been removed, DRBD 8 is the only
308
    supported version (when migrating from Ganeti 1.2 to 2.0, you need
309
    to migrate to DRBD 8 first while still running Ganeti 1.2)
310
  - DRBD devices are using statically allocated minor numbers, which
311
    will be assigned to existing instances during the migration process
312
  - there is support for both Xen PVM and Xen HVM instances running on
313
    the same cluster
314
  - KVM virtualization is supported too
315
  - file-based storage has been implemented, which means that it is
316
    possible to run the cluster without LVM and DRBD storage, for
317
    example using a shared filesystem exported from shared storage (and
318
    still have live migration)
319

    
320

    
321
Version 1.2.7
322
-------------
323

    
324
- Change the default reboot type in ``gnt-instance reboot`` to "hard"
325
- Reuse the old instance mac address by default on instance import, if
326
  the instance name is the same.
327
- Handle situations in which the node info rpc returns incomplete
328
  results (issue 46)
329
- Add checks for tcp/udp ports collisions in ``gnt-cluster verify``
330
- Improved version of batcher:
331

    
332
  - state file support
333
  - instance mac address support
334
  - support for HVM clusters/instances
335

    
336
- Add an option to show the number of cpu sockets and nodes in
337
  ``gnt-node list``
338
- Support OSes that handle more than one version of the OS api (but do
339
  not change the current API in any other way)
340
- Fix ``gnt-node migrate``
341
- ``gnt-debug`` man page
342
- Fixes various more typos and small issues
343
- Increase disk resync maximum speed to 60MB/s (from 30MB/s)
344

    
345

    
346
Version 1.2.6
347
-------------
348

    
349
- new ``--hvm-nic-type`` and ``--hvm-disk-type`` flags to control the
350
  type of disk exported to fully virtualized instances.
351
- provide access to the serial console of HVM instances
352
- instance auto_balance flag, set by default. If turned off it will
353
  avoid warnings on cluster verify if there is not enough memory to fail
354
  over an instance. in the future it will prevent automatically failing
355
  it over when we will support that.
356
- batcher tool for instance creation, see ``tools/README.batcher``
357
- ``gnt-instance reinstall --select-os`` to interactively select a new
358
  operating system when reinstalling an instance.
359
- when changing the memory amount on instance modify a check has been
360
  added that the instance will be able to start. also warnings are
361
  emitted if the instance will not be able to fail over, if auto_balance
362
  is true.
363
- documentation fixes
364
- sync fields between ``gnt-instance list/modify/add/import``
365
- fix a race condition in drbd when the sync speed was set after giving
366
  the device a remote peer.
367

    
368

    
369
Version 1.2.5
370
-------------
371

    
372
- note: the allowed size and number of tags per object were reduced
373
- fix a bug in ``gnt-cluster verify`` with inconsistent volume groups
374
- fixed twisted 8.x compatibility
375
- fixed ``gnt-instance replace-disks`` with iallocator
376
- add TCP keepalives on twisted connections to detect restarted nodes
377
- disk increase support, see ``gnt-instance grow-disk``
378
- implement bulk node/instance query for RAPI
379
- add tags in node/instance listing (optional)
380
- experimental migration (and live migration) support, read the man page
381
  for ``gnt-instance migrate``
382
- the ``ganeti-watcher`` logs are now timestamped, and the watcher also
383
  has some small improvements in handling its state file
384

    
385

    
386
Version 1.2.4
387
-------------
388

    
389
- Experimental readonly, REST-based remote API implementation;
390
  automatically started on master node, TCP port 5080, if enabled by
391
  ``--enable-rapi`` parameter to configure script.
392
- Instance allocator support. Add and import instance accept a
393
  ``--iallocator`` parameter, and call that instance allocator to decide
394
  which node to use for the instance. The iallocator document describes
395
  what's expected from an allocator script.
396
- ``gnt-cluster verify`` N+1 memory redundancy checks: Unless passed the
397
  ``--no-nplus1-mem`` option ``gnt-cluster verify`` now checks that if a
398
  node is lost there is still enough memory to fail over the instances
399
  that reside on it.
400
- ``gnt-cluster verify`` hooks: it is now possible to add post-hooks to
401
  ``gnt-cluster verify``, to check for site-specific compliance. All the
402
  hooks will run, and their output, if any, will be displayed. Any
403
  failing hook will make the verification return an error value.
404
- ``gnt-cluster verify`` now checks that its peers are reachable on the
405
  primary and secondary interfaces
406
- ``gnt-node add`` now supports the ``--readd`` option, to readd a node
407
  that is still declared as part of the cluster and has failed.
408
- ``gnt-* list`` commands now accept a new ``-o +field`` way of
409
  specifying output fields, that just adds the chosen fields to the
410
  default ones.
411
- ``gnt-backup`` now has a new ``remove`` command to delete an existing
412
  export from the filesystem.
413
- New per-instance parameters hvm_acpi, hvm_pae and hvm_cdrom_image_path
414
  have been added. Using them you can enable/disable acpi and pae
415
  support, and specify a path for a cd image to be exported to the
416
  instance. These parameters as the name suggest only work on HVM
417
  clusters.
418
- When upgrading an HVM cluster to Ganeti 1.2.4, the values for ACPI and
419
  PAE support will be set to the previously hardcoded values, but the
420
  (previously hardcoded) path to the CDROM ISO image will be unset and
421
  if required, needs to be set manually with ``gnt-instance modify``
422
  after the upgrade.
423
- The address to which an instance's VNC console is bound is now
424
  selectable per-instance, rather than being cluster wide. Of course
425
  this only applies to instances controlled via VNC, so currently just
426
  applies to HVM clusters.
427

    
428

    
429
Version 1.2.3
430
-------------
431

    
432
- more tweaks to the disk activation code (especially helpful for DRBD)
433
- change the default ``gnt-instance list`` output format, now there is
434
  one combined status field (see the manpage for the exact values this
435
  field will have)
436
- some more fixes for the mac export to hooks change
437
- make Ganeti not break with DRBD 8.2.x (which changed the version
438
  format in ``/proc/drbd``) (issue 24)
439
- add an upgrade tool from "remote_raid1" disk template to "drbd" disk
440
  template, allowing migration from DRBD0.7+MD to DRBD8
441

    
442

    
443
Version 1.2.2
444
-------------
445

    
446
- fix ``gnt-instance modify`` breakage introduced in 1.2.1 with the HVM
447
  support (issue 23)
448
- add command aliases infrastructure and a few aliases
449
- allow listing of VCPUs in the ``gnt-instance list`` and improve the
450
  man pages and the ``--help`` option of ``gnt-node
451
  list``/``gnt-instance list``
452
- fix ``gnt-backup list`` with down nodes (issue 21)
453
- change the tools location (move from $pkgdatadir to $pkglibdir/tools)
454
- fix the dist archive and add a check for including svn/git files in
455
  the future
456
- some developer-related changes: improve the burnin and the QA suite,
457
  add an upload script for testing during development
458

    
459

    
460
Version 1.2.1
461
-------------
462

    
463
- experimental HVM support, read the install document, section
464
  "Initializing the cluster"
465
- allow for the PVM hypervisor per-instance kernel and initrd paths
466
- add a new command ``gnt-cluster verify-disks`` which uses a new
467
  algorithm to improve the reconnection of the DRBD pairs if the device
468
  on the secondary node has gone away
469
- make logical volume code auto-activate LVs at disk activation time
470
- slightly improve the speed of activating disks
471
- allow specification of the MAC address at instance creation time, and
472
  changing it later via ``gnt-instance modify``
473
- fix handling of external commands that generate lots of output on
474
  stderr
475
- update documentation with regard to minimum version of DRBD8 supported
476

    
477

    
478
Version 1.2.0
479
-------------
480

    
481
- Log the ``xm create`` output to the node daemon log on failure (to
482
  help diagnosing the error)
483
- In debug mode, log all external commands output if failed to the logs
484
- Change parsing of lvm commands to ignore stderr
485

    
486

    
487
Version 1.2b3
488
-------------
489

    
490
- Another round of updates to the DRBD 8 code to deal with more failures
491
  in the replace secondary node operation
492
- Some more logging of failures in disk operations (lvm, drbd)
493
- A few documentation updates
494
- QA updates
495

    
496

    
497
Version 1.2b2
498
-------------
499

    
500
- Change configuration file format from Python's Pickle to JSON.
501
  Upgrading is possible using the cfgupgrade utility.
502
- Add support for DRBD 8.0 (new disk template ``drbd``) which allows for
503
  faster replace disks and is more stable (DRBD 8 has many improvements
504
  compared to DRBD 0.7)
505
- Added command line tags support (see man pages for ``gnt-instance``,
506
  ``gnt-node``, ``gnt-cluster``)
507
- Added instance rename support
508
- Added multi-instance startup/shutdown
509
- Added cluster rename support
510
- Added ``gnt-node evacuate`` to simplify some node operations
511
- Added instance reboot operation that can speedup reboot as compared to
512
  stop and start
513
- Soften the requirement that hostnames are in FQDN format
514
- The ``ganeti-watcher`` now activates drbd pairs after secondary node
515
  reboots
516
- Removed dependency on debian's patched fping that uses the
517
  non-standard ``-S`` option
518
- Now the OS definitions are searched for in multiple, configurable
519
  paths (easier for distros to package)
520
- Some changes to the hooks infrastructure (especially the new
521
  post-configuration update hook)
522
- Other small bugfixes
523

    
524
.. vim: set textwidth=72 :