8 *(Released Mon, 25 Jun 2012)*
16 The current ``admin_up`` field, which used to denote whether an instance
17 should be running or not, has been removed. Instead, ``admin_state`` is
18 introduced, with 3 possible values -- ``up``, ``down`` and ``offline``.
20 The rational behind this is that an instance being “down” can have
23 - it could be down during a reboot
24 - it could be temporarily be down for a reinstall
25 - or it could be down because it is deprecated and kept just for its
28 The previous Boolean state was making it difficult to do capacity
29 calculations: should Ganeti reserve memory for a down instance? Now, the
30 tri-state field makes it clear:
32 - in ``up`` and ``down`` state, all resources are reserved for the
33 instance, and it can be at any time brought up if it is down
34 - in ``offline`` state, only disk space is reserved for it, but not
37 The field can have an extra use: since the transition between ``up`` and
38 ``down`` and vice-versus is done via ``gnt-instance start/stop``, but
39 transition between ``offline`` and ``down`` is done via ``gnt-instance
40 modify``, it is possible to given different rights to users. For
41 example, owners of an instance could be allowed to start/stop it, but
42 not transition it out of the offline state.
44 Instance policies and specs
45 +++++++++++++++++++++++++++
47 In previous Ganeti versions, an instance creation request was not
48 limited on the minimum size and on the maximum size just by the cluster
49 resources. As such, any policy could be implemented only in third-party
50 clients (RAPI clients, or shell wrappers over ``gnt-*``
51 tools). Furthermore, calculating cluster capacity via ``hspace`` again
52 required external input with regards to instance sizes.
54 In order to improve these workflows and to allow for example better
55 per-node group differentiation, we introduced instance specs, which
58 - minimum instance disk size, disk count, memory size, cpu count
59 - maximum values for the above metrics
60 - and “standard” values (used in ``hspace`` to calculate the standard
63 The minimum/maximum values can be also customised at node-group level,
64 for example allowing more powerful hardware to support bigger instance
67 Beside the instance specs, there are a few other settings belonging to
68 the instance policy framework. It is possible now to customise, per
69 cluster and node-group:
71 - the list of allowed disk templates
72 - the maximum ratio of VCPUs per PCPUs (to control CPU oversubscription)
73 - the maximum ratio of instance to spindles (see below for more
74 information) for local storage
76 All these together should allow all tools that talk to Ganeti to know
77 what are the ranges of allowed values for instances and the
78 over-subscription that is allowed.
80 For the VCPU/PCPU ratio, we already have the VCPU configuration from the
81 instance configuration, and the physical CPU configuration from the
82 node. For the spindle ratios however, we didn't track before these
83 values, so new parameters have been added:
85 - a new node parameter ``spindle_count``, defaults to 1, customisable at
86 node group or node level
87 - at new backend parameter (for instances), ``spindle_use`` defaults to 1
89 Note that spindles in this context doesn't need to mean actual
90 mechanical hard-drives; it's just a relative number for both the node
91 I/O capacity and instance I/O consumption.
93 Instance migration behaviour
94 ++++++++++++++++++++++++++++
96 While live-migration is in general desirable over failover, it is
97 possible that for some workloads it is actually worse, due to the
98 variable time of the “suspend” phase during live migration.
100 To allow the tools to work consistently over such instances (without
101 having to hard-code instance names), a new backend parameter
102 ``always_failover`` has been added to control the migration/failover
103 behaviour. When set to True, all migration requests for an instance will
104 instead fall-back to failover.
106 Instance memory ballooning
107 ++++++++++++++++++++++++++
109 Initial support for memory ballooning has been added. The memory for an
110 instance is no longer fixed (backend parameter ``memory``), but instead
111 can vary between minimum and maximum values (backend parameters
112 ``minmem`` and ``maxmem``). Currently we only change an instance's
115 - live migrating or failing over and instance and the target node
116 doesn't have enough memory
117 - user requests changing the memory via ``gnt-instance modify
123 In order to control the use of specific CPUs by instance, support for
124 controlling CPU pinning has been added for the Xen, HVM and LXC
125 hypervisors. This is controlled by a new hypervisor parameter
126 ``cpu_mask``; details about possible values for this are in the
127 :manpage:`gnt-instance(8)`. Note that use of the most specific (precise
128 VCPU-to-CPU mapping) form will work well only when all nodes in your
129 cluster have the same amount of CPUs.
134 Another area in which Ganeti was not customisable were the parameters
135 used for storage configuration, e.g. how many stripes to use for LVM,
136 DRBD resync configuration, etc.
138 To improve this area, we've added disks parameters, which are
139 customisable at cluster and node group level, and which allow to
140 specify various parameters for disks (DRBD has the most parameters
141 currently), for example:
143 - DRBD resync algorithm and parameters (e.g. speed)
144 - the default VG for meta-data volumes for DRBD
145 - number of stripes for LVM (plain disk template)
148 These parameters can be modified via ``gnt-cluster modify -D …`` and
149 ``gnt-group modify -D …``, and are used at either instance creation (in
150 case of LVM stripes, for example) or at disk “activation” time
153 Rados block device support
154 ++++++++++++++++++++++++++
156 A Rados (http://ceph.com/wiki/Rbd) storage backend has been added,
157 denoted by the ``rbd`` disk template type. This is considered
158 experimental, feedback is welcome. For details on configuring it, see
159 the :doc:`install` document and the :manpage:`gnt-cluster(8)` man page.
164 The existing master IP functionality works well only in simple setups (a
165 single network shared by all nodes); however, if nodes belong to
166 different networks, then the ``/32`` setup and lack of routing
167 information is not enough.
169 To allow the master IP to function well in more complex cases, the
170 system was reworked as follows:
172 - a master IP netmask setting has been added
173 - the master IP activation/turn-down code was moved from the node daemon
175 - whether to run the Ganeti-supplied master IP script or a user-supplied
176 on is a ``gnt-cluster init`` setting
178 Details about the location of the standard and custom setup scripts are
179 in the man page :manpage:`gnt-cluster(8)`; for information about the
180 setup script protocol, look at the Ganeti-supplied script.
185 The `SPICE <http://www.linux-kvm.org/page/SPICE>`_ support has been
188 It is now possible to use TLS-protected connections, and when renewing
189 or changing the cluster certificates (via ``gnt-cluster renew-crypto``,
190 it is now possible to specify spice or spice CA certificates. Also, it
191 is possible to configure a password for SPICE sessions via the
192 hypervisor parameter ``spice_password_file``.
194 There are also new parameters to control the compression and streaming
195 options (e.g. ``spice_image_compression``, ``spice_streaming_video``,
196 etc.). For details, see the man page :manpage:`gnt-instance(8)` and look
197 for the spice parameters.
199 Lastly, it is now possible to see the SPICE connection information via
200 ``gnt-instance console``.
205 A new tool (``tools/ovfconverter``) has been added that supports
206 conversion between Ganeti and the `Open Virtualization Format
207 <http://en.wikipedia.org/wiki/Open_Virtualization_Format>`_ (both to and
210 This relies on the ``qemu-img`` tool to convert the disk formats, so the
211 actual compatibility with other virtualization solutions depends on it.
216 The configuration query daemon (``ganeti-confd``) is now optional, and
217 has been rewritten in Haskell; whether to use the daemon at all, use the
218 Python (default) or the Haskell version is selectable at configure time
219 via the ``--enable-confd`` parameter, which can take one of the
220 ``haskell``, ``python`` or ``no`` values. If not used, disabling the
221 daemon will result in a smaller footprint; for larger systems, we
222 welcome feedback on the Haskell version which might become the default
226 User interface changes
227 ~~~~~~~~~~~~~~~~~~~~~~
229 We have replaced the ``--disks`` option of ``gnt-instance
230 replace-disks`` with a more flexible ``--disk`` option, which allows
231 adding and removing disks at arbitrary indices (Issue 188). Furthermore,
232 disk size and mode can be changed upon recreation (via ``gnt-instance
233 recreate-disks``, which accepts the same ``--disk`` option).
235 As many people are used to a ``show`` command, we have added that as an
236 alias to ``info`` on all ``gnt-*`` commands.
238 The ``gnt-instance grow-disk`` command has a new mode in which it can
239 accept the target size of the disk, instead of the delta; this can be
240 more safe since two runs in absolute mode will be idempotent, and
241 sometimes it's also easier to specify the desired size directly.
246 RAPI coverage has improved, with (for example) new resources for
247 recreate-disks, node power-cycle, etc.
252 There is partial support for ``xl`` in the Xen hypervisor; feedback is
255 Python 2.7 is better supported, and after Ganeti 2.6 we will investigate
256 whether to still support Python 2.4 or move to Python 2.6 as minimum
259 Support for Fedora has been slightly improved; the provided example
260 init.d script should work better on it and the INSTALL file should
261 document the needed dependencies.
266 The deprecated ``QueryLocks`` LUXI request has been removed. Use
267 ``Query(what=QR_LOCK, ...)`` instead.
269 The LUXI requests :pyeval:`luxi.REQ_QUERY_JOBS`,
270 :pyeval:`luxi.REQ_QUERY_INSTANCES`, :pyeval:`luxi.REQ_QUERY_NODES`,
271 :pyeval:`luxi.REQ_QUERY_GROUPS`, :pyeval:`luxi.REQ_QUERY_EXPORTS` and
272 :pyeval:`luxi.REQ_QUERY_TAGS` are deprecated and will be removed in a
273 future version. :pyeval:`luxi.REQ_QUERY` should be used instead.
275 RAPI client: ``CertificateError`` now derives from
276 ``GanetiApiError``. This should make it more easy to handle Ganeti
279 Deprecation warnings due to PyCrypto/paramiko import in
280 ``tools/setup-ssh`` have been silenced, as usually they are safe; please
281 make sure to run an up-to-date paramiko version, if you use this tool.
283 The QA scripts now depend on Python 2.5 or above (the main code base
284 still works with Python 2.4).
286 The configuration file (``config.data``) is now written without
287 indentation for performance reasons; if you want to edit it, it can be
288 re-formatted via ``tools/fmtjson``.
290 A number of bugs has been fixed in the cluster merge tool.
292 ``x509`` certification verification (used in import-export) has been
293 changed to allow the same clock skew as permitted by the cluster
294 verification. This will remove some rare but hard to diagnose errors in
301 *(Released Mon, 11 Jun 2012)*
303 Second beta release of 2.6. The following changes were done from beta2
306 - Fixed ``daemon-util`` with non-root user models
307 - Fixed creation of plain instances with ``--no-wait-for-sync``
308 - Fix wrong iv_names when running ``cfgupgrade``
309 - Export more information in RAPI group queries
310 - Fixed bug when changing instance network interfaces
311 - Extended burnin to do NIC changes
312 - query: Added ``<``, ``>``, ``<=``, ``>=`` comparison operators
313 - Changed default for DRBD barriers
314 - Fixed DRBD error reporting for syncer rate
315 - Verify the options on disk parameters
317 And of course various fixes to documentation and improved unittests and
324 *(Released Wed, 23 May 2012)*
326 First beta release of 2.6. The following changes were done from beta1 to
329 - integrated patch for distributions without ``start-stop-daemon``
330 - adapted example init.d script to work on Fedora
331 - fixed log handling in Haskell daemons
332 - adapted checks in the watcher for pycurl linked against libnss
333 - add partial support for ``xl`` instead of ``xm`` for Xen
334 - fixed a type issue in cluster verification
335 - fixed ssconf handling in the Haskell code (was breaking confd in IPv6
338 Plus integrated fixes from the 2.5 branch:
340 - fixed ``kvm-ifup`` to use ``/bin/bash``
341 - fixed parallel build failures
342 - KVM live migration when using a custom keymap
348 *(Released Fri, 11 May 2012)*
350 A small bugfix release.
352 The main issues solved are on the topic of compatibility with newer LVM
355 - fixed parsing of ``lv_attr`` field
356 - adapted to new ``vgreduce --removemissing`` behaviour where sometimes
357 the ``--force`` flag is needed
359 Also on the topic of compatibility, ``tools/lvmstrap`` has been changed
360 to accept kernel 3.x too (was hardcoded to 2.6.*).
362 A regression present in 2.5.0 that broke handling (in the gnt-* scripts)
363 of hook results and that also made display of other errors suboptimal
364 was fixed; the code behaves now like 2.4 and earlier.
366 Another change in 2.5, the cleanup of the OS scripts environment, is too
367 aggressive: it removed even the ``PATH`` variable, which requires the OS
368 scripts to *always* need to export it. Since this is a bit too strict,
369 we now export a minimal PATH, the same that we export for hooks.
371 The fix for issue 201 (Preserve bridge MTU in KVM ifup script) was
372 integrated into this release.
374 Finally, a few other miscellaneous changes were done (no new features,
375 just small improvements):
377 - Fix ``gnt-group --help`` display
378 - Fix hardcoded Xen kernel path
379 - Fix grow-disk handling of invalid units
380 - Update synopsis for ``gnt-cluster repair-disk-sizes``
381 - Accept both PUT and POST in noded (makes future upgrade to 2.6 easier)
387 *(Released Thu, 12 Apr 2012)*
389 Incompatible/important changes and bugfixes
390 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
392 - The default of the ``/2/instances/[instance_name]/rename`` RAPI
393 resource's ``ip_check`` parameter changed from ``True`` to ``False``
394 to match the underlying LUXI interface.
395 - The ``/2/nodes/[node_name]/evacuate`` RAPI resource was changed to use
396 body parameters, see :doc:`RAPI documentation <rapi>`. The server does
397 not maintain backwards-compatibility as the underlying operation
398 changed in an incompatible way. The RAPI client can talk to old
399 servers, but it needs to be told so as the return value changed.
400 - When creating file-based instances via RAPI, the ``file_driver``
401 parameter no longer defaults to ``loop`` and must be specified.
402 - The deprecated ``bridge`` NIC parameter is no longer supported. Use
404 - Support for the undocumented and deprecated RAPI instance creation
405 request format version 0 has been dropped. Use version 1, supported
406 since Ganeti 2.1.3 and :doc:`documented <rapi>`, instead.
407 - Pyparsing 1.4.6 or above is required, see :doc:`installation
408 documentation <install>`.
409 - The "cluster-verify" hooks are now executed per group by the
410 ``OP_CLUSTER_VERIFY_GROUP`` opcode. This maintains the same behavior
411 if you just run ``gnt-cluster verify``, which generates one opcode per
413 - The environment as passed to the OS scripts is cleared, and thus no
414 environment variables defined in the node daemon's environment will be
415 inherited by the scripts.
416 - The :doc:`iallocator <iallocator>` mode ``multi-evacuate`` has been
418 - :doc:`New iallocator modes <design-multi-reloc>` have been added to
419 support operations involving multiple node groups.
420 - Offline nodes are ignored when failing over an instance.
421 - Support for KVM version 1.0, which changed the version reporting format
423 - TCP/IP ports used by DRBD disks are returned to a pool upon instance
425 - ``Makefile`` is now compatible with Automake 1.11.2
426 - Includes all bugfixes made in the 2.4 series
431 - The ganeti-htools project has been merged into the ganeti-core source
432 tree and will be built as part of Ganeti (see :doc:`install-quick`).
433 - Implemented support for :doc:`shared storage <design-shared-storage>`.
434 - Add support for disks larger than 2 TB in ``lvmstrap`` by supporting
435 GPT-style partition tables (requires `parted
436 <http://www.gnu.org/s/parted/>`_).
437 - Added support for floppy drive and 2nd CD-ROM drive in KVM hypervisor.
438 - Allowed adding tags on instance creation.
439 - Export instance tags to hooks (``INSTANCE_TAGS``, see :doc:`hooks`)
440 - Allow instances to be started in a paused state, enabling the user to
441 see the complete console output on boot using the console.
442 - Added new hypervisor flag to control default reboot behaviour
443 (``reboot_behavior``).
444 - Added support for KVM keymaps (hypervisor parameter ``keymap``).
445 - Improved out-of-band management support:
447 - Added ``gnt-node health`` command reporting the health status of
449 - Added ``gnt-node power`` command to manage power status of nodes.
450 - Added command for emergency power-off (EPO), ``gnt-cluster epo``.
452 - Instance migration can fall back to failover if instance is not
454 - Filters can be used when listing nodes, instances, groups and locks;
455 see *ganeti(7)* manpage.
456 - Added post-execution status as variables to :doc:`hooks <hooks>`
458 - Instance tags are exported/imported together with the instance.
459 - When given an explicit job ID, ``gnt-job info`` will work for archived
461 - Jobs can define dependencies on other jobs (not yet supported via
462 RAPI or command line, but used by internal commands and usable via
465 - Lock monitor (``gnt-debug locks``) shows jobs waiting for
468 - Instance failover is now available as a RAPI resource
469 (``/2/instances/[instance_name]/failover``).
470 - ``gnt-instance info`` defaults to static information if primary node
472 - Opcodes have a new ``comment`` attribute.
473 - Added basic SPICE support to KVM hypervisor.
474 - ``tools/ganeti-listrunner`` allows passing of arguments to executable.
476 Node group improvements
477 ~~~~~~~~~~~~~~~~~~~~~~~
479 - ``gnt-cluster verify`` has been modified to check groups separately,
480 thereby improving performance.
481 - Node group support has been added to ``gnt-cluster verify-disks``,
482 which now operates per node group.
483 - Watcher has been changed to work better with node groups.
485 - One process and state file per node group.
486 - Slow watcher in one group doesn't block other group's watcher.
488 - Added new command, ``gnt-group evacuate``, to move all instances in a
489 node group to other groups.
490 - Added ``gnt-instance change-group`` to move an instance to another
492 - ``gnt-cluster command`` and ``gnt-cluster copyfile`` now support
493 per-group operations.
494 - Node groups can be tagged.
495 - Some operations switch from an exclusive to a shared lock as soon as
497 - Instance's primary and secondary nodes' groups are now available as
498 query fields (``pnode.group``, ``pnode.group.uuid``, ``snodes.group``
499 and ``snodes.group.uuid``).
504 - Numerous updates to documentation and manpages.
506 - :doc:`RAPI <rapi>` documentation now has detailed parameter
508 - Some opcode/job results are now also documented, see :doc:`RAPI
511 - A lockset's internal lock is now also visible in lock monitor.
512 - Log messages from job queue workers now contain information about the
513 opcode they're processing.
514 - ``gnt-instance console`` no longer requires the instance lock.
515 - A short delay when waiting for job changes reduces the number of LUXI
516 requests significantly.
517 - DRBD metadata volumes are overwritten with zeros during disk creation.
518 - Out-of-band commands no longer acquire the cluster lock in exclusive
520 - ``devel/upload`` now uses correct permissions for directories.
526 *(Released Fri, 23 Mar 2012)*
528 This was the sixth release candidate of the 2.5 series.
534 *(Released Mon, 9 Jan 2012)*
536 This was the fifth release candidate of the 2.5 series.
542 *(Released Thu, 27 Oct 2011)*
544 This was the fourth release candidate of the 2.5 series.
550 *(Released Wed, 26 Oct 2011)*
552 This was the third release candidate of the 2.5 series.
558 *(Released Tue, 18 Oct 2011)*
560 This was the second release candidate of the 2.5 series.
566 *(Released Tue, 4 Oct 2011)*
568 This was the first release candidate of the 2.5 series.
574 *(Released Wed, 31 Aug 2011)*
576 This was the third beta release of the 2.5 series.
582 *(Released Mon, 22 Aug 2011)*
584 This was the second beta release of the 2.5 series.
590 *(Released Fri, 12 Aug 2011)*
592 This was the first beta release of the 2.5 series.
598 *(Released Thu, 27 Oct 2011)*
600 - Fixed bug when parsing command line parameter values ending in
602 - Fixed assertion error after unclean master shutdown
603 - Disable HTTP client pool for RPC, significantly reducing memory usage
605 - Fixed queue archive creation with wrong permissions
611 *(Released Tue, 23 Aug 2011)*
615 - Fixed documentation for importing with ``--src-dir`` option
616 - Fixed a bug in ``ensure-dirs`` with queue/archive permissions
617 - Fixed a parsing issue with DRBD 8.3.11 in the Linux kernel
623 *(Released Fri, 5 Aug 2011)*
625 Many bug-fixes and a few small features:
627 - Fixed argument order in ``ReserveLV`` and ``ReserveMAC`` which caused
628 issues when you tried to add an instance with two MAC addresses in one
630 - KVM: fixed per-instance stored UID value
631 - KVM: configure bridged NICs at migration start
632 - KVM: Fix a bug where instance will not start with never KVM versions
634 - Added OS search path to ``gnt-cluster info``
635 - Fixed an issue with ``file_storage_dir`` where you were forced to
636 provide an absolute path, but the documentation states it is a
637 relative path, the documentation was right
638 - Added a new parameter to instance stop/start called ``--no-remember``
639 that will make the state change to not be remembered
640 - Implemented ``no_remember`` at RAPI level
641 - Improved the documentation
642 - Node evacuation: don't call IAllocator if node is already empty
643 - Fixed bug in DRBD8 replace disks on current nodes
644 - Fixed bug in recreate-disks for DRBD instances
645 - Moved assertion checking locks in ``gnt-instance replace-disks``
646 causing it to abort with not owning the right locks for some situation
647 - Job queue: Fixed potential race condition when cancelling queued jobs
648 - Fixed off-by-one bug in job serial generation
649 - ``gnt-node volumes``: Fix instance names
650 - Fixed aliases in bash completion
651 - Fixed a bug in reopening log files after being sent a SIGHUP
652 - Added a flag to burnin to allow specifying VCPU count
653 - Bugfixes to non-root Ganeti configuration
659 *(Released Thu, 12 May 2011)*
661 Many bug-fixes and a few new small features:
663 - Fixed a bug related to log opening failures
664 - Fixed a bug in instance listing with orphan instances
665 - Fixed a bug which prevented resetting the cluster-level node parameter
666 ``oob_program`` to the default
667 - Many fixes related to the ``cluster-merge`` tool
668 - Fixed a race condition in the lock monitor, which caused failures
669 during (at least) creation of many instances in parallel
670 - Improved output for gnt-job info
671 - Removed the quiet flag on some ssh calls which prevented debugging
673 - Improved the N+1 failure messages in cluster verify by actually
674 showing the memory values (needed and available)
675 - Increased lock attempt timeouts so that when executing long operations
676 (e.g. DRBD replace-disks) other jobs do not enter 'blocking acquire'
677 too early and thus prevent the use of the 'fair' mechanism
678 - Changed instance query data (``gnt-instance info``) to not acquire
679 locks unless needed, thus allowing its use on locked instance if only
680 static information is asked for
681 - Improved behaviour with filesystems that do not support rename on an
683 - Fixed the behaviour of ``prealloc_wipe_disks`` cluster parameter which
684 kept locks on all nodes during the wipe, which is unneeded
685 - Fixed ``gnt-watcher`` handling of errors during hooks execution
686 - Fixed bug in ``prealloc_wipe_disks`` with small disk sizes (less than
687 10GiB) which caused the wipe to fail right at the end in some cases
688 - Fixed master IP activation when doing master failover with no-voting
689 - Fixed bug in ``gnt-node add --readd`` which allowed the re-adding of
690 the master node itself
691 - Fixed potential data-loss in under disk full conditions, where Ganeti
692 wouldn't check correctly the return code and would consider
693 partially-written files 'correct'
694 - Fixed bug related to multiple VGs and DRBD disk replacing
695 - Added new disk parameter ``metavg`` that allows placement of the meta
696 device for DRBD in a different volume group
697 - Fixed error handling in the node daemon when the system libc doesn't
698 have major number 6 (i.e. if ``libc.so.6`` is not the actual libc)
699 - Fixed lock release during replace-disks, which kept cluster-wide locks
700 when doing disk replaces with an iallocator script
701 - Added check for missing bridges in cluster verify
702 - Handle EPIPE errors while writing to the terminal better, so that
703 piping the output to e.g. ``less`` doesn't cause a backtrace
704 - Fixed rare case where a ^C during Luxi calls could have been
705 interpreted as server errors, instead of simply terminating
706 - Fixed a race condition in LUGroupAssignNodes (``gnt-group
708 - Added a few more parameters to the KVM hypervisor, allowing a second
709 CDROM, custom disk type for CDROMs and a floppy image
710 - Removed redundant message in instance rename when the name is given
712 - Added option to ``gnt-instance recreate-disks`` to allow creating the
713 disks on new nodes, allowing recreation when the original instance
714 nodes are completely gone
715 - Added option when converting disk templates to DRBD to skip waiting
716 for the resync, in order to make the instance available sooner
717 - Added two new variables to the OS scripts environment (containing the
719 - Made the root_path and optional parameter for the xen-pvm hypervisor,
720 to allow use of ``pvgrub`` as bootloader
721 - Changed the instance memory modifications to only check out-of-memory
722 conditions on memory increases, and turned the secondary node warnings
723 into errors (they can still be overridden via ``--force``)
724 - Fixed the handling of a corner case when the Python installation gets
725 corrupted (e.g. a bad disk) while ganeti-noded is running and we try
726 to execute a command that doesn't exist
727 - Fixed a bug in ``gnt-instance move`` (LUInstanceMove) when the primary
728 node of the instance returned failures during instance shutdown; this
729 adds the option ``--ignore-consistency`` to gnt-instance move
731 And as usual, various improvements to the error messages, documentation
738 *(Released Wed, 09 Mar 2011)*
740 Emergency bug-fix release. ``tools/cfgupgrade`` was broken and overwrote
741 the RAPI users file if run twice (even with ``--dry-run``).
743 The release fixes that bug (nothing else changed).
749 *(Released Mon, 07 Mar 2011)*
751 Final 2.4.0 release. Just a few small fixes:
753 - Fixed RAPI node evacuate
754 - Fixed the kvm-ifup script
755 - Fixed internal error handling for special job cases
756 - Updated man page to specify the escaping feature for options
762 *(Released Mon, 28 Feb 2011)*
764 A critical fix for the ``prealloc_wipe_disks`` feature: it is possible
765 that this feature wiped the disks of the wrong instance, leading to loss
770 - Fixed title of query field containing instance name
771 - Expanded the glossary in the documentation
772 - Fixed one unittest (internal issue)
778 *(Released Mon, 21 Feb 2011)*
780 A number of bug fixes plus just a couple functionality changes.
782 On the user-visible side, the ``gnt-* list`` command output has changed
783 with respect to "special" field states. The current rc1 style of display
784 can be re-enabled by passing a new ``--verbose`` (``-v``) flag, but in
785 the default output mode special fields states are displayed as follows:
787 - Offline resource: ``*``
788 - Unavailable/not applicable: ``-``
789 - Data missing (RPC failure): ``?``
790 - Unknown field: ``??``
792 Another user-visible change is the addition of ``--force-join`` to
797 - ``tools/cluster-merge`` has seen many fixes and is now enabled again
798 - Fixed regression in RAPI/instance reinstall where all parameters were
799 required (instead of optional)
800 - Fixed ``gnt-cluster repair-disk-sizes``, was broken since Ganeti 2.2
801 - Fixed iallocator usage (offline nodes were not considered offline)
802 - Fixed ``gnt-node list`` with respect to non-vm_capable nodes
803 - Fixed hypervisor and OS parameter validation with respect to
805 - Fixed ``gnt-cluster verify`` with respect to offline nodes (mostly
807 - Fixed ``tools/listrunner`` with respect to agent-based usage
813 *(Released Fri, 4 Feb 2011)*
815 Many changes and fixes since the beta1 release. While there were some
816 internal changes, the code has been mostly stabilised for the RC
819 Note: the dumb allocator was removed in this release, as it was not kept
820 up-to-date with the IAllocator protocol changes. It is recommended to
821 use the ``hail`` command from the ganeti-htools package.
823 Note: the 2.4 and up versions of Ganeti are not compatible with the
824 0.2.x branch of ganeti-htools. You need to upgrade to
825 ganeti-htools-0.3.0 (or later).
827 Regressions fixed from 2.3
828 ~~~~~~~~~~~~~~~~~~~~~~~~~~
830 - Fixed the ``gnt-cluster verify-disks`` command
831 - Made ``gnt-cluster verify-disks`` work in parallel (as opposed to
833 - Fixed disk adoption breakage
834 - Fixed wrong headers in instance listing for field aliases
839 - Fixed corner case in KVM handling of NICs
840 - Fixed many cases of wrong handling of non-vm_capable nodes
841 - Fixed a bug where a missing instance symlink was not possible to
842 recreate with any ``gnt-*`` command (now ``gnt-instance
843 activate-disks`` does it)
844 - Fixed the volume group name as reported by ``gnt-cluster
846 - Increased timeouts for the import-export code, hopefully leading to
847 fewer aborts due network or instance timeouts
848 - Fixed bug in ``gnt-node list-storage``
849 - Fixed bug where not all daemons were started on cluster
850 initialisation, but only at the first watcher run
851 - Fixed many bugs in the OOB implementation
852 - Fixed watcher behaviour in presence of instances with offline
854 - Fixed instance list output for instances running on the wrong node
855 - a few fixes to the cluster-merge tool, but it still cannot merge
856 multi-node groups (currently it is not recommended to use this tool)
862 - Improved network configuration for the KVM hypervisor
863 - Added e1000 as a supported NIC for Xen-HVM
864 - Improved the lvmstrap tool to also be able to use partitions, as
865 opposed to full disks
866 - Improved speed of disk wiping (the cluster parameter
867 ``prealloc_wipe_disks``, so that it has a low impact on the total time
868 of instance creations
869 - Added documentation for the OS parameters
870 - Changed ``gnt-instance deactivate-disks`` so that it can work if the
871 hypervisor is not responding
872 - Added display of blacklisted and hidden OS information in
874 - Extended ``gnt-cluster verify`` to also validate hypervisor, backend,
875 NIC and node parameters, which might create problems with currently
876 invalid (but undetected) configuration files, but prevents validation
877 failures when unrelated parameters are modified
878 - Changed cluster initialisation to wait for the master daemon to become
880 - Expanded the RAPI interface:
882 - Added config redistribution resource
883 - Added activation/deactivation of instance disks
884 - Added export of console information
886 - Implemented log file reopening on SIGHUP, which allows using
887 logrotate(8) for the Ganeti log files
888 - Added a basic OOB helper script as an example
894 *(Released Fri, 14 Jan 2011)*
899 - Fixed timezone issues when formatting timestamps
900 - Added support for node groups, available via ``gnt-group`` and other
902 - Added out-of-band framework and management, see :doc:`design
903 document <design-oob>`
904 - Removed support for roman numbers from ``gnt-node list`` and
905 ``gnt-instance list``.
906 - Allowed modification of master network interface via ``gnt-cluster
907 modify --master-netdev``
908 - Accept offline secondaries while shutting down instance disks
909 - Added ``blockdev_prefix`` parameter to Xen PVM and HVM hypervisors
910 - Added support for multiple LVM volume groups
911 - Avoid sorting nodes for ``gnt-node list`` if specific nodes are
913 - Added commands to list available fields:
915 - ``gnt-node list-fields``
916 - ``gnt-group list-fields``
917 - ``gnt-instance list-fields``
919 - Updated documentation and man pages
924 - Moved ``rapi_users`` file into separate directory, now named
925 ``.../ganeti/rapi/users``, ``cfgupgrade`` moves the file and creates a
927 - Added new tool for running commands on many machines,
928 ``tools/ganeti-listrunner``
929 - Implemented more verbose result in ``OpInstanceConsole`` opcode, also
930 improving the ``gnt-instance console`` output
931 - Allowed customisation of disk index separator at ``configure`` time
932 - Export node group allocation policy to :doc:`iallocator <iallocator>`
933 - Added support for non-partitioned md disks in ``lvmstrap``
934 - Added script to gracefully power off KVM instances
935 - Split ``utils`` module into smaller parts
936 - Changed query operations to return more detailed information, e.g.
937 whether an information is unavailable due to an offline node. To use
938 this new functionality, the LUXI call ``Query`` must be used. Field
939 information is now stored by the master daemon and can be retrieved
940 using ``QueryFields``. Instances, nodes and groups can also be queried
941 using the new opcodes ``OpQuery`` and ``OpQueryFields`` (not yet
942 exposed via RAPI). The following commands make use of this
943 infrastructure change:
946 - ``gnt-group list-fields``
948 - ``gnt-node list-fields``
949 - ``gnt-instance list``
950 - ``gnt-instance list-fields``
951 - ``gnt-debug locks``
956 - New RAPI resources (see :doc:`rapi`):
960 - ``/2/groups/[group_name]``
961 - ``/2/groups/[group_name]/assign-nodes``
962 - ``/2/groups/[group_name]/modify``
963 - ``/2/groups/[group_name]/rename``
964 - ``/2/instances/[instance_name]/disk/[disk_index]/grow``
968 - Implemented ``no_install`` for instance creation
969 - Implemented OS parameters for instance reinstallation, allowing
970 use of special settings on reinstallation (e.g. for preserving data)
975 - Added IPv6 support in import/export
976 - Pause DRBD synchronization while wiping disks on instance creation
977 - Updated unittests and QA scripts
978 - Improved network parameters passed to KVM
979 - Converted man pages from docbook to reStructuredText
985 *(Released Mon, 20 Dec 2010)*
987 Released version 2.3.1~rc1 without any changes.
993 *(Released Wed, 1 Dec 2010)*
995 - impexpd: Disable OpenSSL compression in socat if possible (backport
996 from master, commit e90739d625b, see :doc:`installation guide
997 <install-quick>` for details)
998 - Changed unittest coverage report to exclude test scripts
999 - Added script to check version format
1005 *(Released Wed, 1 Dec 2010)*
1007 Released version 2.3.0~rc1 without any changes.
1013 *(Released Fri, 19 Nov 2010)*
1015 A number of bugfixes and documentation updates:
1017 - Update ganeti-os-interface documentation
1018 - Fixed a bug related to duplicate MACs or similar items which should be
1020 - Fix breakage in OS state modify
1021 - Reinstall instance: disallow offline secondaries (fixes bug related to
1022 OS changing but reinstall failing)
1023 - plus all the other fixes between 2.2.1 and 2.2.2
1029 *(Released Tue, 2 Nov 2010)*
1031 - Fixed clearing of the default iallocator using ``gnt-cluster modify``
1032 - Fixed master failover race with watcher
1033 - Fixed a bug in ``gnt-node modify`` which could lead to an inconsistent
1035 - Accept previously stopped instance for export with instance removal
1036 - Simplify and extend the environment variables for instance OS scripts
1037 - Added new node flags, ``master_capable`` and ``vm_capable``
1038 - Added optional instance disk wiping prior during allocation. This is a
1039 cluster-wide option and can be set/modified using
1040 ``gnt-cluster {init,modify} --prealloc-wipe-disks``.
1041 - Added IPv6 support, see :doc:`design document <design-2.3>` and
1042 :doc:`install-quick`
1043 - Added a new watcher option (``--ignore-pause``)
1044 - Added option to ignore offline node on instance start/stop
1045 (``--ignore-offline``)
1046 - Allow overriding OS parameters with ``gnt-instance reinstall``
1047 - Added ability to change node's secondary IP address using ``gnt-node
1049 - Implemented privilege separation for all daemons except
1050 ``ganeti-noded``, see ``configure`` options
1051 - Complain if an instance's disk is marked faulty in ``gnt-cluster
1053 - Implemented job priorities (see ``ganeti(7)`` manpage)
1054 - Ignore failures while shutting down instances during failover from
1056 - Exit daemon's bootstrap process only once daemon is ready
1057 - Export more information via ``LUInstanceQuery``/remote API
1058 - Improved documentation, QA and unittests
1059 - RAPI daemon now watches ``rapi_users`` all the time and doesn't need a
1060 restart if the file was created or changed
1061 - Added LUXI protocol version sent with each request and response,
1062 allowing detection of server/client mismatches
1063 - Moved the Python scripts among gnt-* and ganeti-* into modules
1064 - Moved all code related to setting up SSH to an external script,
1066 - Infrastructure changes for node group support in future versions
1072 *(Released Fri, 19 Nov 2010)*
1074 A few small bugs fixed, and some improvements to the build system:
1076 - Fix documentation regarding conversion to drbd
1077 - Fix validation of parameters in cluster modify (``gnt-cluster modify
1079 - Fix error handling in node modify with multiple changes
1080 - Allow remote imports without checked names
1086 *(Released Tue, 19 Oct 2010)*
1088 - Disable SSL session ID cache in RPC client
1094 *(Released Thu, 14 Oct 2010)*
1096 - Fix interaction between Curl/GnuTLS and the Python's HTTP server
1097 (thanks Apollon Oikonomopoulos!), finally allowing the use of Curl
1099 - Fix problems with interaction between Curl and Python's HTTP server,
1100 resulting in increased speed in many RPC calls
1101 - Improve our release script to prevent breakage with older aclocal and
1108 *(Released Thu, 7 Oct 2010)*
1110 - Fixed issue 125, replace hardcoded "xenvg" in ``gnt-cluster`` with
1111 value retrieved from master
1112 - Added support for blacklisted or hidden OS definitions
1113 - Added simple lock monitor (accessible via (``gnt-debug locks``)
1114 - Added support for -mem-path in KVM hypervisor abstraction layer
1115 - Allow overriding instance parameters in tool for inter-cluster
1116 instance moves (``tools/move-instance``)
1117 - Improved opcode summaries (e.g. in ``gnt-job list``)
1118 - Improve consistency of OS listing by sorting it
1119 - Documentation updates
1125 *(Released Fri, 8 Oct 2010)*
1127 - Rebuild with a newer autotools version, to fix python 2.6 compatibility
1133 *(Released Mon, 4 Oct 2010)*
1135 - Fixed regression in ``gnt-instance rename``
1141 *(Released Wed, 22 Sep 2010)*
1143 - Fixed OS_VARIANT variable for OS scripts
1144 - Fixed cluster tag operations via RAPI
1145 - Made ``setup-ssh`` exit with non-zero code if an error occurred
1146 - Disabled RAPI CA checks in watcher
1152 *(Released Mon, 23 Aug 2010)*
1154 - Support DRBD versions of the format "a.b.c.d"
1156 - Re-introduce support for usage from multiple threads in RAPI client
1157 - Instance renames and modify via RAPI
1158 - Work around race condition between processing and archival in job
1160 - Mark opcodes following failed one as failed, too
1161 - Job field ``lock_status`` was removed due to difficulties making it
1162 work with the changed job queue in Ganeti 2.2; a better way to monitor
1163 locks is expected for a later 2.2.x release
1164 - Fixed dry-run behaviour with many commands
1165 - Support ``ssh-agent`` again when adding nodes
1166 - Many additional bugfixes
1172 *(Released Fri, 30 Jul 2010)*
1174 Important change: the internal RPC mechanism between Ganeti nodes has
1175 changed from using a home-grown http library (based on the Python base
1176 libraries) to use the PycURL library. This requires that PycURL is
1177 installed on nodes. Please note that on Debian/Ubuntu, PycURL is linked
1178 against GnuTLS by default. cURL's support for GnuTLS had known issues
1179 before cURL 7.21.0 and we recommend using the latest cURL release or
1180 linking against OpenSSL. Most other distributions already link PycURL
1181 and cURL against OpenSSL. The command::
1183 python -c 'import pycurl; print pycurl.version'
1185 can be used to determine the libraries PycURL and cURL are linked
1188 Other significant changes:
1190 - Rewrote much of the internals of the job queue, in order to achieve
1191 better parallelism; this decouples job query operations from the job
1192 processing, and it should allow much nicer behaviour of the master
1193 daemon under load, and it also has uncovered some long-standing bugs
1194 related to the job serialisation (now fixed)
1195 - Added a default iallocator setting to the cluster parameters,
1196 eliminating the need to always pass nodes or an iallocator for
1197 operations that require selection of new node(s)
1198 - Added experimental support for the LXC virtualization method
1199 - Added support for OS parameters, which allows the installation of
1200 instances to pass parameter to OS scripts in order to customise the
1202 - Added a hypervisor parameter controlling the migration type (live or
1203 non-live), since hypervisors have various levels of reliability; this
1204 has renamed the 'live' parameter to 'mode'
1205 - Added a cluster parameter ``reserved_lvs`` that denotes reserved
1206 logical volumes, meaning that cluster verify will ignore them and not
1207 flag their presence as errors
1208 - The watcher will now reset the error count for failed instances after
1209 8 hours, thus allowing self-healing if the problem that caused the
1210 instances to be down/fail to start has cleared in the meantime
1211 - Added a cluster parameter ``drbd_usermode_helper`` that makes Ganeti
1212 check for, and warn, if the drbd module parameter ``usermode_helper``
1213 is not consistent with the cluster-wide setting; this is needed to
1214 make diagnose easier of failed drbd creations
1215 - Started adding base IPv6 support, but this is not yet
1216 enabled/available for use
1217 - Rename operations (cluster, instance) will now return the new name,
1218 which is especially useful if a short name was passed in
1219 - Added support for instance migration in RAPI
1220 - Added a tool to pre-configure nodes for the SSH setup, before joining
1221 them to the cluster; this will allow in the future a simplified model
1222 for node joining (but not yet fully enabled in 2.2); this needs the
1223 paramiko python library
1224 - Fixed handling of name-resolving errors
1225 - Fixed consistency of job results on the error path
1226 - Fixed master-failover race condition when executed multiple times in
1228 - Fixed many bugs related to the job queue (mostly introduced during the
1229 2.2 development cycle, so not all are impacting 2.1)
1230 - Fixed instance migration with missing disk symlinks
1231 - Fixed handling of unknown jobs in ``gnt-job archive``
1232 - And many other small fixes/improvements
1236 - Enhanced both the unittest and the QA coverage
1237 - Switched the opcode validation to a generic model, and extended the
1238 validation to all opcode parameters
1239 - Changed more parts of the code that write shell scripts to use the
1241 - Switched the master daemon to use the asyncore library for the Luxi
1248 *(Released Thu, 17 Jun 2010)*
1250 - Added tool (``move-instance``) and infrastructure to move instances
1251 between separate clusters (see :doc:`separate documentation
1252 <move-instance>` and :doc:`design document <design-2.2>`)
1253 - Added per-request RPC timeout
1254 - RAPI now requires a Content-Type header for requests with a body (e.g.
1255 ``PUT`` or ``POST``) which must be set to ``application/json`` (see
1256 :rfc:`2616` (HTTP/1.1), section 7.2.1)
1257 - ``ganeti-watcher`` attempts to restart ``ganeti-rapi`` if RAPI is not
1259 - Implemented initial support for running Ganeti daemons as separate
1260 users, see configure-time flags ``--with-user-prefix`` and
1261 ``--with-group-prefix`` (only ``ganeti-rapi`` is supported at this
1263 - Instances can be removed after export (``gnt-backup export
1264 --remove-instance``)
1265 - Self-signed certificates generated by Ganeti now use a 2048 bit RSA
1266 key (instead of 1024 bit)
1267 - Added new cluster configuration file for cluster domain secret
1268 - Import/export now use SSL instead of SSH
1269 - Added support for showing estimated time when exporting an instance,
1270 see the ``ganeti-os-interface(7)`` manpage and look for
1277 *(Released Tue, 16 Nov 2010)*
1279 Some more bugfixes. Unless critical bugs occur, this will be the last
1282 - Fix case of MAC special-values
1283 - Fix mac checker regex
1284 - backend: Fix typo causing "out of range" error
1285 - Add missing --units in gnt-instance list man page
1291 *(Released Tue, 24 Aug 2010)*
1294 - Don't ignore secondary node silently on non-mirrored disk templates
1296 - Fix --master-netdev arg name in gnt-cluster(8) (issue 114)
1297 - Fix usb_mouse parameter breaking with vnc_console (issue 109)
1298 - Properly document the usb_mouse parameter
1299 - Fix path in ganeti-rapi(8) (issue 116)
1300 - Adjust error message when the ganeti user's .ssh directory is
1302 - Add same-node-check when changing the disk template to drbd
1308 *(Released Fri, 16 Jul 2010)*
1311 - Add an option to only select some reboot types during qa/burnin.
1312 (on some hypervisors consequent reboots are not supported)
1313 - Fix infrequent race condition in master failover. Sometimes the old
1314 master ip address would be still detected as up for a short time
1315 after it was removed, causing failover to fail.
1316 - Decrease mlockall warnings when the ctypes module is missing. On
1317 Python 2.4 we support running even if no ctypes module is installed,
1318 but we were too verbose about this issue.
1319 - Fix building on old distributions, on which man doesn't have a
1321 - Fix RAPI not to ignore the MAC address on instance creation
1322 - Implement the old instance creation format in the RAPI client.
1328 *(Released Thu, 01 Jul 2010)*
1330 A small bugfix release:
1331 - Fix disk adoption: broken by strict --disk option checking in 2.1.4
1332 - Fix batch-create: broken in the whole 2.1 series due to a lookup on
1333 a non-existing option
1334 - Fix instance create: the --force-variant option was ignored
1335 - Improve pylint 0.21 compatibility and warnings with Python 2.6
1336 - Fix modify node storage with non-FQDN arguments
1337 - Fix RAPI client to authenticate under Python 2.6 when used
1338 for more than 5 requests needing authentication
1339 - Fix gnt-instance modify -t (storage) giving a wrong error message
1340 when converting a non-shutdown drbd instance to plain
1346 *(Released Fri, 18 Jun 2010)*
1348 A small bugfix release:
1350 - Fix live migration of KVM instances started with older Ganeti
1351 versions which had fewer hypervisor parameters
1352 - Fix gnt-instance grow-disk on down instances
1353 - Fix an error-reporting bug during instance migration
1354 - Better checking of the ``--net`` and ``--disk`` values, to avoid
1355 silently ignoring broken ones
1356 - Fix an RPC error reporting bug affecting, for example, RAPI client
1358 - Fix bug triggered by different API version os-es on different nodes
1359 - Fix a bug in instance startup with custom hvparams: OS level
1360 parameters would fail to be applied.
1361 - Fix the RAPI client under Python 2.6 (but more work is needed to
1362 make it work completely well with OpenSSL)
1363 - Fix handling of errors when resolving names from DNS
1369 *(Released Thu, 3 Jun 2010)*
1371 A medium sized development cycle. Some new features, and some
1372 fixes/small improvements/cleanups.
1374 Significant features
1375 ~~~~~~~~~~~~~~~~~~~~
1377 The node deamon now tries to mlock itself into memory, unless the
1378 ``--no-mlock`` flag is passed. It also doesn't fail if it can't write
1379 its logs, and falls back to console logging. This allows emergency
1380 features such as ``gnt-node powercycle`` to work even in the event of a
1381 broken node disk (tested offlining the disk hosting the node's
1382 filesystem and dropping its memory caches; don't try this at home)
1384 KVM: add vhost-net acceleration support. It can be tested with a new
1385 enough version of the kernel and of qemu-kvm.
1387 KVM: Add instance chrooting feature. If you use privilege dropping for
1388 your VMs you can also now force them to chroot to an empty directory,
1389 before starting the emulated guest.
1391 KVM: Add maximum migration bandwith and maximum downtime tweaking
1392 support (requires a new-enough version of qemu-kvm).
1394 Cluster verify will now warn if the master node doesn't have the master
1395 ip configured on it.
1397 Add a new (incompatible) instance creation request format to RAPI which
1398 supports all parameters (previously only a subset was supported, and it
1399 wasn't possible to extend the old format to accomodate all the new
1400 features. The old format is still supported, and a client can check for
1401 this feature, before using it, by checking for its presence in the
1402 ``features`` RAPI resource.
1404 Now with ancient latin support. Try it passing the ``--roman`` option to
1405 ``gnt-instance info``, ``gnt-cluster info`` or ``gnt-node list``
1406 (requires the python-roman module to be installed, in order to work).
1411 As usual many internal code refactorings, documentation updates, and
1414 - Lots of improvements and cleanups to the experimental Remote API
1415 (RAPI) client library.
1416 - A new unit test suite for the core daemon libraries.
1417 - A fix to creating missing directories makes sure the umask is not
1418 applied anymore. This enforces the same directory permissions
1420 - Better handling terminating daemons with ctrl+c (used when running
1421 them in debugging mode).
1422 - Fix a race condition in live migrating a KVM instance, when stat()
1423 on the old proc status file returned EINVAL, which is an unexpected
1425 - Fixed manpage checking with newer man and utf-8 charachters. But now
1426 you need the en_US.UTF-8 locale enabled to build Ganeti from git.
1432 *(Released Fri, 7 May 2010)*
1434 Fix a bug which prevented untagged KVM instances from starting.
1440 *(Released Fri, 7 May 2010)*
1442 Another release with a long development cycle, during which many
1443 different features were added.
1445 Significant features
1446 ~~~~~~~~~~~~~~~~~~~~
1448 The KVM hypervisor now can run the individual instances as non-root, to
1449 reduce the impact of a VM being hijacked due to bugs in the
1450 hypervisor. It is possible to run all instances as a single (non-root)
1451 user, to manually specify a user for each instance, or to dynamically
1452 allocate a user out of a cluster-wide pool to each instance, with the
1453 guarantee that no two instances will run under the same user ID on any
1456 An experimental RAPI client library, that can be used standalone
1457 (without the other Ganeti libraries), is provided in the source tree as
1458 ``lib/rapi/client.py``. Note this client might change its interface in
1459 the future, as we iterate on its capabilities.
1461 A new command, ``gnt-cluster renew-crypto`` has been added to easily
1462 replace the cluster's certificates and crypto keys. This might help in
1463 case they have been compromised, or have simply expired.
1465 A new disk option for instance creation has been added that allows one
1466 to "adopt" currently existing logical volumes, with data
1467 preservation. This should allow easier migration to Ganeti from
1468 unmanaged (or managed via other software) instances.
1470 Another disk improvement is the possibility to convert between redundant
1471 (DRBD) and plain (LVM) disk configuration for an instance. This should
1472 allow better scalability (starting with one node and growing the
1473 cluster, or shrinking a two-node cluster to one node).
1475 A new feature that could help with automated node failovers has been
1476 implemented: if a node sees itself as offline (by querying the master
1477 candidates), it will try to shutdown (hard) all instances and any active
1478 DRBD devices. This reduces the risk of duplicate instances if an
1479 external script automatically failovers the instances on such nodes. To
1480 enable this, the cluster parameter ``maintain_node_health`` should be
1481 enabled; in the future this option (per the name) will enable other
1482 automatic maintenance features.
1484 Instance export/import now will reuse the original instance
1485 specifications for all parameters; that means exporting an instance,
1486 deleting it and the importing it back should give an almost identical
1487 instance. Note that the default import behaviour has changed from
1488 before, where it created only one NIC; now it recreates the original
1491 Cluster verify has added a few new checks: SSL certificates validity,
1492 /etc/hosts consistency across the cluster, etc.
1497 As usual, many internal changes were done, documentation fixes,
1500 - Fixed cluster initialization with disabled cluster storage (regression
1501 introduced in 2.1.1)
1502 - File-based storage supports growing the disks
1503 - Fixed behaviour of node role changes
1504 - Fixed cluster verify for some corner cases, plus a general rewrite of
1505 cluster verify to allow future extension with more checks
1506 - Fixed log spamming by watcher and node daemon (regression introduced
1508 - Fixed possible validation issues when changing the list of enabled
1510 - Fixed cleanup of /etc/hosts during node removal
1511 - Fixed RAPI response for invalid methods
1512 - Fixed bug with hashed passwords in ``ganeti-rapi`` daemon
1513 - Multiple small improvements to the KVM hypervisor (VNC usage, booting
1514 from ide disks, etc.)
1515 - Allow OS changes without re-installation (to record a changed OS
1516 outside of Ganeti, or to allow OS renames)
1517 - Allow instance creation without OS installation (useful for example if
1518 the OS will be installed manually, or restored from a backup not in
1520 - Implemented option to make cluster ``copyfile`` use the replication
1522 - Added list of enabled hypervisors to ssconf (possibly useful for
1524 - Added a new tool (``tools/cfgupgrade12``) that allows upgrading from
1526 - A partial form of node re-IP is possible via node readd, which now
1527 allows changed node primary IP
1528 - Command line utilities now show an informational message if the job is
1530 - The logs of the master daemon now show the PID/UID/GID of the
1537 *(Released Fri, 12 Mar 2010)*
1539 During the 2.1.0 long release candidate cycle, a lot of improvements and
1540 changes have accumulated with were released later as 2.1.1.
1545 The node evacuate command (``gnt-node evacuate``) was significantly
1546 rewritten, and as such the IAllocator protocol was changed - a new
1547 request type has been added. This unfortunate change during a stable
1548 series is designed to improve performance of node evacuations; on
1549 clusters with more than about five nodes and which are well-balanced,
1550 evacuation should proceed in parallel for all instances of the node
1551 being evacuated. As such, any existing IAllocator scripts need to be
1552 updated, otherwise the above command will fail due to the unknown
1553 request. The provided "dumb" allocator has not been updated; but the
1554 ganeti-htools package supports the new protocol since version 0.2.4.
1556 Another important change is increased validation of node and instance
1557 names. This might create problems in special cases, if invalid host
1558 names are being used.
1560 Also, a new layer of hypervisor parameters has been added, that sits at
1561 OS level between the cluster defaults and the instance ones. This allows
1562 customisation of virtualization parameters depending on the installed
1563 OS. For example instances with OS 'X' may have a different KVM kernel
1564 (or any other parameter) than the cluster defaults. This is intended to
1565 help managing a multiple OSes on the same cluster, without manual
1566 modification of each instance's parameters.
1568 A tool for merging clusters, ``cluster-merge``, has been added in the
1569 tools sub-directory.
1574 - Improved the int/float conversions that should make the code more
1575 robust in face of errors from the node daemons
1576 - Fixed the remove node code in case of internal configuration errors
1577 - Fixed the node daemon behaviour in face of inconsistent queue
1578 directory (e.g. read-only file-system where we can't open the files
1580 - Fixed the behaviour of gnt-node modify for master candidate demotion;
1581 now it either aborts cleanly or, if given the new "auto_promote"
1582 parameter, will automatically promote other nodes as needed
1583 - Fixed compatibility with (unreleased yet) Python 2.6.5 that would
1584 completely prevent Ganeti from working
1585 - Fixed bug for instance export when not all disks were successfully
1587 - Fixed behaviour of node add when the new node is slow in starting up
1589 - Fixed handling of signals in the LUXI client, which should improve
1590 behaviour of command-line scripts
1591 - Added checks for invalid node/instance names in the configuration (now
1592 flagged during cluster verify)
1593 - Fixed watcher behaviour for disk activation errors
1594 - Fixed two potentially endless loops in http library, which led to the
1595 RAPI daemon hanging and consuming 100% CPU in some cases
1596 - Fixed bug in RAPI daemon related to hashed passwords
1597 - Fixed bug for unintended qemu-level bridging of multi-NIC KVM
1599 - Enhanced compatibility with non-Debian OSes, but not using absolute
1600 path in some commands and allowing customisation of the ssh
1601 configuration directory
1602 - Fixed possible future issue with new Python versions by abiding to the
1603 proper use of ``__slots__`` attribute on classes
1604 - Added checks that should prevent directory traversal attacks
1605 - Many documentation fixes based on feedback from users
1610 - Added an "early_release" more for instance replace disks and node
1611 evacuate, where we release locks earlier and thus allow higher
1612 parallelism within the cluster
1613 - Added watcher hooks, intended to allow the watcher to restart other
1614 daemons (e.g. from the ganeti-nbma project), but they can be used of
1615 course for any other purpose
1616 - Added a compile-time disable for DRBD barriers, to increase
1617 performance if the administrator trusts the power supply or the
1618 storage system to not lose writes
1619 - Added the option of using syslog for logging instead of, or in
1620 addition to, Ganeti's own log files
1621 - Removed boot restriction for paravirtual NICs for KVM, recent versions
1622 can indeed boot from a paravirtual NIC
1623 - Added a generic debug level for many operations; while this is not
1624 used widely yet, it allows one to pass the debug value all the way to
1626 - Enhanced the hooks environment for instance moves (failovers,
1627 migrations) where the primary/secondary nodes changed during the
1628 operation, by adding {NEW,OLD}_{PRIMARY,SECONDARY} vars
1629 - Enhanced data validations for many user-supplied values; one important
1630 item is the restrictions imposed on instance and node names, which
1631 might reject some (invalid) host names
1632 - Add a configure-time option to disable file-based storage, if it's not
1633 needed; this allows greater security separation between the master
1634 node and the other nodes from the point of view of the inter-node RPC
1636 - Added user notification in interactive tools if job is waiting in the
1637 job queue or trying to acquire locks
1638 - Added log messages when a job is waiting for locks
1639 - Added filtering by node tags in instance operations which admit
1640 multiple instances (start, stop, reboot, reinstall)
1641 - Added a new tool for cluster mergers, ``cluster-merge``
1642 - Parameters from command line which are of the form ``a=b,c=d`` can now
1643 use backslash escapes to pass in values which contain commas,
1644 e.g. ``a=b\\c,d=e`` where the 'a' parameter would get the value
1646 - For KVM, the instance name is the first parameter passed to KVM, so
1647 that it's more visible in the process list
1653 *(Released Tue, 2 Mar 2010)*
1655 Ganeti 2.1 brings many improvements with it. Major changes:
1657 - Added infrastructure to ease automated disk repairs
1658 - Added new daemon to export configuration data in a cheaper way than
1659 using the remote API
1660 - Instance NICs can now be routed instead of being associated with a
1662 - Improved job locking logic to reduce impact of jobs acquiring multiple
1663 locks waiting for other long-running jobs
1665 In-depth implementation details can be found in the Ganeti 2.1 design
1671 - Added chroot hypervisor
1672 - Added more options to xen-hvm hypervisor (``kernel_path`` and
1674 - Added more options to xen-pvm hypervisor (``use_bootloader``,
1675 ``bootloader_path`` and ``bootloader_args``)
1676 - Added the ``use_localtime`` option for the xen-hvm and kvm
1677 hypervisors, and the default value for this has changed to false (in
1678 2.0 xen-hvm always enabled it)
1679 - Added luxi call to submit multiple jobs in one go
1680 - Added cluster initialization option to not modify ``/etc/hosts``
1682 - Added network interface parameters
1683 - Added dry run mode to some LUs
1684 - Added RAPI resources:
1686 - ``/2/instances/[instance_name]/info``
1687 - ``/2/instances/[instance_name]/replace-disks``
1688 - ``/2/nodes/[node_name]/evacuate``
1689 - ``/2/nodes/[node_name]/migrate``
1690 - ``/2/nodes/[node_name]/role``
1691 - ``/2/nodes/[node_name]/storage``
1692 - ``/2/nodes/[node_name]/storage/modify``
1693 - ``/2/nodes/[node_name]/storage/repair``
1695 - Added OpCodes to evacuate or migrate all instances on a node
1696 - Added new command to list storage elements on nodes (``gnt-node
1697 list-storage``) and modify them (``gnt-node modify-storage``)
1698 - Added new ssconf files with master candidate IP address
1699 (``ssconf_master_candidates_ips``), node primary IP address
1700 (``ssconf_node_primary_ips``) and node secondary IP address
1701 (``ssconf_node_secondary_ips``)
1702 - Added ``ganeti-confd`` and a client library to query the Ganeti
1703 configuration via UDP
1704 - Added ability to run hooks after cluster initialization and before
1706 - Added automatic mode for disk replace (``gnt-instance replace-disks
1708 - Added ``gnt-instance recreate-disks`` to re-create (empty) disks
1709 after catastrophic data-loss
1710 - Added ``gnt-node repair-storage`` command to repair damaged LVM volume
1712 - Added ``gnt-instance move`` command to move instances
1713 - Added ``gnt-cluster watcher`` command to control watcher
1714 - Added ``gnt-node powercycle`` command to powercycle nodes
1715 - Added new job status field ``lock_status``
1716 - Added parseable error codes to cluster verification (``gnt-cluster
1717 verify --error-codes``) and made output less verbose (use
1718 ``--verbose`` to restore previous behaviour)
1719 - Added UUIDs to the main config entities (cluster, nodes, instances)
1720 - Added support for OS variants
1721 - Added support for hashed passwords in the Ganeti remote API users file
1723 - Added option to specify maximum timeout on instance shutdown
1724 - Added ``--no-ssh-init`` option to ``gnt-cluster init``
1725 - Added new helper script to start and stop Ganeti daemons
1726 (``daemon-util``), with the intent to reduce the work necessary to
1727 adjust Ganeti for non-Debian distributions and to start/stop daemons
1729 - Added more unittests
1730 - Fixed critical bug in ganeti-masterd startup
1731 - Removed the configure-time ``kvm-migration-port`` parameter, this is
1732 now customisable at the cluster level for both the KVM and Xen
1733 hypervisors using the new ``migration_port`` parameter
1734 - Pass ``INSTANCE_REINSTALL`` variable to OS installation script when
1735 reinstalling an instance
1736 - Allowed ``@`` in tag names
1737 - Migrated to Sphinx (http://sphinx.pocoo.org/) for documentation
1738 - Many documentation updates
1739 - Distribute hypervisor files on ``gnt-cluster redist-conf``
1740 - ``gnt-instance reinstall`` can now reinstall multiple instances
1741 - Updated many command line parameters
1742 - Introduced new OS API version 15
1743 - No longer support a default hypervisor
1744 - Treat virtual LVs as inexistent
1745 - Improved job locking logic to reduce lock contention
1746 - Match instance and node names case insensitively
1747 - Reimplemented bash completion script to be more complete
1754 *(Released Thu, 4 Feb 2010)*
1756 - Fix cleaner behaviour on nodes not in a cluster (Debian bug 568105)
1757 - Fix a string formatting bug
1758 - Improve safety of the code in some error paths
1759 - Improve data validation in the master of values returned from nodes
1765 *(Released Thu, 17 Dec 2009)*
1767 - Fix security issue due to missing validation of iallocator names; this
1768 allows local and remote execution of arbitrary executables
1769 - Fix failure of gnt-node list during instance removal
1770 - Ship the RAPI documentation in the archive
1776 *(Released Wed, 30 Sep 2009)*
1778 - Fixed many wrong messages
1779 - Fixed a few bugs related to the locking library
1780 - Fixed MAC checking at instance creation time
1781 - Fixed a DRBD parsing bug related to gaps in /proc/drbd
1782 - Fixed a few issues related to signal handling in both daemons and
1784 - Fixed the example startup script provided
1785 - Fixed insserv dependencies in the example startup script (patch from
1787 - Fixed handling of drained nodes in the iallocator framework
1788 - Fixed handling of KERNEL_PATH parameter for xen-hvm (Debian bug
1790 - Fixed error related to invalid job IDs in job polling
1791 - Fixed job/opcode persistence on unclean master shutdown
1792 - Fixed handling of partial job processing after unclean master
1794 - Fixed error reporting from LUs, previously all errors were converted
1795 into execution errors
1796 - Fixed error reporting from burnin
1797 - Decreased significantly the memory usage of the job queue
1798 - Optimised slightly multi-job submission
1799 - Optimised slightly opcode loading
1800 - Backported the multi-job submit framework from the development
1801 branch; multi-instance start and stop should be faster
1802 - Added script to clean archived jobs after 21 days; this will reduce
1803 the size of the queue directory
1804 - Added some extra checks in disk size tracking
1805 - Added an example ethers hook script
1806 - Added a cluster parameter that prevents Ganeti from modifying of
1808 - Added more node information to RAPI responses
1809 - Added a ``gnt-job watch`` command that allows following the ouput of a
1811 - Added a bind-address option to ganeti-rapi
1812 - Added more checks to the configuration verify
1813 - Enhanced the burnin script such that some operations can be retried
1815 - Converted instance reinstall to multi-instance model
1821 *(Released Fri, 7 Aug 2009)*
1823 - Added ``--ignore-size`` to the ``gnt-instance activate-disks`` command
1824 to allow using the pre-2.0.2 behaviour in activation, if any existing
1825 instances have mismatched disk sizes in the configuration
1826 - Added ``gnt-cluster repair-disk-sizes`` command to check and update
1827 any configuration mismatches for disk sizes
1828 - Added ``gnt-master cluste-failover --no-voting`` to allow master
1829 failover to work on two-node clusters
1830 - Fixed the ``--net`` option of ``gnt-backup import``, which was
1832 - Fixed detection of OS script errors in ``gnt-backup export``
1833 - Fixed exit code of ``gnt-backup export``
1839 *(Released Fri, 17 Jul 2009)*
1841 - Added experimental support for stripped logical volumes; this should
1842 enhance performance but comes with a higher complexity in the block
1843 device handling; stripping is only enabled when passing
1844 ``--with-lvm-stripecount=N`` to ``configure``, but codepaths are
1845 affected even in the non-stripped mode
1846 - Improved resiliency against transient failures at the end of DRBD
1847 resyncs, and in general of DRBD resync checks
1848 - Fixed a couple of issues with exports and snapshot errors
1849 - Fixed a couple of issues in instance listing
1850 - Added display of the disk size in ``gnt-instance info``
1851 - Fixed checking for valid OSes in instance creation
1852 - Fixed handling of the "vcpus" parameter in instance listing and in
1853 general of invalid parameters
1854 - Fixed http server library, and thus RAPI, to handle invalid
1855 username/password combinations correctly; this means that now they
1856 report unauthorized for queries too, not only for modifications,
1857 allowing earlier detect of configuration problems
1858 - Added a new "role" node list field, equivalent to the master/master
1859 candidate/drained/offline flags combinations
1860 - Fixed cluster modify and changes of candidate pool size
1861 - Fixed cluster verify error messages for wrong files on regular nodes
1862 - Fixed a couple of issues with node demotion from master candidate role
1863 - Fixed node readd issues
1864 - Added non-interactive mode for ``ganeti-masterd --no-voting`` startup
1865 - Added a new ``--no-voting`` option for masterfailover to fix failover
1866 on two-nodes clusters when the former master node is unreachable
1867 - Added instance reinstall over RAPI
1873 *(Released Tue, 16 Jun 2009)*
1875 - added ``-H``/``-B`` startup parameters to ``gnt-instance``, which will
1876 allow re-adding the start in single-user option (regression from 1.2)
1877 - the watcher writes the instance status to a file, to allow monitoring
1878 to report the instance status (from the master) based on cached
1879 results of the watcher's queries; while this can get stale if the
1880 watcher is being locked due to other work on the cluster, this is
1881 still an improvement
1882 - the watcher now also restarts the node daemon and the rapi daemon if
1884 - fixed the watcher to handle full and drained queue cases
1885 - hooks export more instance data in the environment, which helps if
1886 hook scripts need to take action based on the instance's properties
1887 (no longer need to query back into ganeti)
1888 - instance failovers when the instance is stopped do not check for free
1889 RAM, so that failing over a stopped instance is possible in low memory
1891 - rapi uses queries for tags instead of jobs (for less job traffic), and
1892 for cluster tags it won't talk to masterd at all but read them from
1894 - a couple of error handling fixes in RAPI
1895 - drbd handling: improved the error handling of inconsistent disks after
1896 resync to reduce the frequency of "there are some degraded disks for
1897 this instance" messages
1898 - fixed a bug in live migration when DRBD doesn't want to reconnect (the
1899 error handling path called a wrong function name)
1905 *(Released Wed, 27 May 2009)*
1907 - no changes from rc5
1913 *(Released Wed, 20 May 2009)*
1915 - fix a couple of bugs (validation, argument checks)
1916 - fix ``gnt-cluster getmaster`` on non-master nodes (regression)
1917 - some small improvements to RAPI and IAllocator
1918 - make watcher automatically start the master daemon if down
1924 *(Released Mon, 27 Apr 2009)*
1926 - change the OS list to not require locks; this helps with big clusters
1927 - fix ``gnt-cluster verify`` and ``gnt-cluster verify-disks`` when the
1928 volume group is broken
1929 - ``gnt-instance info``, without any arguments, doesn't run for all
1930 instances anymore; either pass ``--all`` or pass the desired
1931 instances; this helps against mistakes on big clusters where listing
1932 the information for all instances takes a long time
1933 - miscellaneous doc and man pages fixes
1939 *(Released Wed, 8 Apr 2009)*
1941 - Change the internal locking model of some ``gnt-node`` commands, in
1942 order to reduce contention (and blocking of master daemon) when
1943 batching many creation/reinstall jobs
1944 - Fixes to Xen soft reboot
1945 - No longer build documentation at build time, instead distribute it in
1946 the archive, in order to reduce the need for the whole docbook/rst
1953 *(Released Fri, 27 Mar 2009)*
1955 - Now the cfgupgrade scripts works and can upgrade 1.2.7 clusters to 2.0
1956 - Fix watcher startup sequence, improves the behaviour of busy clusters
1957 - Some other fixes in ``gnt-cluster verify``, ``gnt-instance
1958 replace-disks``, ``gnt-instance add``, ``gnt-cluster queue``, KVM VNC
1959 bind address and other places
1960 - Some documentation fixes and updates
1966 *(Released Mon, 2 Mar 2009)*
1968 - More documentation updates, now all docs should be more-or-less
1970 - A couple of small fixes (mixed hypervisor clusters, offline nodes,
1972 - Added a customizable HV_KERNEL_ARGS hypervisor parameter (for Xen PVM
1974 - Fix an issue related to $libdir/run/ganeti and cluster creation
1980 *(Released Thu, 19 Feb 2009)*
1982 - Xen PVM and KVM have switched the default value for the instance root
1983 disk to the first partition on the first drive, instead of the whole
1984 drive; this means that the OS installation scripts must be changed
1986 - Man pages have been updated
1987 - RAPI has been switched by default to HTTPS, and the exported functions
1988 should all work correctly
1989 - RAPI v1 has been removed
1990 - Many improvements to the KVM hypervisor
1991 - Block device errors are now better reported
1992 - Many other bugfixes and small improvements
1998 *(Released Mon, 26 Jan 2009)*
2000 - Version 2 is a general rewrite of the code and therefore the
2001 differences are too many to list, see the design document for 2.0 in
2002 the ``doc/`` subdirectory for more details
2003 - In this beta version there is not yet a migration path from 1.2 (there
2004 will be one in the final 2.0 release)
2005 - A few significant changes are:
2007 - all commands are executed by a daemon (``ganeti-masterd``) and the
2008 various ``gnt-*`` commands are just front-ends to it
2009 - all the commands are entered into, and executed from a job queue,
2010 see the ``gnt-job(8)`` manpage
2011 - the RAPI daemon supports read-write operations, secured by basic
2012 HTTP authentication on top of HTTPS
2013 - DRBD version 0.7 support has been removed, DRBD 8 is the only
2014 supported version (when migrating from Ganeti 1.2 to 2.0, you need
2015 to migrate to DRBD 8 first while still running Ganeti 1.2)
2016 - DRBD devices are using statically allocated minor numbers, which
2017 will be assigned to existing instances during the migration process
2018 - there is support for both Xen PVM and Xen HVM instances running on
2020 - KVM virtualization is supported too
2021 - file-based storage has been implemented, which means that it is
2022 possible to run the cluster without LVM and DRBD storage, for
2023 example using a shared filesystem exported from shared storage (and
2024 still have live migration)
2030 *(Released Tue, 13 Jan 2009)*
2032 - Change the default reboot type in ``gnt-instance reboot`` to "hard"
2033 - Reuse the old instance mac address by default on instance import, if
2034 the instance name is the same.
2035 - Handle situations in which the node info rpc returns incomplete
2037 - Add checks for tcp/udp ports collisions in ``gnt-cluster verify``
2038 - Improved version of batcher:
2040 - state file support
2041 - instance mac address support
2042 - support for HVM clusters/instances
2044 - Add an option to show the number of cpu sockets and nodes in
2046 - Support OSes that handle more than one version of the OS api (but do
2047 not change the current API in any other way)
2048 - Fix ``gnt-node migrate``
2049 - ``gnt-debug`` man page
2050 - Fixes various more typos and small issues
2051 - Increase disk resync maximum speed to 60MB/s (from 30MB/s)
2057 *(Released Wed, 24 Sep 2008)*
2059 - new ``--hvm-nic-type`` and ``--hvm-disk-type`` flags to control the
2060 type of disk exported to fully virtualized instances.
2061 - provide access to the serial console of HVM instances
2062 - instance auto_balance flag, set by default. If turned off it will
2063 avoid warnings on cluster verify if there is not enough memory to fail
2064 over an instance. in the future it will prevent automatically failing
2065 it over when we will support that.
2066 - batcher tool for instance creation, see ``tools/README.batcher``
2067 - ``gnt-instance reinstall --select-os`` to interactively select a new
2068 operating system when reinstalling an instance.
2069 - when changing the memory amount on instance modify a check has been
2070 added that the instance will be able to start. also warnings are
2071 emitted if the instance will not be able to fail over, if auto_balance
2073 - documentation fixes
2074 - sync fields between ``gnt-instance list/modify/add/import``
2075 - fix a race condition in drbd when the sync speed was set after giving
2076 the device a remote peer.
2082 *(Released Tue, 22 Jul 2008)*
2084 - note: the allowed size and number of tags per object were reduced
2085 - fix a bug in ``gnt-cluster verify`` with inconsistent volume groups
2086 - fixed twisted 8.x compatibility
2087 - fixed ``gnt-instance replace-disks`` with iallocator
2088 - add TCP keepalives on twisted connections to detect restarted nodes
2089 - disk increase support, see ``gnt-instance grow-disk``
2090 - implement bulk node/instance query for RAPI
2091 - add tags in node/instance listing (optional)
2092 - experimental migration (and live migration) support, read the man page
2093 for ``gnt-instance migrate``
2094 - the ``ganeti-watcher`` logs are now timestamped, and the watcher also
2095 has some small improvements in handling its state file
2101 *(Released Fri, 13 Jun 2008)*
2103 - Experimental readonly, REST-based remote API implementation;
2104 automatically started on master node, TCP port 5080, if enabled by
2105 ``--enable-rapi`` parameter to configure script.
2106 - Instance allocator support. Add and import instance accept a
2107 ``--iallocator`` parameter, and call that instance allocator to decide
2108 which node to use for the instance. The iallocator document describes
2109 what's expected from an allocator script.
2110 - ``gnt-cluster verify`` N+1 memory redundancy checks: Unless passed the
2111 ``--no-nplus1-mem`` option ``gnt-cluster verify`` now checks that if a
2112 node is lost there is still enough memory to fail over the instances
2114 - ``gnt-cluster verify`` hooks: it is now possible to add post-hooks to
2115 ``gnt-cluster verify``, to check for site-specific compliance. All the
2116 hooks will run, and their output, if any, will be displayed. Any
2117 failing hook will make the verification return an error value.
2118 - ``gnt-cluster verify`` now checks that its peers are reachable on the
2119 primary and secondary interfaces
2120 - ``gnt-node add`` now supports the ``--readd`` option, to readd a node
2121 that is still declared as part of the cluster and has failed.
2122 - ``gnt-* list`` commands now accept a new ``-o +field`` way of
2123 specifying output fields, that just adds the chosen fields to the
2125 - ``gnt-backup`` now has a new ``remove`` command to delete an existing
2126 export from the filesystem.
2127 - New per-instance parameters hvm_acpi, hvm_pae and hvm_cdrom_image_path
2128 have been added. Using them you can enable/disable acpi and pae
2129 support, and specify a path for a cd image to be exported to the
2130 instance. These parameters as the name suggest only work on HVM
2132 - When upgrading an HVM cluster to Ganeti 1.2.4, the values for ACPI and
2133 PAE support will be set to the previously hardcoded values, but the
2134 (previously hardcoded) path to the CDROM ISO image will be unset and
2135 if required, needs to be set manually with ``gnt-instance modify``
2137 - The address to which an instance's VNC console is bound is now
2138 selectable per-instance, rather than being cluster wide. Of course
2139 this only applies to instances controlled via VNC, so currently just
2140 applies to HVM clusters.
2146 *(Released Mon, 18 Feb 2008)*
2148 - more tweaks to the disk activation code (especially helpful for DRBD)
2149 - change the default ``gnt-instance list`` output format, now there is
2150 one combined status field (see the manpage for the exact values this
2152 - some more fixes for the mac export to hooks change
2153 - make Ganeti not break with DRBD 8.2.x (which changed the version
2154 format in ``/proc/drbd``) (issue 24)
2155 - add an upgrade tool from "remote_raid1" disk template to "drbd" disk
2156 template, allowing migration from DRBD0.7+MD to DRBD8
2162 *(Released Wed, 30 Jan 2008)*
2164 - fix ``gnt-instance modify`` breakage introduced in 1.2.1 with the HVM
2166 - add command aliases infrastructure and a few aliases
2167 - allow listing of VCPUs in the ``gnt-instance list`` and improve the
2168 man pages and the ``--help`` option of ``gnt-node
2169 list``/``gnt-instance list``
2170 - fix ``gnt-backup list`` with down nodes (issue 21)
2171 - change the tools location (move from $pkgdatadir to $pkglibdir/tools)
2172 - fix the dist archive and add a check for including svn/git files in
2174 - some developer-related changes: improve the burnin and the QA suite,
2175 add an upload script for testing during development
2181 *(Released Wed, 16 Jan 2008)*
2183 - experimental HVM support, read the install document, section
2184 "Initializing the cluster"
2185 - allow for the PVM hypervisor per-instance kernel and initrd paths
2186 - add a new command ``gnt-cluster verify-disks`` which uses a new
2187 algorithm to improve the reconnection of the DRBD pairs if the device
2188 on the secondary node has gone away
2189 - make logical volume code auto-activate LVs at disk activation time
2190 - slightly improve the speed of activating disks
2191 - allow specification of the MAC address at instance creation time, and
2192 changing it later via ``gnt-instance modify``
2193 - fix handling of external commands that generate lots of output on
2195 - update documentation with regard to minimum version of DRBD8 supported
2201 *(Released Tue, 4 Dec 2007)*
2203 - Log the ``xm create`` output to the node daemon log on failure (to
2204 help diagnosing the error)
2205 - In debug mode, log all external commands output if failed to the logs
2206 - Change parsing of lvm commands to ignore stderr
2212 *(Released Wed, 28 Nov 2007)*
2214 - Another round of updates to the DRBD 8 code to deal with more failures
2215 in the replace secondary node operation
2216 - Some more logging of failures in disk operations (lvm, drbd)
2217 - A few documentation updates
2224 *(Released Tue, 13 Nov 2007)*
2226 - Change configuration file format from Python's Pickle to JSON.
2227 Upgrading is possible using the cfgupgrade utility.
2228 - Add support for DRBD 8.0 (new disk template ``drbd``) which allows for
2229 faster replace disks and is more stable (DRBD 8 has many improvements
2230 compared to DRBD 0.7)
2231 - Added command line tags support (see man pages for ``gnt-instance``,
2232 ``gnt-node``, ``gnt-cluster``)
2233 - Added instance rename support
2234 - Added multi-instance startup/shutdown
2235 - Added cluster rename support
2236 - Added ``gnt-node evacuate`` to simplify some node operations
2237 - Added instance reboot operation that can speedup reboot as compared to
2239 - Soften the requirement that hostnames are in FQDN format
2240 - The ``ganeti-watcher`` now activates drbd pairs after secondary node
2242 - Removed dependency on debian's patched fping that uses the
2243 non-standard ``-S`` option
2244 - Now the OS definitions are searched for in multiple, configurable
2245 paths (easier for distros to package)
2246 - Some changes to the hooks infrastructure (especially the new
2247 post-configuration update hook)
2248 - Other small bugfixes
2250 .. vim: set textwidth=72 syntax=rst :