7 - RAPI now requires a Content-Type header for requests with a body (e.g.
8 ``PUT`` or ``POST``) which must be set to ``application/json`` (see
9 RFC2616 (HTTP/1.1), section 7.2.1)
15 During the 2.1.0 long release candidate cycle, a lot of improvements and
16 changes have accumulated with were released later as 2.1.1.
21 The node evacuate command (``gnt-node evacuate``) was significantly
22 rewritten, and as such the IAllocator protocol was changed - a new
23 request type has been added. This unfortunate change during a stable
24 series is designed to improve performance of node evacuations; on
25 clusters with more than about five nodes and which are well-balanced,
26 evacuation should proceed in parallel for all instances of the node
27 being evacuated. As such, any existing IAllocator scripts need to be
28 updated, otherwise the above command will fail due to the unknown
29 request. The provided "dumb" allocator has not been updated; but the
30 ganeti-htools package supports the new protocol since version 0.2.4.
32 Another important change is increased validation of node and instance
33 names. This might create problems in special cases, if invalid host
36 Also, a new layer of hypervisor parameters has been added, that sits at
37 OS level between the cluster defaults and the instance ones. This allows
38 customisation of virtualization parameters depending on the installed
39 OS. For example instances with OS 'X' may have a different KVM kernel
40 (or any other parameter) than the cluster defaults. This is intended to
41 help managing a multiple OSes on the same cluster, without manual
42 modification of each instance's parameters.
44 A tool for merging clusters, ``cluster-merge``, has been added in the
50 - Improved the int/float conversions that should make the code more
51 robust in face of errors from the node daemons
52 - Fixed the remove node code in case of internal configuration errors
53 - Fixed the node daemon behaviour in face of inconsistent queue
54 directory (e.g. read-only file-system where we can't open the files
56 - Fixed the behaviour of gnt-node modify for master candidate demotion;
57 now it either aborts cleanly or, if given the new “auto_promote”
58 parameter, will automatically promote other nodes as needed
59 - Fixed compatibility with (unreleased yet) Python 2.6.5 that would
60 completely prevent Ganeti from working
61 - Fixed bug for instance export when not all disks were successfully
63 - Fixed behaviour of node add when the new node is slow in starting up
65 - Fixed handling of signals in the LUXI client, which should improve
66 behaviour of command-line scripts
67 - Added checks for invalid node/instance names in the configuration (now
68 flagged during cluster verify)
69 - Fixed watcher behaviour for disk activation errors
70 - Fixed two potentially endless loops in http library, which led to the
71 RAPI daemon hanging and consuming 100% CPU in some cases
72 - Fixed bug in RAPI daemon related to hashed passwords
73 - Fixed bug for unintended qemu-level bridging of multi-NIC KVM
75 - Enhanced compatibility with non-Debian OSes, but not using absolute
76 path in some commands and allowing customisation of the ssh
77 configuration directory
78 - Fixed possible future issue with new Python versions by abiding to the
79 proper use of ``__slots__`` attribute on classes
80 - Added checks that should prevent directory traversal attacks
81 - Many documentation fixes based on feedback from users
86 - Added an “early_release” more for instance replace disks and node
87 evacuate, where we release locks earlier and thus allow higher
88 parallelism within the cluster
89 - Added watcher hooks, intended to allow the watcher to restart other
90 daemons (e.g. from the ganeti-nbma project), but they can be used of
91 course for any other purpose
92 - Added a compile-time disable for DRBD barriers, to increase
93 performance if the administrator trusts the power supply or the
94 storage system to not lose writes
95 - Added the option of using syslog for logging instead of, or in
96 addition to, Ganeti's own log files
97 - Removed boot restriction for paravirtual NICs for KVM, recent versions
98 can indeed boot from a paravirtual NIC
99 - Added a generic debug level for many operations; while this is not
100 used widely yet, it allows one to pass the debug value all the way to
102 - Enhanced the hooks environment for instance moves (failovers,
103 migrations) where the primary/secondary nodes changed during the
104 operation, by adding {NEW,OLD}_{PRIMARY,SECONDARY} vars
105 - Enhanced data validations for many user-supplied values; one important
106 item is the restrictions imposed on instance and node names, which
107 might reject some (invalid) host names
108 - Add a configure-time option to disable file-based storage, if it's not
109 needed; this allows greater security separation between the master
110 node and the other nodes from the point of view of the inter-node RPC
112 - Added user notification in interactive tools if job is waiting in the
113 job queue or trying to acquire locks
114 - Added log messages when a job is waiting for locks
115 - Added filtering by node tags in instance operations which admit
116 multiple instances (start, stop, reboot, reinstall)
117 - Added a new tool for cluster mergers, ``cluster-merge``
118 - Parameters from command line which are of the form ``a=b,c=d`` can now
119 use backslash escapes to pass in values which contain commas,
120 e.g. ``a=b\\c,d=e`` where the 'a' parameter would get the value
122 - For KVM, the instance name is the first parameter passed to KVM, so
123 that it's more visible in the process list
129 Ganeti 2.1 brings many improvements with it. Major changes:
131 - Added infrastructure to ease automated disk repairs
132 - Added new daemon to export configuration data in a cheaper way than
134 - Instance NICs can now be routed instead of being associated with a
136 - Improved job locking logic to reduce impact of jobs acquiring multiple
137 locks waiting for other long-running jobs
139 In-depth implementation details can be found in the Ganeti 2.1 design
145 - Added chroot hypervisor
146 - Added more options to xen-hvm hypervisor (``kernel_path`` and
148 - Added more options to xen-pvm hypervisor (``use_bootloader``,
149 ``bootloader_path`` and ``bootloader_args``)
150 - Added the ``use_localtime`` option for the xen-hvm and kvm
151 hypervisors, and the default value for this has changed to false (in
152 2.0 xen-hvm always enabled it)
153 - Added luxi call to submit multiple jobs in one go
154 - Added cluster initialization option to not modify ``/etc/hosts``
156 - Added network interface parameters
157 - Added dry run mode to some LUs
158 - Added RAPI resources:
160 - ``/2/instances/[instance_name]/info``
161 - ``/2/instances/[instance_name]/replace-disks``
162 - ``/2/nodes/[node_name]/evacuate``
163 - ``/2/nodes/[node_name]/migrate``
164 - ``/2/nodes/[node_name]/role``
165 - ``/2/nodes/[node_name]/storage``
166 - ``/2/nodes/[node_name]/storage/modify``
167 - ``/2/nodes/[node_name]/storage/repair``
169 - Added OpCodes to evacuate or migrate all instances on a node
170 - Added new command to list storage elements on nodes (``gnt-node
171 list-storage``) and modify them (``gnt-node modify-storage``)
172 - Added new ssconf files with master candidate IP address
173 (``ssconf_master_candidates_ips``), node primary IP address
174 (``ssconf_node_primary_ips``) and node secondary IP address
175 (``ssconf_node_secondary_ips``)
176 - Added ``ganeti-confd`` and a client library to query the Ganeti
177 configuration via UDP
178 - Added ability to run hooks after cluster initialization and before
180 - Added automatic mode for disk replace (``gnt-instance replace-disks
182 - Added ``gnt-instance recreate-disks`` to re-create (empty) disks
183 after catastrophic data-loss
184 - Added ``gnt-node repair-storage`` command to repair damaged LVM volume
186 - Added ``gnt-instance move`` command to move instances
187 - Added ``gnt-cluster watcher`` command to control watcher
188 - Added ``gnt-node powercycle`` command to powercycle nodes
189 - Added new job status field ``lock_status``
190 - Added parseable error codes to cluster verification (``gnt-cluster
191 verify --error-codes``) and made output less verbose (use
192 ``--verbose`` to restore previous behaviour)
193 - Added UUIDs to the main config entities (cluster, nodes, instances)
194 - Added support for OS variants
195 - Added support for hashed passwords in the Ganeti remote API users file
197 - Added option to specify maximum timeout on instance shutdown
198 - Added ``--no-ssh-init`` option to ``gnt-cluster init``
199 - Added new helper script to start and stop Ganeti daemons
200 (``daemon-util``), with the intent to reduce the work necessary to
201 adjust Ganeti for non-Debian distributions and to start/stop daemons
203 - Added more unittests
204 - Fixed critical bug in ganeti-masterd startup
205 - Removed the configure-time ``kvm-migration-port`` parameter, this is
206 now customisable at the cluster level for both the KVM and Xen
207 hypervisors using the new ``migration_port`` parameter
208 - Pass ``INSTANCE_REINSTALL`` variable to OS installation script when
209 reinstalling an instance
210 - Allowed ``@`` in tag names
211 - Migrated to Sphinx (http://sphinx.pocoo.org/) for documentation
212 - Many documentation updates
213 - Distribute hypervisor files on ``gnt-cluster redist-conf``
214 - ``gnt-instance reinstall`` can now reinstall multiple instances
215 - Updated many command line parameters
216 - Introduced new OS API version 15
217 - No longer support a default hypervisor
218 - Treat virtual LVs as inexistent
219 - Improved job locking logic to reduce lock contention
220 - Match instance and node names case insensitively
221 - Reimplemented bash completion script to be more complete
228 - Fix cleaner behaviour on nodes not in a cluster (Debian bug 568105)
229 - Fix a string formatting bug
230 - Improve safety of the code in some error paths
231 - Improve data validation in the master of values returned from nodes
237 - Fix security issue due to missing validation of iallocator names; this
238 allows local and remote execution of arbitrary executables
239 - Fix failure of gnt-node list during instance removal
240 - Ship the RAPI documentation in the archive
246 - Fixed many wrong messages
247 - Fixed a few bugs related to the locking library
248 - Fixed MAC checking at instance creation time
249 - Fixed a DRBD parsing bug related to gaps in /proc/drbd
250 - Fixed a few issues related to signal handling in both daemons and
252 - Fixed the example startup script provided
253 - Fixed insserv dependencies in the example startup script (patch from
255 - Fixed handling of drained nodes in the iallocator framework
256 - Fixed handling of KERNEL_PATH parameter for xen-hvm (Debian bug
258 - Fixed error related to invalid job IDs in job polling
259 - Fixed job/opcode persistence on unclean master shutdown
260 - Fixed handling of partial job processing after unclean master
262 - Fixed error reporting from LUs, previously all errors were converted
263 into execution errors
264 - Fixed error reporting from burnin
265 - Decreased significantly the memory usage of the job queue
266 - Optimised slightly multi-job submission
267 - Optimised slightly opcode loading
268 - Backported the multi-job submit framework from the development
269 branch; multi-instance start and stop should be faster
270 - Added script to clean archived jobs after 21 days; this will reduce
271 the size of the queue directory
272 - Added some extra checks in disk size tracking
273 - Added an example ethers hook script
274 - Added a cluster parameter that prevents Ganeti from modifying of
276 - Added more node information to RAPI responses
277 - Added a ``gnt-job watch`` command that allows following the ouput of a
279 - Added a bind-address option to ganeti-rapi
280 - Added more checks to the configuration verify
281 - Enhanced the burnin script such that some operations can be retried
283 - Converted instance reinstall to multi-instance model
289 - Added ``--ignore-size`` to the ``gnt-instance activate-disks`` command
290 to allow using the pre-2.0.2 behaviour in activation, if any existing
291 instances have mismatched disk sizes in the configuration
292 - Added ``gnt-cluster repair-disk-sizes`` command to check and update
293 any configuration mismatches for disk sizes
294 - Added ``gnt-master cluste-failover --no-voting`` to allow master
295 failover to work on two-node clusters
296 - Fixed the ``--net`` option of ``gnt-backup import``, which was
298 - Fixed detection of OS script errors in ``gnt-backup export``
299 - Fixed exit code of ``gnt-backup export``
305 - Added experimental support for stripped logical volumes; this should
306 enhance performance but comes with a higher complexity in the block
307 device handling; stripping is only enabled when passing
308 ``--with-lvm-stripecount=N`` to ``configure``, but codepaths are
309 affected even in the non-stripped mode
310 - Improved resiliency against transient failures at the end of DRBD
311 resyncs, and in general of DRBD resync checks
312 - Fixed a couple of issues with exports and snapshot errors
313 - Fixed a couple of issues in instance listing
314 - Added display of the disk size in ``gnt-instance info``
315 - Fixed checking for valid OSes in instance creation
316 - Fixed handling of the "vcpus" parameter in instance listing and in
317 general of invalid parameters
318 - Fixed http server library, and thus RAPI, to handle invalid
319 username/password combinations correctly; this means that now they
320 report unauthorized for queries too, not only for modifications,
321 allowing earlier detect of configuration problems
322 - Added a new "role" node list field, equivalent to the master/master
323 candidate/drained/offline flags combinations
324 - Fixed cluster modify and changes of candidate pool size
325 - Fixed cluster verify error messages for wrong files on regular nodes
326 - Fixed a couple of issues with node demotion from master candidate role
327 - Fixed node readd issues
328 - Added non-interactive mode for ``ganeti-masterd --no-voting`` startup
329 - Added a new ``--no-voting`` option for masterfailover to fix failover
330 on two-nodes clusters when the former master node is unreachable
331 - Added instance reinstall over RAPI
337 - added ``-H``/``-B`` startup parameters to ``gnt-instance``, which will
338 allow re-adding the start in single-user option (regression from 1.2)
339 - the watcher writes the instance status to a file, to allow monitoring
340 to report the instance status (from the master) based on cached
341 results of the watcher's queries; while this can get stale if the
342 watcher is being locked due to other work on the cluster, this is
344 - the watcher now also restarts the node daemon and the rapi daemon if
346 - fixed the watcher to handle full and drained queue cases
347 - hooks export more instance data in the environment, which helps if
348 hook scripts need to take action based on the instance's properties
349 (no longer need to query back into ganeti)
350 - instance failovers when the instance is stopped do not check for free
351 RAM, so that failing over a stopped instance is possible in low memory
353 - rapi uses queries for tags instead of jobs (for less job traffic), and
354 for cluster tags it won't talk to masterd at all but read them from
356 - a couple of error handling fixes in RAPI
357 - drbd handling: improved the error handling of inconsistent disks after
358 resync to reduce the frequency of "there are some degraded disks for
359 this instance" messages
360 - fixed a bug in live migration when DRBD doesn't want to reconnect (the
361 error handling path called a wrong function name)
367 - no changes from rc5
370 Version 2.0 release candidate 5
371 -------------------------------
373 - fix a couple of bugs (validation, argument checks)
374 - fix ``gnt-cluster getmaster`` on non-master nodes (regression)
375 - some small improvements to RAPI and IAllocator
376 - make watcher automatically start the master daemon if down
379 Version 2.0 release candidate 4
380 -------------------------------
382 - change the OS list to not require locks; this helps with big clusters
383 - fix ``gnt-cluster verify`` and ``gnt-cluster verify-disks`` when the
384 volume group is broken
385 - ``gnt-instance info``, without any arguments, doesn't run for all
386 instances anymore; either pass ``--all`` or pass the desired
387 instances; this helps against mistakes on big clusters where listing
388 the information for all instances takes a long time
389 - miscellaneous doc and man pages fixes
392 Version 2.0 release candidate 3
393 -------------------------------
395 - Change the internal locking model of some ``gnt-node`` commands, in
396 order to reduce contention (and blocking of master daemon) when
397 batching many creation/reinstall jobs
398 - Fixes to Xen soft reboot
399 - No longer build documentation at build time, instead distribute it in
400 the archive, in order to reduce the need for the whole docbook/rst
404 Version 2.0 release candidate 2
405 -------------------------------
407 - Now the cfgupgrade scripts works and can upgrade 1.2.7 clusters to 2.0
408 - Fix watcher startup sequence, improves the behaviour of busy clusters
409 - Some other fixes in ``gnt-cluster verify``, ``gnt-instance
410 replace-disks``, ``gnt-instance add``, ``gnt-cluster queue``, KVM VNC
411 bind address and other places
412 - Some documentation fixes and updates
415 Version 2.0 release candidate 1
416 -------------------------------
418 - More documentation updates, now all docs should be more-or-less
420 - A couple of small fixes (mixed hypervisor clusters, offline nodes,
422 - Added a customizable HV_KERNEL_ARGS hypervisor parameter (for Xen PVM
424 - Fix an issue related to $libdir/run/ganeti and cluster creation
430 - Xen PVM and KVM have switched the default value for the instance root
431 disk to the first partition on the first drive, instead of the whole
432 drive; this means that the OS installation scripts must be changed
434 - Man pages have been updated
435 - RAPI has been switched by default to HTTPS, and the exported functions
436 should all work correctly
437 - RAPI v1 has been removed
438 - Many improvements to the KVM hypervisor
439 - Block device errors are now better reported
440 - Many other bugfixes and small improvements
446 - Version 2 is a general rewrite of the code and therefore the
447 differences are too many to list, see the design document for 2.0 in
448 the ``doc/`` subdirectory for more details
449 - In this beta version there is not yet a migration path from 1.2 (there
450 will be one in the final 2.0 release)
451 - A few significant changes are:
453 - all commands are executed by a daemon (``ganeti-masterd``) and the
454 various ``gnt-*`` commands are just front-ends to it
455 - all the commands are entered into, and executed from a job queue,
456 see the ``gnt-job(8)`` manpage
457 - the RAPI daemon supports read-write operations, secured by basic
458 HTTP authentication on top of HTTPS
459 - DRBD version 0.7 support has been removed, DRBD 8 is the only
460 supported version (when migrating from Ganeti 1.2 to 2.0, you need
461 to migrate to DRBD 8 first while still running Ganeti 1.2)
462 - DRBD devices are using statically allocated minor numbers, which
463 will be assigned to existing instances during the migration process
464 - there is support for both Xen PVM and Xen HVM instances running on
466 - KVM virtualization is supported too
467 - file-based storage has been implemented, which means that it is
468 possible to run the cluster without LVM and DRBD storage, for
469 example using a shared filesystem exported from shared storage (and
470 still have live migration)
476 - Change the default reboot type in ``gnt-instance reboot`` to "hard"
477 - Reuse the old instance mac address by default on instance import, if
478 the instance name is the same.
479 - Handle situations in which the node info rpc returns incomplete
481 - Add checks for tcp/udp ports collisions in ``gnt-cluster verify``
482 - Improved version of batcher:
485 - instance mac address support
486 - support for HVM clusters/instances
488 - Add an option to show the number of cpu sockets and nodes in
490 - Support OSes that handle more than one version of the OS api (but do
491 not change the current API in any other way)
492 - Fix ``gnt-node migrate``
493 - ``gnt-debug`` man page
494 - Fixes various more typos and small issues
495 - Increase disk resync maximum speed to 60MB/s (from 30MB/s)
501 - new ``--hvm-nic-type`` and ``--hvm-disk-type`` flags to control the
502 type of disk exported to fully virtualized instances.
503 - provide access to the serial console of HVM instances
504 - instance auto_balance flag, set by default. If turned off it will
505 avoid warnings on cluster verify if there is not enough memory to fail
506 over an instance. in the future it will prevent automatically failing
507 it over when we will support that.
508 - batcher tool for instance creation, see ``tools/README.batcher``
509 - ``gnt-instance reinstall --select-os`` to interactively select a new
510 operating system when reinstalling an instance.
511 - when changing the memory amount on instance modify a check has been
512 added that the instance will be able to start. also warnings are
513 emitted if the instance will not be able to fail over, if auto_balance
515 - documentation fixes
516 - sync fields between ``gnt-instance list/modify/add/import``
517 - fix a race condition in drbd when the sync speed was set after giving
518 the device a remote peer.
524 - note: the allowed size and number of tags per object were reduced
525 - fix a bug in ``gnt-cluster verify`` with inconsistent volume groups
526 - fixed twisted 8.x compatibility
527 - fixed ``gnt-instance replace-disks`` with iallocator
528 - add TCP keepalives on twisted connections to detect restarted nodes
529 - disk increase support, see ``gnt-instance grow-disk``
530 - implement bulk node/instance query for RAPI
531 - add tags in node/instance listing (optional)
532 - experimental migration (and live migration) support, read the man page
533 for ``gnt-instance migrate``
534 - the ``ganeti-watcher`` logs are now timestamped, and the watcher also
535 has some small improvements in handling its state file
541 - Experimental readonly, REST-based remote API implementation;
542 automatically started on master node, TCP port 5080, if enabled by
543 ``--enable-rapi`` parameter to configure script.
544 - Instance allocator support. Add and import instance accept a
545 ``--iallocator`` parameter, and call that instance allocator to decide
546 which node to use for the instance. The iallocator document describes
547 what's expected from an allocator script.
548 - ``gnt-cluster verify`` N+1 memory redundancy checks: Unless passed the
549 ``--no-nplus1-mem`` option ``gnt-cluster verify`` now checks that if a
550 node is lost there is still enough memory to fail over the instances
552 - ``gnt-cluster verify`` hooks: it is now possible to add post-hooks to
553 ``gnt-cluster verify``, to check for site-specific compliance. All the
554 hooks will run, and their output, if any, will be displayed. Any
555 failing hook will make the verification return an error value.
556 - ``gnt-cluster verify`` now checks that its peers are reachable on the
557 primary and secondary interfaces
558 - ``gnt-node add`` now supports the ``--readd`` option, to readd a node
559 that is still declared as part of the cluster and has failed.
560 - ``gnt-* list`` commands now accept a new ``-o +field`` way of
561 specifying output fields, that just adds the chosen fields to the
563 - ``gnt-backup`` now has a new ``remove`` command to delete an existing
564 export from the filesystem.
565 - New per-instance parameters hvm_acpi, hvm_pae and hvm_cdrom_image_path
566 have been added. Using them you can enable/disable acpi and pae
567 support, and specify a path for a cd image to be exported to the
568 instance. These parameters as the name suggest only work on HVM
570 - When upgrading an HVM cluster to Ganeti 1.2.4, the values for ACPI and
571 PAE support will be set to the previously hardcoded values, but the
572 (previously hardcoded) path to the CDROM ISO image will be unset and
573 if required, needs to be set manually with ``gnt-instance modify``
575 - The address to which an instance's VNC console is bound is now
576 selectable per-instance, rather than being cluster wide. Of course
577 this only applies to instances controlled via VNC, so currently just
578 applies to HVM clusters.
584 - more tweaks to the disk activation code (especially helpful for DRBD)
585 - change the default ``gnt-instance list`` output format, now there is
586 one combined status field (see the manpage for the exact values this
588 - some more fixes for the mac export to hooks change
589 - make Ganeti not break with DRBD 8.2.x (which changed the version
590 format in ``/proc/drbd``) (issue 24)
591 - add an upgrade tool from "remote_raid1" disk template to "drbd" disk
592 template, allowing migration from DRBD0.7+MD to DRBD8
598 - fix ``gnt-instance modify`` breakage introduced in 1.2.1 with the HVM
600 - add command aliases infrastructure and a few aliases
601 - allow listing of VCPUs in the ``gnt-instance list`` and improve the
602 man pages and the ``--help`` option of ``gnt-node
603 list``/``gnt-instance list``
604 - fix ``gnt-backup list`` with down nodes (issue 21)
605 - change the tools location (move from $pkgdatadir to $pkglibdir/tools)
606 - fix the dist archive and add a check for including svn/git files in
608 - some developer-related changes: improve the burnin and the QA suite,
609 add an upload script for testing during development
615 - experimental HVM support, read the install document, section
616 "Initializing the cluster"
617 - allow for the PVM hypervisor per-instance kernel and initrd paths
618 - add a new command ``gnt-cluster verify-disks`` which uses a new
619 algorithm to improve the reconnection of the DRBD pairs if the device
620 on the secondary node has gone away
621 - make logical volume code auto-activate LVs at disk activation time
622 - slightly improve the speed of activating disks
623 - allow specification of the MAC address at instance creation time, and
624 changing it later via ``gnt-instance modify``
625 - fix handling of external commands that generate lots of output on
627 - update documentation with regard to minimum version of DRBD8 supported
633 - Log the ``xm create`` output to the node daemon log on failure (to
634 help diagnosing the error)
635 - In debug mode, log all external commands output if failed to the logs
636 - Change parsing of lvm commands to ignore stderr
642 - Another round of updates to the DRBD 8 code to deal with more failures
643 in the replace secondary node operation
644 - Some more logging of failures in disk operations (lvm, drbd)
645 - A few documentation updates
652 - Change configuration file format from Python's Pickle to JSON.
653 Upgrading is possible using the cfgupgrade utility.
654 - Add support for DRBD 8.0 (new disk template ``drbd``) which allows for
655 faster replace disks and is more stable (DRBD 8 has many improvements
656 compared to DRBD 0.7)
657 - Added command line tags support (see man pages for ``gnt-instance``,
658 ``gnt-node``, ``gnt-cluster``)
659 - Added instance rename support
660 - Added multi-instance startup/shutdown
661 - Added cluster rename support
662 - Added ``gnt-node evacuate`` to simplify some node operations
663 - Added instance reboot operation that can speedup reboot as compared to
665 - Soften the requirement that hostnames are in FQDN format
666 - The ``ganeti-watcher`` now activates drbd pairs after secondary node
668 - Removed dependency on debian's patched fping that uses the
669 non-standard ``-S`` option
670 - Now the OS definitions are searched for in multiple, configurable
671 paths (easier for distros to package)
672 - Some changes to the hooks infrastructure (especially the new
673 post-configuration update hook)
674 - Other small bugfixes
676 .. vim: set textwidth=72 :