9 Another release with a long development cycle, during which many
10 different features were added.
15 The KVM hypervisor now can run the individual instances as non-root, to
16 reduce the impact of a VM being hijacked due to bugs in the
17 hypervisor. It is possible to run all instances as a single (non-root)
18 user, to manually specify a user for each instance, or to dynamically
19 allocate a user out of a cluster-wide pool to each instance, with the
20 guarantee that no two instances will run under the same user ID on any
23 An experimental RAPI client library, that can be used standalone
24 (without the other Ganeti libraries), is provided in the source tree as
25 ``lib/rapi/client.py``. Note this client might change its interface in
26 the future, as we iterate on its capabilities.
28 A new command, ``gnt-cluster renew-crypto`` has been added to easily
29 replace the cluster's certificates and crypto keys. This might help in
30 case they have been compromised, or have simply expired.
32 A new disk option for instance creation has been added that allows one
33 to "adopt" currently existing logical volumes, with data
34 preservation. This should allow easier migration to Ganeti from
35 unmanaged (or managed via other software) instances.
37 Another disk improvement is the possibility to convert between redundant
38 (DRBD) and plain (LVM) disk configuration for an instance. This should
39 allow better scalability (starting with one node and growing the
40 cluster, or shrinking a two-node cluster to one node).
42 A new feature that could help with automated node failovers has been
43 implemented: if a node sees itself as offline (by querying the master
44 candidates), it will try to shutdown (hard) all instances and any active
45 DRBD devices. This reduces the risk of duplicate instances if an
46 external script automatically failovers the instances on such nodes. To
47 enable this, the cluster parameter ``maintain_node_health`` should be
48 enabled; in the future this option (per the name) will enable other
49 automatic maintenance features.
51 Instance export/import now will reuse the original instance
52 specifications for all parameters; that means exporting an instance,
53 deleting it and the importing it back should give an almost identical
54 instance. Note that the default import behaviour has changed from
55 before, where it created only one NIC; now it recreates the original
58 Cluster verify has added a few new checks: SSL certificates validity,
59 /etc/hosts consistency across the cluster, etc.
64 As usual, many internal changes were done, documentation fixes,
67 - Fixed cluster initialization with disabled cluster storage (regression
69 - File-based storage supports growing the disks
70 - Fixed behaviour of node role changes
71 - Fixed cluster verify for some corner cases, plus a general rewrite of
72 cluster verify to allow future extension with more checks
73 - Fixed log spamming by watcher and node daemon (regression introduced
75 - Fixed possible validation issues when changing the list of enabled
77 - Fixed cleanup of /etc/hosts during node removal
78 - Fixed RAPI response for invalid methods
79 - Fixed bug with hashed passwords in ``ganeti-rapi`` daemon
80 - Multiple small improvements to the KVM hypervisor (VNC usage, booting
82 - Allow OS changes without re-installation (to record a changed OS
83 outside of Ganeti, or to allow OS renames)
84 - Allow instance creation without OS installation (useful for example if
85 the OS will be installed manually, or restored from a backup not in
87 - Implemented option to make cluster ``copyfile`` use the replication
89 - Added list of enabled hypervisors to ssconf (possibly useful for
91 - Added a new tool (``tools/cfgupgrade12``) that allows upgrading from
93 - A partial form of node re-IP is possible via node readd, which now
94 allows changed node primary IP
95 - Command line utilities now show an informational message if the job is
97 - The logs of the master daemon now show the PID/UID/GID of the
105 *(Released Fri, 12 Mar 2010)*
107 During the 2.1.0 long release candidate cycle, a lot of improvements and
108 changes have accumulated with were released later as 2.1.1.
113 The node evacuate command (``gnt-node evacuate``) was significantly
114 rewritten, and as such the IAllocator protocol was changed - a new
115 request type has been added. This unfortunate change during a stable
116 series is designed to improve performance of node evacuations; on
117 clusters with more than about five nodes and which are well-balanced,
118 evacuation should proceed in parallel for all instances of the node
119 being evacuated. As such, any existing IAllocator scripts need to be
120 updated, otherwise the above command will fail due to the unknown
121 request. The provided "dumb" allocator has not been updated; but the
122 ganeti-htools package supports the new protocol since version 0.2.4.
124 Another important change is increased validation of node and instance
125 names. This might create problems in special cases, if invalid host
126 names are being used.
128 Also, a new layer of hypervisor parameters has been added, that sits at
129 OS level between the cluster defaults and the instance ones. This allows
130 customisation of virtualization parameters depending on the installed
131 OS. For example instances with OS 'X' may have a different KVM kernel
132 (or any other parameter) than the cluster defaults. This is intended to
133 help managing a multiple OSes on the same cluster, without manual
134 modification of each instance's parameters.
136 A tool for merging clusters, ``cluster-merge``, has been added in the
142 - Improved the int/float conversions that should make the code more
143 robust in face of errors from the node daemons
144 - Fixed the remove node code in case of internal configuration errors
145 - Fixed the node daemon behaviour in face of inconsistent queue
146 directory (e.g. read-only file-system where we can't open the files
148 - Fixed the behaviour of gnt-node modify for master candidate demotion;
149 now it either aborts cleanly or, if given the new “auto_promote”
150 parameter, will automatically promote other nodes as needed
151 - Fixed compatibility with (unreleased yet) Python 2.6.5 that would
152 completely prevent Ganeti from working
153 - Fixed bug for instance export when not all disks were successfully
155 - Fixed behaviour of node add when the new node is slow in starting up
157 - Fixed handling of signals in the LUXI client, which should improve
158 behaviour of command-line scripts
159 - Added checks for invalid node/instance names in the configuration (now
160 flagged during cluster verify)
161 - Fixed watcher behaviour for disk activation errors
162 - Fixed two potentially endless loops in http library, which led to the
163 RAPI daemon hanging and consuming 100% CPU in some cases
164 - Fixed bug in RAPI daemon related to hashed passwords
165 - Fixed bug for unintended qemu-level bridging of multi-NIC KVM
167 - Enhanced compatibility with non-Debian OSes, but not using absolute
168 path in some commands and allowing customisation of the ssh
169 configuration directory
170 - Fixed possible future issue with new Python versions by abiding to the
171 proper use of ``__slots__`` attribute on classes
172 - Added checks that should prevent directory traversal attacks
173 - Many documentation fixes based on feedback from users
178 - Added an “early_release” more for instance replace disks and node
179 evacuate, where we release locks earlier and thus allow higher
180 parallelism within the cluster
181 - Added watcher hooks, intended to allow the watcher to restart other
182 daemons (e.g. from the ganeti-nbma project), but they can be used of
183 course for any other purpose
184 - Added a compile-time disable for DRBD barriers, to increase
185 performance if the administrator trusts the power supply or the
186 storage system to not lose writes
187 - Added the option of using syslog for logging instead of, or in
188 addition to, Ganeti's own log files
189 - Removed boot restriction for paravirtual NICs for KVM, recent versions
190 can indeed boot from a paravirtual NIC
191 - Added a generic debug level for many operations; while this is not
192 used widely yet, it allows one to pass the debug value all the way to
194 - Enhanced the hooks environment for instance moves (failovers,
195 migrations) where the primary/secondary nodes changed during the
196 operation, by adding {NEW,OLD}_{PRIMARY,SECONDARY} vars
197 - Enhanced data validations for many user-supplied values; one important
198 item is the restrictions imposed on instance and node names, which
199 might reject some (invalid) host names
200 - Add a configure-time option to disable file-based storage, if it's not
201 needed; this allows greater security separation between the master
202 node and the other nodes from the point of view of the inter-node RPC
204 - Added user notification in interactive tools if job is waiting in the
205 job queue or trying to acquire locks
206 - Added log messages when a job is waiting for locks
207 - Added filtering by node tags in instance operations which admit
208 multiple instances (start, stop, reboot, reinstall)
209 - Added a new tool for cluster mergers, ``cluster-merge``
210 - Parameters from command line which are of the form ``a=b,c=d`` can now
211 use backslash escapes to pass in values which contain commas,
212 e.g. ``a=b\\c,d=e`` where the 'a' parameter would get the value
214 - For KVM, the instance name is the first parameter passed to KVM, so
215 that it's more visible in the process list
221 *(Released Tue, 2 Mar 2010)*
223 Ganeti 2.1 brings many improvements with it. Major changes:
225 - Added infrastructure to ease automated disk repairs
226 - Added new daemon to export configuration data in a cheaper way than
228 - Instance NICs can now be routed instead of being associated with a
230 - Improved job locking logic to reduce impact of jobs acquiring multiple
231 locks waiting for other long-running jobs
233 In-depth implementation details can be found in the Ganeti 2.1 design
239 - Added chroot hypervisor
240 - Added more options to xen-hvm hypervisor (``kernel_path`` and
242 - Added more options to xen-pvm hypervisor (``use_bootloader``,
243 ``bootloader_path`` and ``bootloader_args``)
244 - Added the ``use_localtime`` option for the xen-hvm and kvm
245 hypervisors, and the default value for this has changed to false (in
246 2.0 xen-hvm always enabled it)
247 - Added luxi call to submit multiple jobs in one go
248 - Added cluster initialization option to not modify ``/etc/hosts``
250 - Added network interface parameters
251 - Added dry run mode to some LUs
252 - Added RAPI resources:
254 - ``/2/instances/[instance_name]/info``
255 - ``/2/instances/[instance_name]/replace-disks``
256 - ``/2/nodes/[node_name]/evacuate``
257 - ``/2/nodes/[node_name]/migrate``
258 - ``/2/nodes/[node_name]/role``
259 - ``/2/nodes/[node_name]/storage``
260 - ``/2/nodes/[node_name]/storage/modify``
261 - ``/2/nodes/[node_name]/storage/repair``
263 - Added OpCodes to evacuate or migrate all instances on a node
264 - Added new command to list storage elements on nodes (``gnt-node
265 list-storage``) and modify them (``gnt-node modify-storage``)
266 - Added new ssconf files with master candidate IP address
267 (``ssconf_master_candidates_ips``), node primary IP address
268 (``ssconf_node_primary_ips``) and node secondary IP address
269 (``ssconf_node_secondary_ips``)
270 - Added ``ganeti-confd`` and a client library to query the Ganeti
271 configuration via UDP
272 - Added ability to run hooks after cluster initialization and before
274 - Added automatic mode for disk replace (``gnt-instance replace-disks
276 - Added ``gnt-instance recreate-disks`` to re-create (empty) disks
277 after catastrophic data-loss
278 - Added ``gnt-node repair-storage`` command to repair damaged LVM volume
280 - Added ``gnt-instance move`` command to move instances
281 - Added ``gnt-cluster watcher`` command to control watcher
282 - Added ``gnt-node powercycle`` command to powercycle nodes
283 - Added new job status field ``lock_status``
284 - Added parseable error codes to cluster verification (``gnt-cluster
285 verify --error-codes``) and made output less verbose (use
286 ``--verbose`` to restore previous behaviour)
287 - Added UUIDs to the main config entities (cluster, nodes, instances)
288 - Added support for OS variants
289 - Added support for hashed passwords in the Ganeti remote API users file
291 - Added option to specify maximum timeout on instance shutdown
292 - Added ``--no-ssh-init`` option to ``gnt-cluster init``
293 - Added new helper script to start and stop Ganeti daemons
294 (``daemon-util``), with the intent to reduce the work necessary to
295 adjust Ganeti for non-Debian distributions and to start/stop daemons
297 - Added more unittests
298 - Fixed critical bug in ganeti-masterd startup
299 - Removed the configure-time ``kvm-migration-port`` parameter, this is
300 now customisable at the cluster level for both the KVM and Xen
301 hypervisors using the new ``migration_port`` parameter
302 - Pass ``INSTANCE_REINSTALL`` variable to OS installation script when
303 reinstalling an instance
304 - Allowed ``@`` in tag names
305 - Migrated to Sphinx (http://sphinx.pocoo.org/) for documentation
306 - Many documentation updates
307 - Distribute hypervisor files on ``gnt-cluster redist-conf``
308 - ``gnt-instance reinstall`` can now reinstall multiple instances
309 - Updated many command line parameters
310 - Introduced new OS API version 15
311 - No longer support a default hypervisor
312 - Treat virtual LVs as inexistent
313 - Improved job locking logic to reduce lock contention
314 - Match instance and node names case insensitively
315 - Reimplemented bash completion script to be more complete
322 *(Released Thu, 4 Feb 2010)*
324 - Fix cleaner behaviour on nodes not in a cluster (Debian bug 568105)
325 - Fix a string formatting bug
326 - Improve safety of the code in some error paths
327 - Improve data validation in the master of values returned from nodes
333 *(Released Thu, 17 Dec 2009)*
335 - Fix security issue due to missing validation of iallocator names; this
336 allows local and remote execution of arbitrary executables
337 - Fix failure of gnt-node list during instance removal
338 - Ship the RAPI documentation in the archive
344 *(Released Wed, 30 Sep 2009)*
346 - Fixed many wrong messages
347 - Fixed a few bugs related to the locking library
348 - Fixed MAC checking at instance creation time
349 - Fixed a DRBD parsing bug related to gaps in /proc/drbd
350 - Fixed a few issues related to signal handling in both daemons and
352 - Fixed the example startup script provided
353 - Fixed insserv dependencies in the example startup script (patch from
355 - Fixed handling of drained nodes in the iallocator framework
356 - Fixed handling of KERNEL_PATH parameter for xen-hvm (Debian bug
358 - Fixed error related to invalid job IDs in job polling
359 - Fixed job/opcode persistence on unclean master shutdown
360 - Fixed handling of partial job processing after unclean master
362 - Fixed error reporting from LUs, previously all errors were converted
363 into execution errors
364 - Fixed error reporting from burnin
365 - Decreased significantly the memory usage of the job queue
366 - Optimised slightly multi-job submission
367 - Optimised slightly opcode loading
368 - Backported the multi-job submit framework from the development
369 branch; multi-instance start and stop should be faster
370 - Added script to clean archived jobs after 21 days; this will reduce
371 the size of the queue directory
372 - Added some extra checks in disk size tracking
373 - Added an example ethers hook script
374 - Added a cluster parameter that prevents Ganeti from modifying of
376 - Added more node information to RAPI responses
377 - Added a ``gnt-job watch`` command that allows following the ouput of a
379 - Added a bind-address option to ganeti-rapi
380 - Added more checks to the configuration verify
381 - Enhanced the burnin script such that some operations can be retried
383 - Converted instance reinstall to multi-instance model
389 *(Released Fri, 7 Aug 2009)*
391 - Added ``--ignore-size`` to the ``gnt-instance activate-disks`` command
392 to allow using the pre-2.0.2 behaviour in activation, if any existing
393 instances have mismatched disk sizes in the configuration
394 - Added ``gnt-cluster repair-disk-sizes`` command to check and update
395 any configuration mismatches for disk sizes
396 - Added ``gnt-master cluste-failover --no-voting`` to allow master
397 failover to work on two-node clusters
398 - Fixed the ``--net`` option of ``gnt-backup import``, which was
400 - Fixed detection of OS script errors in ``gnt-backup export``
401 - Fixed exit code of ``gnt-backup export``
407 *(Released Fri, 17 Jul 2009)*
409 - Added experimental support for stripped logical volumes; this should
410 enhance performance but comes with a higher complexity in the block
411 device handling; stripping is only enabled when passing
412 ``--with-lvm-stripecount=N`` to ``configure``, but codepaths are
413 affected even in the non-stripped mode
414 - Improved resiliency against transient failures at the end of DRBD
415 resyncs, and in general of DRBD resync checks
416 - Fixed a couple of issues with exports and snapshot errors
417 - Fixed a couple of issues in instance listing
418 - Added display of the disk size in ``gnt-instance info``
419 - Fixed checking for valid OSes in instance creation
420 - Fixed handling of the "vcpus" parameter in instance listing and in
421 general of invalid parameters
422 - Fixed http server library, and thus RAPI, to handle invalid
423 username/password combinations correctly; this means that now they
424 report unauthorized for queries too, not only for modifications,
425 allowing earlier detect of configuration problems
426 - Added a new "role" node list field, equivalent to the master/master
427 candidate/drained/offline flags combinations
428 - Fixed cluster modify and changes of candidate pool size
429 - Fixed cluster verify error messages for wrong files on regular nodes
430 - Fixed a couple of issues with node demotion from master candidate role
431 - Fixed node readd issues
432 - Added non-interactive mode for ``ganeti-masterd --no-voting`` startup
433 - Added a new ``--no-voting`` option for masterfailover to fix failover
434 on two-nodes clusters when the former master node is unreachable
435 - Added instance reinstall over RAPI
441 *(Released Tue, 16 Jun 2009)*
443 - added ``-H``/``-B`` startup parameters to ``gnt-instance``, which will
444 allow re-adding the start in single-user option (regression from 1.2)
445 - the watcher writes the instance status to a file, to allow monitoring
446 to report the instance status (from the master) based on cached
447 results of the watcher's queries; while this can get stale if the
448 watcher is being locked due to other work on the cluster, this is
450 - the watcher now also restarts the node daemon and the rapi daemon if
452 - fixed the watcher to handle full and drained queue cases
453 - hooks export more instance data in the environment, which helps if
454 hook scripts need to take action based on the instance's properties
455 (no longer need to query back into ganeti)
456 - instance failovers when the instance is stopped do not check for free
457 RAM, so that failing over a stopped instance is possible in low memory
459 - rapi uses queries for tags instead of jobs (for less job traffic), and
460 for cluster tags it won't talk to masterd at all but read them from
462 - a couple of error handling fixes in RAPI
463 - drbd handling: improved the error handling of inconsistent disks after
464 resync to reduce the frequency of "there are some degraded disks for
465 this instance" messages
466 - fixed a bug in live migration when DRBD doesn't want to reconnect (the
467 error handling path called a wrong function name)
473 *(Released Wed, 27 May 2009)*
475 - no changes from rc5
478 Version 2.0 release candidate 5
479 -------------------------------
481 *(Released Wed, 20 May 2009)*
483 - fix a couple of bugs (validation, argument checks)
484 - fix ``gnt-cluster getmaster`` on non-master nodes (regression)
485 - some small improvements to RAPI and IAllocator
486 - make watcher automatically start the master daemon if down
489 Version 2.0 release candidate 4
490 -------------------------------
492 *(Released Mon, 27 Apr 2009)*
494 - change the OS list to not require locks; this helps with big clusters
495 - fix ``gnt-cluster verify`` and ``gnt-cluster verify-disks`` when the
496 volume group is broken
497 - ``gnt-instance info``, without any arguments, doesn't run for all
498 instances anymore; either pass ``--all`` or pass the desired
499 instances; this helps against mistakes on big clusters where listing
500 the information for all instances takes a long time
501 - miscellaneous doc and man pages fixes
504 Version 2.0 release candidate 3
505 -------------------------------
507 *(Released Wed, 8 Apr 2009)*
509 - Change the internal locking model of some ``gnt-node`` commands, in
510 order to reduce contention (and blocking of master daemon) when
511 batching many creation/reinstall jobs
512 - Fixes to Xen soft reboot
513 - No longer build documentation at build time, instead distribute it in
514 the archive, in order to reduce the need for the whole docbook/rst
518 Version 2.0 release candidate 2
519 -------------------------------
521 *(Released Fri, 27 Mar 2009)*
523 - Now the cfgupgrade scripts works and can upgrade 1.2.7 clusters to 2.0
524 - Fix watcher startup sequence, improves the behaviour of busy clusters
525 - Some other fixes in ``gnt-cluster verify``, ``gnt-instance
526 replace-disks``, ``gnt-instance add``, ``gnt-cluster queue``, KVM VNC
527 bind address and other places
528 - Some documentation fixes and updates
531 Version 2.0 release candidate 1
532 -------------------------------
534 *(Released Mon, 2 Mar 2009)*
536 - More documentation updates, now all docs should be more-or-less
538 - A couple of small fixes (mixed hypervisor clusters, offline nodes,
540 - Added a customizable HV_KERNEL_ARGS hypervisor parameter (for Xen PVM
542 - Fix an issue related to $libdir/run/ganeti and cluster creation
548 *(Released Thu, 19 Feb 2009)*
550 - Xen PVM and KVM have switched the default value for the instance root
551 disk to the first partition on the first drive, instead of the whole
552 drive; this means that the OS installation scripts must be changed
554 - Man pages have been updated
555 - RAPI has been switched by default to HTTPS, and the exported functions
556 should all work correctly
557 - RAPI v1 has been removed
558 - Many improvements to the KVM hypervisor
559 - Block device errors are now better reported
560 - Many other bugfixes and small improvements
566 *(Released Mon, 26 Jan 2009)*
568 - Version 2 is a general rewrite of the code and therefore the
569 differences are too many to list, see the design document for 2.0 in
570 the ``doc/`` subdirectory for more details
571 - In this beta version there is not yet a migration path from 1.2 (there
572 will be one in the final 2.0 release)
573 - A few significant changes are:
575 - all commands are executed by a daemon (``ganeti-masterd``) and the
576 various ``gnt-*`` commands are just front-ends to it
577 - all the commands are entered into, and executed from a job queue,
578 see the ``gnt-job(8)`` manpage
579 - the RAPI daemon supports read-write operations, secured by basic
580 HTTP authentication on top of HTTPS
581 - DRBD version 0.7 support has been removed, DRBD 8 is the only
582 supported version (when migrating from Ganeti 1.2 to 2.0, you need
583 to migrate to DRBD 8 first while still running Ganeti 1.2)
584 - DRBD devices are using statically allocated minor numbers, which
585 will be assigned to existing instances during the migration process
586 - there is support for both Xen PVM and Xen HVM instances running on
588 - KVM virtualization is supported too
589 - file-based storage has been implemented, which means that it is
590 possible to run the cluster without LVM and DRBD storage, for
591 example using a shared filesystem exported from shared storage (and
592 still have live migration)
598 *(Released Tue, 13 Jan 2009)*
600 - Change the default reboot type in ``gnt-instance reboot`` to "hard"
601 - Reuse the old instance mac address by default on instance import, if
602 the instance name is the same.
603 - Handle situations in which the node info rpc returns incomplete
605 - Add checks for tcp/udp ports collisions in ``gnt-cluster verify``
606 - Improved version of batcher:
609 - instance mac address support
610 - support for HVM clusters/instances
612 - Add an option to show the number of cpu sockets and nodes in
614 - Support OSes that handle more than one version of the OS api (but do
615 not change the current API in any other way)
616 - Fix ``gnt-node migrate``
617 - ``gnt-debug`` man page
618 - Fixes various more typos and small issues
619 - Increase disk resync maximum speed to 60MB/s (from 30MB/s)
625 *(Released Wed, 24 Sep 2008)*
627 - new ``--hvm-nic-type`` and ``--hvm-disk-type`` flags to control the
628 type of disk exported to fully virtualized instances.
629 - provide access to the serial console of HVM instances
630 - instance auto_balance flag, set by default. If turned off it will
631 avoid warnings on cluster verify if there is not enough memory to fail
632 over an instance. in the future it will prevent automatically failing
633 it over when we will support that.
634 - batcher tool for instance creation, see ``tools/README.batcher``
635 - ``gnt-instance reinstall --select-os`` to interactively select a new
636 operating system when reinstalling an instance.
637 - when changing the memory amount on instance modify a check has been
638 added that the instance will be able to start. also warnings are
639 emitted if the instance will not be able to fail over, if auto_balance
641 - documentation fixes
642 - sync fields between ``gnt-instance list/modify/add/import``
643 - fix a race condition in drbd when the sync speed was set after giving
644 the device a remote peer.
650 *(Released Tue, 22 Jul 2008)*
652 - note: the allowed size and number of tags per object were reduced
653 - fix a bug in ``gnt-cluster verify`` with inconsistent volume groups
654 - fixed twisted 8.x compatibility
655 - fixed ``gnt-instance replace-disks`` with iallocator
656 - add TCP keepalives on twisted connections to detect restarted nodes
657 - disk increase support, see ``gnt-instance grow-disk``
658 - implement bulk node/instance query for RAPI
659 - add tags in node/instance listing (optional)
660 - experimental migration (and live migration) support, read the man page
661 for ``gnt-instance migrate``
662 - the ``ganeti-watcher`` logs are now timestamped, and the watcher also
663 has some small improvements in handling its state file
669 *(Released Fri, 13 Jun 2008)*
671 - Experimental readonly, REST-based remote API implementation;
672 automatically started on master node, TCP port 5080, if enabled by
673 ``--enable-rapi`` parameter to configure script.
674 - Instance allocator support. Add and import instance accept a
675 ``--iallocator`` parameter, and call that instance allocator to decide
676 which node to use for the instance. The iallocator document describes
677 what's expected from an allocator script.
678 - ``gnt-cluster verify`` N+1 memory redundancy checks: Unless passed the
679 ``--no-nplus1-mem`` option ``gnt-cluster verify`` now checks that if a
680 node is lost there is still enough memory to fail over the instances
682 - ``gnt-cluster verify`` hooks: it is now possible to add post-hooks to
683 ``gnt-cluster verify``, to check for site-specific compliance. All the
684 hooks will run, and their output, if any, will be displayed. Any
685 failing hook will make the verification return an error value.
686 - ``gnt-cluster verify`` now checks that its peers are reachable on the
687 primary and secondary interfaces
688 - ``gnt-node add`` now supports the ``--readd`` option, to readd a node
689 that is still declared as part of the cluster and has failed.
690 - ``gnt-* list`` commands now accept a new ``-o +field`` way of
691 specifying output fields, that just adds the chosen fields to the
693 - ``gnt-backup`` now has a new ``remove`` command to delete an existing
694 export from the filesystem.
695 - New per-instance parameters hvm_acpi, hvm_pae and hvm_cdrom_image_path
696 have been added. Using them you can enable/disable acpi and pae
697 support, and specify a path for a cd image to be exported to the
698 instance. These parameters as the name suggest only work on HVM
700 - When upgrading an HVM cluster to Ganeti 1.2.4, the values for ACPI and
701 PAE support will be set to the previously hardcoded values, but the
702 (previously hardcoded) path to the CDROM ISO image will be unset and
703 if required, needs to be set manually with ``gnt-instance modify``
705 - The address to which an instance's VNC console is bound is now
706 selectable per-instance, rather than being cluster wide. Of course
707 this only applies to instances controlled via VNC, so currently just
708 applies to HVM clusters.
714 *(Released Mon, 18 Feb 2008)*
716 - more tweaks to the disk activation code (especially helpful for DRBD)
717 - change the default ``gnt-instance list`` output format, now there is
718 one combined status field (see the manpage for the exact values this
720 - some more fixes for the mac export to hooks change
721 - make Ganeti not break with DRBD 8.2.x (which changed the version
722 format in ``/proc/drbd``) (issue 24)
723 - add an upgrade tool from "remote_raid1" disk template to "drbd" disk
724 template, allowing migration from DRBD0.7+MD to DRBD8
730 *(Released Wed, 30 Jan 2008)*
732 - fix ``gnt-instance modify`` breakage introduced in 1.2.1 with the HVM
734 - add command aliases infrastructure and a few aliases
735 - allow listing of VCPUs in the ``gnt-instance list`` and improve the
736 man pages and the ``--help`` option of ``gnt-node
737 list``/``gnt-instance list``
738 - fix ``gnt-backup list`` with down nodes (issue 21)
739 - change the tools location (move from $pkgdatadir to $pkglibdir/tools)
740 - fix the dist archive and add a check for including svn/git files in
742 - some developer-related changes: improve the burnin and the QA suite,
743 add an upload script for testing during development
749 *(Released Wed, 16 Jan 2008)*
751 - experimental HVM support, read the install document, section
752 "Initializing the cluster"
753 - allow for the PVM hypervisor per-instance kernel and initrd paths
754 - add a new command ``gnt-cluster verify-disks`` which uses a new
755 algorithm to improve the reconnection of the DRBD pairs if the device
756 on the secondary node has gone away
757 - make logical volume code auto-activate LVs at disk activation time
758 - slightly improve the speed of activating disks
759 - allow specification of the MAC address at instance creation time, and
760 changing it later via ``gnt-instance modify``
761 - fix handling of external commands that generate lots of output on
763 - update documentation with regard to minimum version of DRBD8 supported
769 *(Released Tue, 4 Dec 2007)*
771 - Log the ``xm create`` output to the node daemon log on failure (to
772 help diagnosing the error)
773 - In debug mode, log all external commands output if failed to the logs
774 - Change parsing of lvm commands to ignore stderr
780 *(Released Wed, 28 Nov 2007)*
782 - Another round of updates to the DRBD 8 code to deal with more failures
783 in the replace secondary node operation
784 - Some more logging of failures in disk operations (lvm, drbd)
785 - A few documentation updates
792 *(Released Tue, 13 Nov 2007)*
794 - Change configuration file format from Python's Pickle to JSON.
795 Upgrading is possible using the cfgupgrade utility.
796 - Add support for DRBD 8.0 (new disk template ``drbd``) which allows for
797 faster replace disks and is more stable (DRBD 8 has many improvements
798 compared to DRBD 0.7)
799 - Added command line tags support (see man pages for ``gnt-instance``,
800 ``gnt-node``, ``gnt-cluster``)
801 - Added instance rename support
802 - Added multi-instance startup/shutdown
803 - Added cluster rename support
804 - Added ``gnt-node evacuate`` to simplify some node operations
805 - Added instance reboot operation that can speedup reboot as compared to
807 - Soften the requirement that hostnames are in FQDN format
808 - The ``ganeti-watcher`` now activates drbd pairs after secondary node
810 - Removed dependency on debian's patched fping that uses the
811 non-standard ``-S`` option
812 - Now the OS definitions are searched for in multiple, configurable
813 paths (easier for distros to package)
814 - Some changes to the hooks infrastructure (especially the new
815 post-configuration update hook)
816 - Other small bugfixes
818 .. vim: set textwidth=72 :