2 - Added “--ignore-size” to the “gnt-instance activate-disks” command
3 to allow using the pre-2.0.2 behaviour in activation, if any
4 existing instances have mismatched disk sizes in the configuration
5 - Added “gnt-cluster repair-disk-sizes” command to check and update
6 any configuration mismatches for disk sizes
7 - Added “gnt-master cluste-failover --no-voting” to allow master
8 failover to work on two-node clusters
9 - Fixed the ‘--net’ option of “gnt-backup import”, which was unusable
10 - Fixed detection of OS script errors in “gnt-backup export”
11 - Fixed exit code of “gnt-backup export”
14 - Added experimental support for stripped logical volumes; this should
15 enhance performance but comes with a higher complexity in the block
16 device handling; stripping is only enabled when passing
17 --with-lvm-stripecount=N to configure, but codepaths are affected
18 even in the non-stripped mode
19 - Improved resiliency against transient failures at the end of DRBD
20 resyncs, and in general of DRBD resync checks
21 - Fixed a couple of issues with exports and snapshot errors
22 - Fixed a couple of issues in instance listing
23 - Added display of the disk size in “gnt-instance info”
24 - Fixed checking for valid OSes in instance creation
25 - Fixed handling of the ‘vcpus’ parameter in instance listing and in
26 general of invalid parameters
27 - Fixed http server library, and thus RAPI, to handle invalid
28 username/password combinations correctly; this means that now they
29 report unauthorized for queries too, not only for modifications,
30 allowing earlier detect of configuration problems
31 - Added a new ‘role’ node list field, equivalent to the master/master
32 candidate/drained/offline flags combinations
33 - Fixed cluster modify and changes of candidate pool size
34 - Fixed cluster verify error messages for wrong files on regular nodes
35 - Fixed a couple of issues with node demotion from master candidate
37 - Fixed node readd issues
38 - Added non-interactive mode for “ganeti-masterd --no-voting” startup
39 - Added a new ‘--no-voting’ option for masterfailover to fix failover
40 on two-nodes clusters when the former master node is unreachable
41 - Added instance reinstall over RAPI
44 - added -H/-B startup parameters to gnt-instance, which will allow
45 re-adding the start in single-user option (regression from 1.2)
46 - the watcher writes the instance status to a file, to allow
47 monitoring to report the instance status (from the master) based on
48 cached results of the watcher's queries; while this can get stale if
49 the watcher is being locked due to other work on the cluster, this
50 is still an improvement
51 - the watcher now also restarts the node daemon and the rapi daemon if
53 - fixed the watcher to handle full and drained queue cases
54 - hooks export more instance data in the environment, which helps if
55 hook scripts need to take action based on the instance's properties
56 (no longer need to query back into ganeti)
57 - instance failovers when the instance is stopped do not check for
58 free RAM, so that failing over a stopped instance is possible in low
60 - rapi uses queries for tags instead of jobs (for less job traffic),
61 and for cluster tags it won't talk to masterd at all but read them
63 - a couple of error handling fixes in RAPI
64 - drbd handling: improved the error handling of inconsistent disks
65 after resync to reduce the frequency of "there are some degraded
66 disks for this instance" message
67 - fixed a bug in live migration when DRBD doesn't want to reconnect
68 (the error handling path called a wrong function name)
73 Version 2.0 release candidate 5
74 - fix a couple of bugs (validation, argument checks)
75 - fix gnt-cluster getmaster on non-master nodes (regression)
76 - some small improvements to RAPI and IAllocator
77 - make watcher automatically start the master daemon if down
79 Version 2.0 release candidate 4
80 - change the OS list to not require locks; this helps with big
82 - fix gnt-cluster verify and gnt-cluster verify-disks when the volume
84 - “gnt-instance info”, without any arguments, doesn't run for all
85 instances anymore; either pass ‘--all’ or pass the desired
86 instances; this helps against mistakes on big clusters where listing
87 the information for all instances takes a long time
88 - miscellaneous doc and man pages fixes
90 Version 2.0 release candidate 3
91 - Change the internal locking model of some gnt-node commands, in
92 order to reduce contention (and blocking of master daemon) when
93 batching many creation/reinstall jobs
94 - Fixes to Xen soft reboot
95 - No longer build documentation at 'make' time, instead distribute it
96 in the archive, in order to reduce the need for the whole
97 docbook/rst toolchains
99 Version 2.0 release candidate 2
100 - Now the cfgupgrade scripts works and can upgrade 1.2.7 clusters to
102 - Fix watcher startup sequence, improves the behaviour of busy
104 - Some other fixes in cluster verify, replace-disks, gnt-instance add,
105 gnt-cluster queue, KVM VNC bind address and other places
106 - Some documentation fixes and updates
108 Version 2.0 release candidate 1
109 - More documentation updates, now all docs should be more-or-less
111 - A couple of small fixes (mixed hypervisor clusters, offline nodes,
113 - Added a customizable HV_KERNEL_ARGS hypervisor parameter (for Xen
115 - Fix an issue related to $libdir/run/ganeti and cluster creation
118 - Xen PVM and KVM have switched the default value for the instance
119 root disk to the first partition on the first drive, instead of the
120 whole drive; this means that the OS installation scripts must be
122 - Man pages have been updated
123 - RAPI has been switched by default to HTTPS, and the exported
124 functions should all work correctly
125 - RAPI v1 has been removed
126 - Many improvements to the KVM hypervisor
127 - Block device errors are now better reported
128 - Many other bugfixes and small improvements
131 - Version 2 is a general rewrite of the code and therefore the
132 differences are too many to list, see the design document for 2.0 in
133 the doc/ subdirectory for more details
134 - In this beta version there is not yet a migration path from 1.2
135 (there will be one in the final 2.0 release)
136 - A few significant changes are:
137 - all commands are executed by a daemon (ganeti-masterd) and the
138 various gnt-* commands are just front-ends to it
139 - all the commands are entered into, and executed from a job queue,
141 - the RAPI daemon supports read-write operations, secured by basic
142 HTTP authentication on top of HTTPS
143 - DRBD version 0.7 support has been removed, DRBD 8 is the only
144 supported version (when migrating from Ganeti 1.2 to 2.0, you need
145 to migrate to DRBD 8 first while still running Ganeti 1.2)
146 - DRBD devices are using statically allocated minor numbers, which
147 will be assigned to existing instances during the migration
149 - there is support for both Xen PVM and Xen HVM instances running on
151 - KVM virtualization is supported too
152 - file-based storage has been implemented, which means that it is
153 possible to run the cluster without LVM and DRBD storage, for
154 example using a shared filesystem exported from shared storage
155 (and still have live migration)
158 - Change the default reboot type in gnt-instance reboot to "hard"
159 - Reuse the old instance mac address by default on instance import,
160 if the instance name is the same.
161 - Handle situations in which the node info rpc returns incomplete results
163 - Add checks for tcp/udp ports collisions in gnt-cluster verify
164 - Improved version of batcher:
166 - instance mac address support
167 - support for HVM clusters/instances
168 - Add an option to show the number of cpu sockets and nodes in gnt-node list
169 - Support OSes that handle more than one version of the OS api
170 (but do not change the current API in any other way)
171 - Fix gnt-node migrate
173 - Fixes various more typos&small issues
174 - Increase disk resync maximum speed to 60MB/s (from 30MB/s)
177 - new --hvm-nic-type and --hvm-disk-type flags to control the type of disk
178 exported to fully virtualized instances.
179 - provide access to the serial console of HVM instances
180 - instance auto_balance flag, set by default. If turned off it will avoid
181 warnings on cluster verify if there is not enough memory to fail over an
182 instance. in the future it will prevent automatically failing it over when
183 we will support that.
184 - batcher tool for instance creation, see /tools/README.batcher
185 - gnt-instance reinstall --select-os to interactively select a new operating
186 system when reinstalling an instance.
187 - when changing the memory amount on instance modify a check has been added
188 that the instance will be able to start. also warnings are emitted if the
189 instance will not be able to fail over, if auto_balance is true.
190 - documentation fixes
191 - sync fields between instance list/modify/add/import
192 - fix a race condition in drbd when the sync speed was set after giving the
193 device a remote peer.
196 - note: the allowed size and number of tags per object were reduced
197 - fix a bug in “gnt-cluster verify” with inconsistent volume groups
198 - fixed twisted 8.x compatibility
199 - fixed gnt-instance replace-disks with iallocator
200 - add TCP keepalives on twisted connections to detect restarted nodes
201 - disk increase support, see “gnt-instance grow-disk”
202 - implement bulk node/instance query for RAPI
203 - add tags in node/instance listing (optional)
204 - experimental migration (and live migration) support, read the man
205 page for “gnt-instance migrate”
206 - the ganeti-watcher logs are now timestamped, and the watcher also
207 has some small improvements in handling its state file
210 - Experimental readonly, REST-based remote API implementation; automatically
211 started on master node, TCP port 5080, if enabled by --enable-rapi
212 parameter to configure script.
213 - Instance allocator support. Add and import instance accept a --iallocator
214 parameter, and call that instance allocator to decide which node to use for
215 the instance. The iallocator document describes what's expected from an
217 - gnt-cluster verify N+1 memory redundancy checks: Unless passed the
218 --no-nplus1-mem option gnt-cluster verify now checks that if a node is lost
219 there is still enough memory to fail over the instances that reside on it.
220 - gnt-cluster verify hooks: it is now possible to add post-hooks to
221 gnt-cluster verify, to check for site-specific compliance. All the hooks
222 will run, and their output, if any, will be displayed. Any failing hook
223 will make the verification return an error value.
224 - gnt-cluster verify now checks that its peers are reachable on the primary
225 and secondary interfaces
226 - gnt-node add now supports the --readd option, to readd a node that is still
227 declared as part of the cluster and has failed.
228 - gnt-* list commands now accept a new -o +field way of specifying output
229 fields, that just adds the chosen fields to the default ones.
230 - gnt-backup now has a new "remove" command to delete an existing export from
232 - New per-instance parameters hvm_acpi, hvm_pae and hvm_cdrom_image_path have
233 been added. Using them you can enable/disable acpi and pae support, and
234 specify a path for a cd image to be exported to the instance. These
235 parameters as the name suggest only work on HVM clusters.
236 - When upgrading an HVM cluster to Ganeti 1.2.4, the values for ACPI and PAE
237 support will be set to the previously hardcoded values, but the (previously
238 hardcoded) path to the CDROM ISO image will be unset and if required, needs
239 to be set manually with gnt-instance modify after the upgrade..
240 - The address to which an instance's VNC console is bound is now selectable
241 per-instance, rather than being cluster wide. Of course this only applies
242 to instances controlled via VNC, so currently just applies to HVM clusters.
245 - more tweaks to the disk activation code (especially helpful for DRBD)
246 - change the default ‘gnt-instance list’ output format, now there is
247 one combined status field (see the manpage for the exact values this
249 - some more fixes for the mac export to hooks change
250 - make Ganeti not break with DRBD 8.2.x (which changed the version
251 format in /proc/drbd) (issue 24)
252 - add an upgrade tool from ‘remote_raid1’ disk template to ‘drbd’ disk
253 template, allowing migration from DRBD0.7+MD to DRBD8
256 - fix ‘gnt-instance modify’ breakage introduced in 1.2.1 with the HVM
258 - add command aliases infrastructure and a few aliases
259 - allow listing of VCPUs in the gnt-instance list and improve the man
260 pages and the --help option of gnt-node/gnt-instance list
261 - fix gnt-backup list with down nodes (issue 21)
262 - change the tools location (move from $pkgdatadir to
264 - fix the dist archive and add a check for including svn/git files in
266 - some developer-related changes: improve the burnin and the QA
267 suite, add an upload script for testing during development
270 - experimental HVM support, read the install document, section
271 "Initializing the cluster"
272 - allow for the PVM hypervisor per-instance kernel and initrd paths
273 - add a new command ‘gnt-cluster verify-disks’ which uses a new
274 algorithm to improve the reconnection of the DRBD pairs if the
275 device on the secondary node has gone away
276 - make logical volume code auto-activate LVs at disk activation time
277 - slightly improve the speed of activating disks
278 - allow specification of the MAC address at instance creation time,
279 and changing it later via ‘gnt-instance modify’
280 - fix handling of external commands that generate lots of output on
282 - update documentation with regard to minimum version of DRBD8
286 - Log the ‘xm create’ output to the node daemon log on failure (to
287 help diagnosing the error)
288 - In debug mode, log all external commands output if failed to the
290 - Change parsing of lvm commands to ignore stderr
293 - Another round of updates to the DRBD 8 code to deal with more
294 failures in the replace secondary node operation
295 - Some more logging of failures in disk operations (lvm, drbd)
296 - A few documentation updates
300 - Change configuration file format from Python's Pickle to JSON.
301 Upgrading is possible using the cfgupgrade utility.
302 - Add support for DRBD 8.0 (new disk template `drbd`) which allows for
303 faster replace disks and is more stable (DRBD 8 has many
304 improvements compared to DRBD 0.7)
305 - Added command line tags support (see man pages for gnt-instance,
306 gnt-node, gnt-cluster)
307 - Added instance rename support
308 - Added multi-instance startup/shutdown
309 - Added cluster rename support
310 - Added `gnt-node evacuate` to simplify some node operations
311 - Added instance reboot operation that can speedup reboot as compared
313 - Soften the requirement that hostnames are in FQDN format
314 - The ganeti-watcher now activates drbd pairs after secondary node
316 - Removed dependency on debian's patched fping that uses the
317 non-standard -S option
318 - Now the OS definitions are searched for in multiple, configurable
319 paths (easier for distros to package)
320 - Some changes to the hooks infrastructure (especially the new
321 post-configuration update hook)
322 - Other small bugfixes