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