4 .. highlight:: shell-example
6 This document details the steps needed to upgrade a cluster to newer versions
9 As a general rule the node daemons need to be restarted after each software
10 upgrade; if using the provided example init.d script, this means running the
11 following command on all nodes::
13 $ /etc/init.d/ganeti restart
19 Starting with Ganeti 2.0, upgrades between revisions (e.g. 2.1.0 to 2.1.1)
20 should not need manual intervention. As a safety measure, minor releases (e.g.
21 2.1.3 to 2.2.0) require the ``cfgupgrade`` command for changing the
22 configuration version. Below you find the steps necessary to upgrade between
25 To run commands on all nodes, the `distributed shell (dsh)
26 <http://www.netfort.gr.jp/~dancer/software/dsh.html.en>`_ can be used, e.g.
27 ``dsh -M -F 8 -f /var/lib/ganeti/ssconf_online_nodes gnt-cluster --version``.
29 #. Ensure no jobs are running (master node only)::
33 #. Pause the watcher for an hour (master node only)::
35 $ gnt-cluster watcher pause 1h
37 #. Stop all daemons on all nodes::
39 $ /etc/init.d/ganeti stop
41 #. Backup old configuration (master node only)::
43 $ tar czf /var/lib/ganeti-$(date +\%FT\%T).tar.gz -C /var/lib ganeti
45 #. Install new Ganeti version on all nodes
46 #. Run cfgupgrade on the master node::
48 $ /usr/lib/ganeti/tools/cfgupgrade --verbose --dry-run
49 $ /usr/lib/ganeti/tools/cfgupgrade --verbose
51 (``cfgupgrade`` supports a number of parameters, run it with
52 ``--help`` for more information)
54 #. Upgrade the directory permissions on all nodes::
56 $ /usr/lib/ganeti/ensure-dirs --full-run
58 #. Restart daemons on all nodes::
60 $ /etc/init.d/ganeti restart
62 #. Re-distribute configuration (master node only)::
64 $ gnt-cluster redist-conf
66 #. If you use file storage, check that the ``/etc/ganeti/file-storage-paths``
67 #. is correct on all nodes. For security reasons it's not copied
68 #. automatically, but it can be copied manually via::
70 $ gnt-cluster copyfile /etc/ganeti/file-storage-paths
72 #. Restart daemons again on all nodes::
74 $ /etc/init.d/ganeti restart
76 #. Enable the watcher again (master node only)::
78 $ gnt-cluster watcher continue
80 #. Verify cluster (master node only)::
87 For going back between revisions (e.g. 2.1.1 to 2.1.0) no manual
88 intervention is required, as for upgrades.
90 Starting from version 2.8, ``cfgupgrade`` supports ``--downgrade``
91 option to bring the configuration back to the previous stable version.
92 This is useful if you upgrade Ganeti and after some time you run into
93 problems with the new version. You can downgrade the configuration
94 without losing the changes made since the upgrade. Any feature not
95 supported by the old version will be removed from the configuration, of
96 course, but you get a warning about it. If there is any new feature and
97 you haven't changed from its default value, you don't have to worry
98 about it, as it will get the same value whenever you'll upgrade again.
100 The procedure is similar to upgrading, but please notice that you have to
101 revert the configuration **before** installing the old version.
103 #. Ensure no jobs are running (master node only)::
107 #. Pause the watcher for an hour (master node only)::
109 $ gnt-cluster watcher pause 1h
111 #. Stop all daemons on all nodes::
113 $ /etc/init.d/ganeti stop
115 #. Backup old configuration (master node only)::
117 $ tar czf /var/lib/ganeti-$(date +\%FT\%T).tar.gz -C /var/lib ganeti
119 #. Run cfgupgrade on the master node::
121 $ /usr/lib/ganeti/tools/cfgupgrade --verbose --downgrade --dry-run
122 $ /usr/lib/ganeti/tools/cfgupgrade --verbose --downgrade
124 You may want to copy all the messages about features that have been
125 removed during the downgrade, in case you want to restore them when
128 #. Install the old Ganeti version on all nodes
129 #. Restart daemons on all nodes::
131 $ /etc/init.d/ganeti restart
133 #. Re-distribute configuration (master node only)::
135 $ gnt-cluster redist-conf
137 #. Restart daemons again on all nodes::
139 $ /etc/init.d/ganeti restart
141 #. Enable the watcher again (master node only)::
143 $ gnt-cluster watcher continue
145 #. Verify cluster (master node only)::
156 No changes needed except restarting the daemon; but rollback to 2.0.3 might
157 require configuration editing.
159 If you're using Xen-HVM instances, please double-check the network
160 configuration (``nic_type`` parameter) as the defaults might have changed:
161 2.0.4 adds any missing configuration items and depending on the version of the
162 software the cluster has been installed with, some new keys might have been
168 Between 2.0.1 and 2.0.2 there have been some changes in the handling of block
169 devices, which can cause some issues. 2.0.3 was then released which adds two
170 new options/commands to fix this issue.
172 If you use DRBD-type instances and see problems in instance start or
173 activate-disks with messages from DRBD about "lower device too small" or
174 similar, it is recoomended to:
176 #. Run ``gnt-instance activate-disks --ignore-size $instance`` for each
177 of the affected instances
178 #. Then run ``gnt-cluster repair-disk-sizes`` which will check that
179 instances have the correct disk sizes
186 - Ganeti 1.2.7 is currently installed
187 - All instances have been migrated from DRBD 0.7 to DRBD 8.x (i.e. no
188 ``remote_raid1`` disk template)
189 - Upgrade to Ganeti 2.0.0~rc2 or later (~rc1 and earlier don't have the needed
192 In the below steps, replace :file:`/var/lib` with ``$libdir`` if Ganeti was not
193 installed with this prefix (e.g. :file:`/usr/local/var`). Same for
196 Execution (all steps are required in the order given):
198 #. Make a backup of the current configuration, for safety::
200 $ cp -a /var/lib/ganeti /var/lib/ganeti-1.2.backup
202 #. Stop all instances::
204 $ gnt-instance stop --all
206 #. Make sure no DRBD device are in use, the following command should show no
209 $ gnt-cluster command grep cs: /proc/drbd | grep -v cs:Unconf
211 #. Stop the node daemons and rapi daemon on all nodes (note: should be logged
212 in not via the cluster name, but the master node name, as the command below
213 will remove the cluster ip from the master node)::
215 $ gnt-cluster command /etc/init.d/ganeti stop
217 #. Install the new software on all nodes, either from packaging (if available)
218 or from sources; the master daemon will not start but give error messages
219 about wrong configuration file, which is normal
220 #. Upgrade the configuration file::
222 $ /usr/lib/ganeti/tools/cfgupgrade12 -v --dry-run
223 $ /usr/lib/ganeti/tools/cfgupgrade12 -v
225 #. Make sure ``ganeti-noded`` is running on all nodes (and start it if
227 #. Start the master daemon::
231 #. Check that a simple node-list works::
235 #. Redistribute updated configuration to all nodes::
237 $ gnt-cluster redist-conf
238 $ gnt-cluster copyfile /var/lib/ganeti/known_hosts
240 #. Optional: if needed, install RAPI-specific certificates under
241 :file:`/var/lib/ganeti/rapi.pem` and run::
243 $ gnt-cluster copyfile /var/lib/ganeti/rapi.pem
245 #. Run a cluster verify, this should show no problems::
249 #. Remove some obsolete files::
251 $ gnt-cluster command rm /var/lib/ganeti/ssconf_node_pass
252 $ gnt-cluster command rm /var/lib/ganeti/ssconf_hypervisor
254 #. Update the xen pvm (if this was a pvm cluster) setting for 1.2
257 $ gnt-cluster modify -H xen-pvm:root_path=/dev/sda
259 #. Depending on your setup, you might also want to reset the initrd parameter::
261 $ gnt-cluster modify -H xen-pvm:initrd_path=/boot/initrd-2.6-xenU
263 #. Reset the instance autobalance setting to default::
265 $ for i in $(gnt-instance list -o name --no-headers); do \
266 gnt-instance modify -B auto_balance=default $i; \
269 #. Optional: start the RAPI demon::
273 #. Restart instances::
275 $ gnt-instance start --force-multiple --all
277 At this point, ``gnt-cluster verify`` should show no errors and the migration
283 1.2.4 to any other higher 1.2 version
284 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
286 No changes needed. Rollback will usually require manual edit of the
292 No changes needed. Note that going back from 1.2.4 to 1.2.3 will require manual
293 edit of the configuration file (since we added some HVM-related new
299 No changes needed. Note that the drbd7-to-8 upgrade tool does a disk format
300 change for the DRBD metadata, so in theory this might be **risky**. It is
301 advised to have (good) backups before doing the upgrade.
311 No changes needed. Only some bugfixes and new additions that don't affect
314 1.2.0 beta 3 to 1.2.0
315 ~~~~~~~~~~~~~~~~~~~~~
319 1.2.0 beta 2 to beta 3
320 ~~~~~~~~~~~~~~~~~~~~~~
322 No changes needed. A new version of the debian-etch-instance OS (0.3) has been
323 released, but upgrading it is not required.
325 1.2.0 beta 1 to beta 2
326 ~~~~~~~~~~~~~~~~~~~~~~
328 Beta 2 switched the config file format to JSON. Steps to upgrade:
330 #. Stop the daemons (``/etc/init.d/ganeti stop``) on all nodes
331 #. Disable the cron job (default is :file:`/etc/cron.d/ganeti`)
332 #. Install the new version
333 #. Make a backup copy of the config file
334 #. Upgrade the config file using the following command::
336 $ /usr/share/ganeti/cfgupgrade --verbose /var/lib/ganeti/config.data
338 #. Start the daemons and run ``gnt-cluster info``, ``gnt-node list`` and
339 ``gnt-instance list`` to check if the upgrade process finished successfully
341 The OS definition also need to be upgraded. There is a new version of the
342 debian-etch-instance OS (0.2) that goes along with beta 2.
344 .. vim: set textwidth=72 :