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 #. Restart daemons on all nodes::
56 $ /etc/init.d/ganeti restart
58 #. Re-distribute configuration (master node only)::
60 $ gnt-cluster redist-conf
62 #. Restart daemons again on all nodes::
64 $ /etc/init.d/ganeti restart
66 #. Enable the watcher again (master node only)::
68 $ gnt-cluster watcher continue
70 #. Verify cluster (master node only)::
77 For going back between revisions (e.g. 2.1.1 to 2.1.0) no manual
78 intervention is required, as for upgrades.
80 Starting from version 2.8, ``cfgupgrade`` supports ``--downgrade``
81 option to bring the configuration back to the previous stable version.
82 This is useful if you upgrade Ganeti and after some time you run into
83 problems with the new version. You can downgrade the configuration
84 without losing the changes made since the upgrade. Any feature not
85 supported by the old version will be removed from the configuration, of
86 course, but you get a warning about it. If there is any new feature and
87 you haven't changed from its default value, you don't have to worry
88 about it, as it will get the same value whenever you'll upgrade again.
90 The procedure is similar to upgrading, but please notice that you have to
91 revert the configuration **before** installing the old version.
93 #. Ensure no jobs are running (master node only)::
97 #. Pause the watcher for an hour (master node only)::
99 $ gnt-cluster watcher pause 1h
101 #. Stop all daemons on all nodes::
103 $ /etc/init.d/ganeti stop
105 #. Backup old configuration (master node only)::
107 $ tar czf /var/lib/ganeti-$(date +\%FT\%T).tar.gz -C /var/lib ganeti
109 #. Run cfgupgrade on the master node::
111 $ /usr/lib/ganeti/tools/cfgupgrade --verbose --downgrade --dry-run
112 $ /usr/lib/ganeti/tools/cfgupgrade --verbose --downgrade
114 You may want to copy all the messages about features that have been
115 removed during the downgrade, in case you want to restore them when
118 #. Install the old Ganeti version on all nodes
119 #. Restart daemons on all nodes::
121 $ /etc/init.d/ganeti restart
123 #. Re-distribute configuration (master node only)::
125 $ gnt-cluster redist-conf
127 #. Restart daemons again on all nodes::
129 $ /etc/init.d/ganeti restart
131 #. Enable the watcher again (master node only)::
133 $ gnt-cluster watcher continue
135 #. Verify cluster (master node only)::
146 No changes needed except restarting the daemon; but rollback to 2.0.3 might
147 require configuration editing.
149 If you're using Xen-HVM instances, please double-check the network
150 configuration (``nic_type`` parameter) as the defaults might have changed:
151 2.0.4 adds any missing configuration items and depending on the version of the
152 software the cluster has been installed with, some new keys might have been
158 Between 2.0.1 and 2.0.2 there have been some changes in the handling of block
159 devices, which can cause some issues. 2.0.3 was then released which adds two
160 new options/commands to fix this issue.
162 If you use DRBD-type instances and see problems in instance start or
163 activate-disks with messages from DRBD about "lower device too small" or
164 similar, it is recoomended to:
166 #. Run ``gnt-instance activate-disks --ignore-size $instance`` for each
167 of the affected instances
168 #. Then run ``gnt-cluster repair-disk-sizes`` which will check that
169 instances have the correct disk sizes
176 - Ganeti 1.2.7 is currently installed
177 - All instances have been migrated from DRBD 0.7 to DRBD 8.x (i.e. no
178 ``remote_raid1`` disk template)
179 - Upgrade to Ganeti 2.0.0~rc2 or later (~rc1 and earlier don't have the needed
182 In the below steps, replace :file:`/var/lib` with ``$libdir`` if Ganeti was not
183 installed with this prefix (e.g. :file:`/usr/local/var`). Same for
186 Execution (all steps are required in the order given):
188 #. Make a backup of the current configuration, for safety::
190 $ cp -a /var/lib/ganeti /var/lib/ganeti-1.2.backup
192 #. Stop all instances::
194 $ gnt-instance stop --all
196 #. Make sure no DRBD device are in use, the following command should show no
199 $ gnt-cluster command grep cs: /proc/drbd | grep -v cs:Unconf
201 #. Stop the node daemons and rapi daemon on all nodes (note: should be logged
202 in not via the cluster name, but the master node name, as the command below
203 will remove the cluster ip from the master node)::
205 $ gnt-cluster command /etc/init.d/ganeti stop
207 #. Install the new software on all nodes, either from packaging (if available)
208 or from sources; the master daemon will not start but give error messages
209 about wrong configuration file, which is normal
210 #. Upgrade the configuration file::
212 $ /usr/lib/ganeti/tools/cfgupgrade12 -v --dry-run
213 $ /usr/lib/ganeti/tools/cfgupgrade12 -v
215 #. Make sure ``ganeti-noded`` is running on all nodes (and start it if
217 #. Start the master daemon::
221 #. Check that a simple node-list works::
225 #. Redistribute updated configuration to all nodes::
227 $ gnt-cluster redist-conf
228 $ gnt-cluster copyfile /var/lib/ganeti/known_hosts
230 #. Optional: if needed, install RAPI-specific certificates under
231 :file:`/var/lib/ganeti/rapi.pem` and run::
233 $ gnt-cluster copyfile /var/lib/ganeti/rapi.pem
235 #. Run a cluster verify, this should show no problems::
239 #. Remove some obsolete files::
241 $ gnt-cluster command rm /var/lib/ganeti/ssconf_node_pass
242 $ gnt-cluster command rm /var/lib/ganeti/ssconf_hypervisor
244 #. Update the xen pvm (if this was a pvm cluster) setting for 1.2
247 $ gnt-cluster modify -H xen-pvm:root_path=/dev/sda
249 #. Depending on your setup, you might also want to reset the initrd parameter::
251 $ gnt-cluster modify -H xen-pvm:initrd_path=/boot/initrd-2.6-xenU
253 #. Reset the instance autobalance setting to default::
255 $ for i in $(gnt-instance list -o name --no-headers); do \
256 gnt-instance modify -B auto_balance=default $i; \
259 #. Optional: start the RAPI demon::
263 #. Restart instances::
265 $ gnt-instance start --force-multiple --all
267 At this point, ``gnt-cluster verify`` should show no errors and the migration
273 1.2.4 to any other higher 1.2 version
274 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
276 No changes needed. Rollback will usually require manual edit of the
282 No changes needed. Note that going back from 1.2.4 to 1.2.3 will require manual
283 edit of the configuration file (since we added some HVM-related new
289 No changes needed. Note that the drbd7-to-8 upgrade tool does a disk format
290 change for the DRBD metadata, so in theory this might be **risky**. It is
291 advised to have (good) backups before doing the upgrade.
301 No changes needed. Only some bugfixes and new additions that don't affect
304 1.2.0 beta 3 to 1.2.0
305 ~~~~~~~~~~~~~~~~~~~~~
309 1.2.0 beta 2 to beta 3
310 ~~~~~~~~~~~~~~~~~~~~~~
312 No changes needed. A new version of the debian-etch-instance OS (0.3) has been
313 released, but upgrading it is not required.
315 1.2.0 beta 1 to beta 2
316 ~~~~~~~~~~~~~~~~~~~~~~
318 Beta 2 switched the config file format to JSON. Steps to upgrade:
320 #. Stop the daemons (``/etc/init.d/ganeti stop``) on all nodes
321 #. Disable the cron job (default is :file:`/etc/cron.d/ganeti`)
322 #. Install the new version
323 #. Make a backup copy of the config file
324 #. Upgrade the config file using the following command::
326 $ /usr/share/ganeti/cfgupgrade --verbose /var/lib/ganeti/config.data
328 #. Start the daemons and run ``gnt-cluster info``, ``gnt-node list`` and
329 ``gnt-instance list`` to check if the upgrade process finished successfully
331 The OS definition also need to be upgraded. There is a new version of the
332 debian-etch-instance OS (0.2) that goes along with beta 2.
334 .. vim: set textwidth=72 :