Commit Graph

3384 Commits (d7dadc3e7b9d2e218d85784df72e4cd008ecb1ee)
 

Author SHA1 Message Date
Konstantin Shalygin d7dadc3e7b ceph-osd: respect nvme partitions when device is a disk. 2017-12-12 09:03:18 +01:00
Guillaume Abrioux 6a9b5c9632 defaults: fix CI issue with ceph_uid fact
The CI complains because of `ceph_uid` fact which doesn't exist since
the docker image tag used in the CI doesn't match with this condition.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-12-12 09:02:37 +01:00
Andrew Schoen 788c3f351a ceph-osd: adds osd_objectstore to the name when using the ceph_volume module
This allows for easier debugging if verbosity is not set high enough.

Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2017-12-11 09:58:06 -06:00
Andrew Schoen 15db3323a5 tests for the ceph_volume module
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2017-12-11 09:58:06 -06:00
Andrew Schoen 4b4680d1e6 refactor ceph_volume.py so it's easier to test
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2017-12-11 09:58:06 -06:00
Andrew Schoen 5e3d8dbf63 ceph-osd: use the cluster param with the ceph_volume module
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2017-12-11 09:58:06 -06:00
Andrew Schoen 7721aa375e ceph_volume: adds the cluster param
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2017-12-11 09:58:06 -06:00
Andrew Schoen 423166f671 ceph-osd: use the new ceph_volume module for the lvm scenario
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2017-12-11 09:58:06 -06:00
Andrew Schoen 61b4a1bedc adds a ceph_volume module
This module uses ceph-volume to create OSDs. Currently
it only supports the 'lvm' subcommand and 'create'.

Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2017-12-11 09:58:06 -06:00
Prisacari Dmitrii a9981b4131 Vagrantfile: Fixed repeated OSD controller creation 2017-12-11 11:24:55 +01:00
Sébastien Han 0ea1811f6f
Merge pull request #2226 from andymcc/gpt_mklabel
Skip mklabel gpt if already gpt
2017-12-11 03:12:46 -06:00
Andy McCrae 4f1e854c79 Use parted module instead of command 2017-12-11 17:33:40 +10:00
Guillaume Abrioux e8338c87b0
Merge pull request #2211 from fultonj/admin-key-perms
Set tighter permissions on keyrings when containerized
2017-12-07 08:43:02 +01:00
John Fulton ffae294288 Set tighter permissions on keyrings when containerized
During a containerized deployment, set the permissions
of ceph.client.admin.keyring and other keyrings to
chmod 600 and chown it to ceph.
2017-12-06 19:22:28 -05:00
Sébastien Han 29d21303d7
Merge pull request #2221 from ceph/fix_purge_waitfor
purge: fix bug on 'wait_for' task
2017-11-29 14:56:54 +01:00
Guillaume Abrioux aaaf980140 purge: fix bug on 'wait_for' task
this task hangs because `{{ inventory_hostname }}` doesn't resolv to an
actual ip address.
Using `hostvars[inventory_hostname]['ansible_default_ipv4']['address']`
should fix this because it will reach the node with its actual IP
address.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-11-29 11:10:56 +01:00
Guillaume Abrioux b449b16edd
Merge pull request #2215 from squidboylan/support_loopback_devices
Add support for using loopback devices as OSDs
2017-11-28 14:04:47 +01:00
Sébastien Han f94b9040eb
Merge pull request #2214 from ceph/bz-1510555
handlers: restart daemons only if docker is running
2017-11-28 12:22:50 +01:00
Sébastien Han ef581f807d
Merge pull request #2202 from ceph/remove_leftover
osd: remove leftover and fix a typo
2017-11-28 12:21:13 +01:00
Guillaume Abrioux 7852fe2115
Merge pull request #2212 from wintamute/master
Group_vars mon sample - replaced hardcoded pool names for Openstack
2017-11-28 12:14:09 +01:00
wintamute ebe0e60235 Openstack: replaced hardcoded pool names with variables for openstack (nova) user
(cherry picked from commit 2bf48f1)
2017-11-28 09:06:51 +01:00
Caleb Boylan 8f02bb007f Add support for using loopback devices as OSDs
This is particularly useful in CI environments where you dont have
the option of adding extra devices or volumes to the host. It is also
a simple change to support loopback devices
2017-11-27 16:02:36 -08:00
Guillaume Abrioux b26a840002 handlers: restart daemons only if docker is running
In case where docker CLI is available but docker is not running, we
don't want to trigger the restart of the daemons.

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1510555

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-11-27 14:59:30 +01:00
Sébastien Han d9cfe5f6df
Merge pull request #2177 from jprovaznik/rados
Allow to use rados for ganesha exports
2017-11-23 10:36:58 +01:00
Sébastien Han d5399e8d02
Merge pull request #2207 from ceph/mds-debian
common: install ceph-common on all the machines
2017-11-22 17:45:29 +01:00
Sébastien Han bb7b29a9fc common: install ceph-common on all the machines
Since some daemons now install their own packages the task checking the
ceph version fails on Debian systems. So the 'ceph-common' package must
be installed on all the machines.

Signed-off-by: Sébastien Han <seb@redhat.com>
2017-11-22 17:11:50 +01:00
Jan Provaznik 2435c48cd5 Allow to use rados for ganesha exports 2017-11-21 15:21:32 +01:00
Guillaume Abrioux 3af03f0f3e
Merge pull request #2185 from ceph/fix_purge-cluster
purge-cluster: remove usage of `with_fileglob`
2017-11-21 13:30:15 +01:00
Guillaume Abrioux 1cba626484 osd: remove leftover and fix a typo
This task was originally needed to fix a docker installation issue
(see: #1030). This has been fixed, therefore it can be removed.

Fixes: #2199

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-11-21 11:11:34 +01:00
Guillaume Abrioux 947766e294 purge-cluster: remove usage of `with_fileglob`
`with_fileglob` loops over files on the machine where ansible-playbook
is being run.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-11-21 08:24:11 +01:00
Sébastien Han 387fd49086
Merge pull request #2197 from ceph/fix_gpt_prepare
osd: ensure a gpt label is set on device
2017-11-20 10:40:18 +01:00
Guillaume Abrioux efe06be10f osd: ensure a gpt label is set on device
ceph-disk prepare will fail on jewel if a GPT label is not present on
device.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-11-17 17:32:23 +01:00
Guillaume Abrioux 3c6f2854fe
Merge pull request #2189 from fultonj/empty-acl
Make openstack_keys param support no acls list
2017-11-16 19:39:01 +01:00
Guillaume Abrioux 3d4d71ade0
Merge pull request #2184 from ceph/fix_wildcard_remove
purge-docker: remove osd disk prepare logs
2017-11-16 18:02:35 +01:00
John Fulton d73f751b63 Make openstack_keys param support no acls list
A recent change [1] required that the openstack_keys
param always containe an acls list. However, it's
possible it might not contain that list. Thus, this
param sets a default for that list to be empty if it
is not in the structure as defined by the user.

[1] d65cbaa539
2017-11-16 11:29:59 -05:00
Sébastien Han f31d8557dd
Merge pull request #2182 from ceph/fix_reboot_rbd
rbd: enable ceph-rbd-mirror.target on releases prior to luminous
2017-11-16 16:55:39 +01:00
Guillaume Abrioux b295e0a11f
Merge pull request #2186 from ceph/dmcrypt-fixfg
osd: multiple fixes
2017-11-16 16:52:13 +01:00
Sébastien Han 932345ab2a osd: remove leftover from osd partition
We used to support osds that are a partition. This is long gone so
removing this task.

Signed-off-by: Sébastien Han <seb@redhat.com>
2017-11-16 14:58:40 +01:00
Sébastien Han b1c1322357 osd: remove failed_when on activation
There is no need to continue if the activation fails.

Signed-off-by: Sébastien Han <seb@redhat.com>
2017-11-16 14:57:49 +01:00
Sébastien Han 80d3a242d0 osd: fix bad activation for dmcrypt
We were activating dmcrypt devices with the wrong command. Basically the
first task execute the wrong activate command. The task fails but
continues because of the 'failed_when: false'. Then the right activation
sequence is being done by the next task.

Signed-off-by: Sébastien Han <seb@redhat.com>
2017-11-16 14:55:08 +01:00
Sébastien Han cc264d6ba6
Merge pull request #2151 from hwoarang/add-opensuse
Add openSUSE Leap 42.3 support
2017-11-16 14:35:28 +01:00
Guillaume Abrioux d9c1b61092 purge-docker: remove osd disk prepare logs
`with_fileglob` loops over files on the machine that runs the playbook.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-11-16 14:27:36 +01:00
Sébastien Han a98f14784a
Merge pull request #2172 from ceph/lvm-raw-device
lvm: add support for --data to be a raw device or partition
2017-11-16 14:14:23 +01:00
Guillaume Abrioux ccad0ebf26 rbd: enable ceph-rbd-mirror.target for releases <= luminous
when `ceph-rbd-mirror.target` is not enabled, the service won't start
after a reboot because there is a dependency between these two units.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-11-16 14:12:59 +01:00
Guillaume Abrioux 17743f3de7
Merge pull request #2181 from yanyixing/fix_nfs_pkg
[skip ci] fix: remove the duplicated code
2017-11-16 10:29:03 +01:00
Yixing Yan 097249371f fix: remove the duplicated code 2017-11-16 16:45:03 +08:00
Guillaume Abrioux 28a158e7a2
Merge pull request #2174 from fultonj/chmod-facl-on-all-mon
Set permissions and ACLs of OpenStack keys on all ceph-mons
2017-11-15 17:17:23 +01:00
Andrew Schoen 6dc090a4c8 update group_vars/osds.yml.sample for the new lvm_volumes examples
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2017-11-15 09:36:17 -06:00
Andrew Schoen 3c604f1115 lvm: support --data as a raw device or partition in ceph-volume
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2017-11-15 09:36:17 -06:00
Andrew Schoen 04f02910a9 lvm: ensure the data_vg exists before using it
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2017-11-15 09:36:17 -06:00