24b82c2679
purge: fix journal purge
...
Using a condition when osd_scenario == 'non-collocated' was wrong since
these partitions can be collocated on a single device also. Removing the
check makes the purge of these partitions.
Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1499871
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-10 09:57:39 +02:00
f147b119ed
Merge pull request #2014 from ceph/fixes-2
...
infra: use the pg check in the right place
2017-10-09 20:14:06 +02:00
9e8204d9e8
nfs: move packages installation to own role
...
Make role `ceph-nfs` handling itself the installation of nfs
packages.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-10-09 19:10:15 +02:00
3c64abe07d
mds: move installation packages in role itself
...
Make role `ceph-mds` handling itself the installation of `ceph-mds`
package.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-10-09 17:25:46 +02:00
4032f102fe
iscsi: move package install to ceph-iscsi-role
...
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-09 17:25:46 +02:00
1581a1c078
mgr: move installation packages in role itself
...
Make role `ceph-mgr` handling itself the installation of `ceph-mgr`
package because it's complicated to manage it regarding we are going to
install `jewel vs. luminous`
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-10-09 17:25:45 +02:00
a1ea6e7f59
tests: adapt current testing for collocation scenario
...
Since we introduced collocation testing scenario, we need to adapt
current tests to this new scenario.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-10-09 17:25:45 +02:00
88a37ca990
ci: disable vnc console
...
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-09 17:25:45 +02:00
bf99751ce1
osd: bindmount /run/udev
...
Ensures that "udevadm" is able to check the status of udev's event queue.
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-09 17:25:45 +02:00
6d7b73fa91
ci: re-add osd_pool_default_size to 1 with the override
...
If we don't do this the client will create pools with a replica 3 since
osd_pool_default_size was gone in ceph-override.json. This was making
switch_to_containers failing.
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-09 17:25:45 +02:00
e15302c284
tox: decrease sleep to 2min instead of 5.
...
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-09 17:25:44 +02:00
450108fab9
infra: add independant purge-iscsi-gateways.yml
...
The current inclusion of purge-iscsi-gateways.yml in purge-cluster.yml
is not working well and blocking the CI too. So removing it from
purge-cluster.yml and re-add the original purge-iscsi-gateways.yml.
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-09 17:25:44 +02:00
1bd891232c
config: do not duplicate sections when doing collocation
...
Prior to this commit, when collocating a RGW and NFS on the same box the
ceph.conf layout was the following:
[client.rgw.rgw0]
host = mds0
host = rgw0
rgw frontends = civetweb port=192.168.15.50:8080
num_threads=100[client.rgw.mds0]
rgw frontends = civetweb port=192.168.15.70:8080 num_threads=100
rgw frontends = civetweb port=192.168.15.50:8080 num_threads=100
keyring = /var/lib/ceph/radosgw/test-rgw.mds0/keyring
keyring = /var/lib/ceph/radosgw/test-rgw.rgw0/keyring
rgw data = /var/lib/ceph/radosgw/test-rgw.rgw0
log file = /var/log/ceph/test-rgw-mds0.log
log file = /var/log/ceph/test-rgw-rgw0.log
[mds.mds0]
host = mds0
[global]
rgw override bucket index max shards = 16
fsid = 70e1d368-57b3-4978-b746-cbffce6e56b5
rgw bucket default quota max objects = 1638400
osd_pool_default_size = 1
public network = 192.168.15.0/24
mon host = 192.168.15.10,192.168.15.11,192.168.15.12
osd_pool_default_pg_num = 8
cluster network = 192.168.16.0/24
[mds.rgw0]
host = rgw0
[client.rgw.mds0]
host = mds0
rgw data = /var/lib/ceph/radosgw/test-rgw.mds0
keyring = /var/lib/ceph/radosgw/test-rgw.mds0/keyring
rgw frontends = civetweb port=192.168.15.70:8080 num_threads=100
log file = /var/log/ceph/test-rgw-mds0.log
Basically appending all the sections. This commits solves that.
Now the sections appear like this:
-bash-4.2# cat /etc/ceph/test.conf
[client.rgw.rgw0]
log file = /var/log/ceph/test-rgw-rgw0.log
host = rgw0
keyring = /var/lib/ceph/radosgw/test-rgw.rgw0/keyring
rgw frontends = civetweb port=192.168.15.50:8080 num_threads=100
[client.rgw.mds0]
log file = /var/log/ceph/test-rgw-mds0.log
host = mds0
keyring = /var/lib/ceph/radosgw/test-rgw.mds0/keyring
rgw frontends = civetweb port=192.168.15.70:8080 num_threads=100
[global]
cluster network = 192.168.16.0/24
mon host = 192.168.15.10,192.168.15.11,192.168.15.12
osd_pool_default_size = 1
public network = 192.168.15.0/24
rgw bucket default quota max objects = 1638400
osd_pool_default_pg_num = 8
rgw override bucket index max shards = 16
fsid = 77a21980-3033-4174-9264-1abc7185bcb3
[mds.rgw0]
host = rgw0
[mds.mds0]
host = mds0
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-09 17:25:44 +02:00
774697ebd8
infra: use the pg check in the right place
...
Use the pg check before doing the pg check, not on the quorum check.
Also never quote int when doing comparaison.
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-09 17:25:41 +02:00
a3e7bcb13f
Merge pull request #2013 from ceph/wip-purge-cluster
...
A couple of purge cluster fixes
2017-10-09 17:18:30 +02:00
88cb568396
Merge pull request #2011 from ceph/switch-pg
...
[skip ci] switch: check pgs only when num_pgs > 0
2017-10-07 03:42:44 +02:00
33a3aa0dda
switch: check pgs only when num_pgs > 0
...
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-07 03:42:09 +02:00
7054abef99
Merge pull request #2009 from ceph/fix-clean-pg
...
[skip ci] handler: do not test if pgs_num = 0
2017-10-07 03:39:26 +02:00
9f1bd3d6dd
handler: add serial restart back
...
We now restart daemons on each machine in a serialized fashion.
Closes: https://github.com/ceph/ceph-ansible/issues/1989
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-07 03:39:10 +02:00
abb8c374cf
ci: use by-id instead of by-path
...
by-id relies on the disk WWID which is more reliable then by-path
(pointing to the PCI info)
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-07 03:39:09 +02:00
05f26031ea
rolling_update: perform pg check when pgs_num > 0
...
If num_pgs = 0 the check will never return 0.
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-07 03:39:09 +02:00
a4dcef73d4
common: fix debian rhcs installation
...
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-07 03:39:09 +02:00
c3c63ae539
switch: rework and fix clean pg wait
...
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-07 03:39:09 +02:00
c693e95cbf
purge-docker: rework device detection
...
we don't need "devices" and other device variable anymore, the playbook
detects that for us.
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-07 03:39:04 +02:00
2fb4981ca9
shrink-osd: admin key not needed for container shrink
...
Also do some clean
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-07 00:20:43 +02:00
64e272d818
purge-cluster: Do not use shell for rm
...
The shell wildcard expansion of non-existing paths fails on zsh making
the whole script fail. We can use file module with with_fileglob to
alleviate the problem instead.
Signed-off-by: Boris Ranto <branto@redhat.com>
2017-10-06 22:54:37 +02:00
f696cb7637
purge-cluster: Do not fail on systemd commands
...
The systemd can't stop services if the unit files were removed before
the cluster was purged. We should just ignore these.
Signed-off-by: Boris Ranto <branto@redhat.com>
2017-10-06 22:52:56 +02:00
ac29e8f977
Merge pull request #1983 from jprovaznik/suffix
...
Allow to override systemd service instance id
2017-10-06 22:40:57 +02:00
5d39f378da
Merge pull request #1984 from jprovaznik/exportdir
...
Include exports dir in ceph-nfs config file
2017-10-06 22:38:13 +02:00
9a46ed8b76
Merge pull request #2010 from ceph/wip-nfs-set-rgw-key-permissions
...
nfs: missing conditional for setting rgw key permissions
2017-10-06 21:47:49 +02:00
28862a99d9
nfs: missing conditional for setting rgw key permissions
...
Signed-off-by: Ali Maredia <amaredia@redhat.com>
2017-10-06 15:21:35 -04:00
11f51df1fc
Merge pull request #2005 from ceph/wip-nfs-export-id
...
nfs: config var changes
2017-10-06 17:05:21 +02:00
779f642fa8
use get to check stdout_lines
...
During the initial play, the docker command doesn't not exist and then
there is no stdout_lines to the command. So get allows us to fix this by
declaring an array if the command fails.
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-06 16:57:46 +02:00
d5ae0a3340
handler: do not test if pgs_num = 0
...
We don't need to wait if they are no PGS.
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-06 16:57:46 +02:00
e3ba5e2f9c
Merge pull request #2006 from ceph/fix_jewel_nightly
...
Fix jewel nightly
2017-10-06 16:06:23 +02:00
6b027557e6
osd: fix `set_fact build dedicated_devices`
...
Use an intermediate variable to build the final `dedicated_devices` list
to avoid duplicate entry in that array. (We need a 1:1 relation between
`dedicated_devices` and `devices` since we are using a `with_together`
later.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-10-06 15:00:32 +02:00
680ec8758e
tests: skip tests for nfs nodes when release is jewel
...
nfs nodes are not deployed on jewel so we should skip the tests on them.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-10-06 12:49:39 +02:00
d363b0f741
rbd: fix bug when trying to fetch key
...
With jewel, `bootstrap_rbd_keyring` is not set because of this condition:
```
when:
- ceph_release_num.{{ ceph_release }} >= ceph_release_num.luminous
```
Therefore, the task `try to fetch ceph config and keys` will fail.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2017-10-06 11:34:29 +02:00
3c16af5ef2
Allow to override systemd service instance id
...
It's useful to have constant service instance id when ceph-nfs
is managed by pacemaker.
2017-10-06 08:20:37 +02:00
0c09cd3e2e
nfs: config var changes
...
- remove unused ganesha config vars,
- set different default Export_ids for each FSAL
Signed-off-by: Ali Maredia <amaredia@redhat.com>
2017-10-05 16:51:23 -04:00
1121a840ef
Merge pull request #2003 from ceph/debian-iso
...
[skip ci] common: iso install on Debian is supported by rhcs
2017-10-05 18:57:47 +02:00
feaf5ff9c6
common: iso install on Debian is supported by rhcs
...
Also adds support for RCSH installation on Debian.
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-05 18:57:09 +02:00
a155c1984f
Merge pull request #2004 from ceph/debian-url
...
common: fix ga verison for debian rhcs
2017-10-05 18:46:13 +02:00
425ecb3c7d
common: fix ga verison for debian rhcs
...
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-05 18:45:30 +02:00
639389b9cd
Merge pull request #1985 from ceph/debian-rhcs
...
[skip ci] common: fix rhcs installation on debian
2017-10-05 18:42:46 +02:00
0d833657c1
Merge pull request #2001 from ceph/iscsi
...
iscsi: fix wrong group name for iscsi
2017-10-05 18:29:06 +02:00
530bf5d134
Merge pull request #2002 from ceph/non-col
...
[skip ci] osd: do not do unique on dedicated_devices
2017-10-05 18:25:16 +02:00
29888649e5
osd: do not do unique on dedicated_devices
...
This is needed later, if we do unique, only the first OSD will get a
journal.
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-05 18:20:18 +02:00
9193e88878
common: fix rhcs installation on debian
...
* Change version from 2 to 3.
* use ceph_rhcs_cdn_debian_repo_version to use other repositories along
* with ceph_rhcs_cdn_debian_repo
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-05 17:42:21 +02:00
b6b24a5ca9
iscsi: fix wrong group name for iscsi
...
Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1498490
Signed-off-by: Sébastien Han <seb@redhat.com>
2017-10-05 17:25:32 +02:00
Sébastien Han
Guillaume Abrioux
Boris Ranto
Ali Maredia
Jan Provaznik