2a2d977e39
make sure we use **fsid** in command
2015-01-22 20:09:06 +08:00
30ac90dcfa
update ceph noarch repo url
2015-01-22 20:07:46 +08:00
7dd08f7bf5
Merge pull request #196 from leseb/fix-travis-2
...
Fix Travis
2015-01-21 16:37:06 +01:00
9336d80cd6
Fix Travis
...
Only do syntax check now.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-21 16:33:26 +01:00
8a7f1f7aca
Merge pull request #195 from ceph/revert-194-fix-travis
...
Revert "Fix Travis run"
2015-01-21 16:17:52 +01:00
c34d16f08f
Revert "Fix Travis run"
2015-01-21 16:17:40 +01:00
5527cf71e6
Merge pull request #194 from leseb/fix-travis
...
Fix Travis run
2015-01-21 16:16:44 +01:00
965df5579c
Fix Travis run
...
Fix travis run and effectively test the playbooks.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-21 16:15:55 +01:00
c7c729101c
Merge pull request #191 from leseb/travis
...
Add test to the playbooks
2015-01-21 10:01:24 +01:00
2c1d98f07d
Add test to the playbooks
...
Through Travis.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-21 09:59:42 +01:00
4e95686eea
Merge pull request #189 from leseb/change-controller-name
...
Change disk controller name
2015-01-19 18:25:45 +01:00
812776bce4
Change disk controller name
...
The recently added Trusty box was build with a Controller named
"SATAController", we must reuse the same name while adding our devices.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-19 18:25:14 +01:00
af7aa53ad3
Merge pull request #188 from leseb/use-trusty-box
...
Upgrade to Trusty box
2015-01-19 17:42:13 +01:00
b31a81b2b5
Upgrade to Trusty box
...
Use a Trusty box instead of Precise.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-19 17:41:19 +01:00
6eafbb871e
Merge pull request #187 from leseb/debug
...
Enable Ceph daemons debug options
2015-01-16 14:42:40 +01:00
74ffbf90fc
Enable Ceph daemons debug options
...
Add debug options for Ceph daemons.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-16 14:42:11 +01:00
e18588b147
Merge pull request #186 from leseb/swapoff
...
Ability to disable swap
2015-01-15 11:44:11 +01:00
c246d8dc80
Ability to disable swap
...
With enough memory we will probably not need any swap.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-15 11:42:51 +01:00
24d37816b7
Merge pull request #185 from leseb/os-tuning-vfs-pressure
...
Decrease vfs cache pressure
2015-01-15 11:14:52 +01:00
9b0b83a354
Decrease vfs cache pressure
...
On OSD servers we want to keep as much inodes and dentries in memory as
we can.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-15 11:08:57 +01:00
0bafee18ad
Merge pull request #184 from leseb/fix-transparent-huge-page-path
...
Fix transparent huge page sys path
2015-01-15 10:59:27 +01:00
30467a8243
Fix transparent huge page sys path
...
The old path was meant for old kernels...
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-15 10:58:45 +01:00
d24decd871
Merge pull request #183 from leseb/xfs-mount-option-sequential
...
Add more option to maximise sequential IO
2015-01-15 10:55:31 +01:00
e9360456f0
Add more option to maximise sequential IO
...
Using the largeio,inode64,swalloc options usually improves sequential
workloads.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-15 10:54:08 +01:00
8c6cba4307
Merge pull request #182 from leseb/ice-fix-epel
...
Add yum priority plugin
2015-01-13 15:50:38 +01:00
d8ab8c0bd3
Add yum priority plugin
...
Without this plugin if a Ceph version is present in a repo (let's say
epel) it will install the epel version and not the ICE version.
We install yum-plugin-priorities.noarch to honor the 'priority=1' flag.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-13 15:46:47 +01:00
8c1f885a15
Merge pull request #181 from leseb/transparent-hugepages
...
Ability to disable transparent hugepage
2015-01-13 11:13:08 +01:00
25cf4d8650
Ability to disable transparent hugepage
...
In storage world it's often recommended to disable transparent hugepages
as they will tend to lower performance.
Note that this change won't survive reboot. There are several ways to
disable this permanently such as:
* rc.local
* grub boot line
It's a bit tricky to do this in Ansible since it really depends on the
OS you're running on.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-13 11:11:11 +01:00
7430c28741
Merge pull request #180 from leseb/zone-reclaim-mode
...
Restructure OS tuning and add new value
2015-01-13 10:42:08 +01:00
2cb8941dcc
Restructure OS tuning and add new value
...
We now use a single task and loop over a set of params, it's easier to
add now params.
Setting zone_reclaim_mode to 0.
See the discussion on the Ceph ML:
http://lists.ceph.com/pipermail/ceph-users-ceph.com/2013-December/036211.html
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-13 10:34:46 +01:00
5c61ca8e8f
Merge pull request #177 from leseb/osd-sections
...
WIP: Implement OSD sections
2015-01-12 16:46:38 +01:00
78cfc3af79
Merge pull request #179 from leseb/force-vars
...
Make the monitor secret a mandatory variable.
2015-01-12 16:17:16 +01:00
289cc5423a
Make the monitor secret a mandatory variable.
...
Default behavior is to fail if a variable is not declared however this
can be disable in your ansible.cfg so we force this variable as
mandatory.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-12 14:36:15 +01:00
f68cd46664
WIP: Implement OSD sections
...
Still WIP, @mwheckmann free to test
As requested by #162
Current known issue, since ceph.conf gets modified during every single
run (at the end during the merge) so this will restart ceph daemons.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-09 11:14:20 -05:00
5047ac7669
Merge pull request #176 from leseb/os-tuning
...
Make OS tuning var configurable and file re-ordering
2015-01-07 11:46:55 +01:00
81b44fcfe2
Make OS tuning var configurable and file re-ordering
...
Depending on the OS you are runnning on you should be able to configure
these values.
Re-ordering file for clarity as well.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-07 11:44:40 +01:00
ea34a21e53
Merge pull request #175 from leseb/fix-ice
...
Fix ICE installation
2015-01-07 11:26:47 +01:00
5e6708394a
Fix ICE installation
...
Wrong ordering and variable.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-07 11:25:34 +01:00
695d6bf133
Merge pull request #174 from leseb/update-group-vars
...
Update group_vars files to match roles defaults vars
2015-01-06 22:50:12 +01:00
12dc4fbb09
Update group_vars files to match roles defaults vars
...
Updating each group_vars file to reflect the content of each role
default variables.
Closes : #169
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-06 22:47:49 +01:00
e72c8c5ac0
Merge pull request #173 from leseb/fix-vagrant-1.8-deployment
...
Fix vagrant deployment for vagrant version > 1.7
2015-01-06 16:25:10 +01:00
84271325d7
Fix vagrant deployment for vagrant version > 1.7
...
See: https://github.com/mitchellh/vagrant/issues/5048 .
Closes : #172
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-06 16:23:43 +01:00
1ac45b966b
Merge pull request #171 from leseb/file-max
...
Bump up maximum number of open files
2015-01-06 15:30:33 +01:00
5c66801dc7
Bump up maximum number of open files
...
Large cluster will easily reach the default value so we increase it.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-06 15:26:44 +01:00
001a82f023
Merge pull request #170 from leseb/max-open-files
...
Bump up max open file for big cluster
2015-01-06 14:35:51 +01:00
79f6415e9d
Bump up max open file for big cluster
...
Big cluster will easily reach the default limit so we need to increase
it and make it configurable.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-06 14:23:23 +01:00
75c4739679
Revert "Remove the fsid vars from mon and osd roles"
...
This reverts commit 216c8e24c8
2015-01-06 00:42:54 +01:00
10e01c6161
Merge pull request #167 from leseb/fix-vagrant-fsid
...
Fix the vagrant deployment
2015-01-06 00:24:29 +01:00
474eee8fad
Fix the vagrant deployment
...
Add a fetch/ceph_cluster_uuid.conf file so we keep the same UUID from
the Vagrantfile and from the ceph-common run.
Prior to that change the Vagrantfile was setting
4a158d27-f750-41d5-9e7f-26ce4c9d2d45 but while playing the ceph-common
role we check if fetch/ceph_cluster_uuid.conf exists, if not we generate
an UUID. Here we ended up with 2 UUIDs...
IMPORTANT NOTE FOR NON-VAGRANT DEPLOYMENT. IF YOU WANT TO USE YOUR OWN
UUID PLEASE REMOVE THAT FILE BEFORE RUNNING THE PLAYBOOK.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
2015-01-06 00:21:16 +01:00
cfe4178bff
Merge pull request #166 from leseb/remove-unsed-vars
...
Remove the fsid vars from mon and osd roles
2015-01-06 00:19:10 +01:00
Yin Jifeng
Leseb