]>
git.apps.os.sepia.ceph.com Git - ceph-ansible.git/log
Leseb [Mon, 12 May 2014 13:42:28 +0000 (06:42 -0700)]
Merge pull request #82 from lorin/master
Use Vagrant's Ansible provisioner
Lorin Hochstein [Sun, 11 May 2014 01:52:26 +0000 (21:52 -0400)]
Use Vagrant's Ansible provisioner
Use Vagrant's built-in support for Ansible provisioner. This eliminates the need
for a hosts file, and simplifies the ansible config file.
Renames config from .ansible.cfg to ansible.cfg since Ansible expects the file
to be called ansible.cfg and be adjacent to Vagrantfile when using the Vagrant
provisioner.
Leseb [Wed, 30 Apr 2014 23:27:20 +0000 (01:27 +0200)]
Merge pull request #81 from leseb/enh-testing-mode
Enable single host mode
Sébastien Han [Wed, 30 Apr 2014 23:24:20 +0000 (01:24 +0200)]
Enable single host mode
This commit introduces a new config option 'osd crush chooseleaf type'.
With the help of this option and by setting it to '0' we tell Ceph to
store all the replicas on a single host. Basically we tell CRUSH to
iterate over disk and not over host.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Wed, 30 Apr 2014 15:58:51 +0000 (17:58 +0200)]
Merge pull request #80 from leseb/change-default-disk
Change the default disk numbering
Sébastien Han [Wed, 30 Apr 2014 15:57:10 +0000 (17:57 +0200)]
Change the default disk numbering
While trying to auto-provision with vagrant, new disks get /dev/sdb and
so forth. So starting from /dev/sdd doesn't make sense.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Wed, 30 Apr 2014 12:58:02 +0000 (14:58 +0200)]
Merge pull request #79 from leseb/fix-rgw-email
Fix non-declared rgw email_address variable
Sébastien Han [Wed, 30 Apr 2014 12:56:37 +0000 (14:56 +0200)]
Fix non-declared rgw email_address variable
Fix: #77
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Wed, 30 Apr 2014 11:42:08 +0000 (13:42 +0200)]
Merge pull request #76 from AcalephStorage/fetch-fsid
Removed dependency on mon[0]
Dexter John Genterone [Wed, 30 Apr 2014 08:02:07 +0000 (16:02 +0800)]
keys are now copied from the first available mon host
since we're now using fsid for the directory name, it should be safe to
just copy the keys from all mon hosts. Once they are copied, the rest of
the hosts will just skip copying. :)
Hunter Nield [Wed, 30 Apr 2014 07:16:11 +0000 (15:16 +0800)]
Avoid ip/fqdn weirdness & use fsid path for keys
Leseb [Fri, 18 Apr 2014 12:01:47 +0000 (14:01 +0200)]
Merge pull request #74 from guestisp/fix-73-use-ansible-fqdn-in-purge-yml
Fix #73 by using `ansible_fqdn`
Leseb [Fri, 18 Apr 2014 12:01:36 +0000 (14:01 +0200)]
Merge pull request #72 from guestisp/fix-71-typo-in-purge-yml
Fix #71. Removed wrong `---`
Leseb [Fri, 18 Apr 2014 12:01:22 +0000 (14:01 +0200)]
Merge pull request #70 from guestisp/fix-69-mon-osd-min-down-reporters
Fix #69 - mon osd min down reporters
Alessandro Corbelli [Fri, 18 Apr 2014 08:55:31 +0000 (10:55 +0200)]
Fix #73 by using `ansible_fqdn`
Alessandro Corbelli [Fri, 18 Apr 2014 08:52:22 +0000 (10:52 +0200)]
Fix #71. Removed wrong `---`
Alessandro Corbelli [Fri, 18 Apr 2014 08:47:30 +0000 (10:47 +0200)]
Fix #69 - mon osd min down reporters
Leseb [Mon, 14 Apr 2014 13:19:03 +0000 (15:19 +0200)]
Merge pull request #68 from leseb/purge-ceph
Add a purge playbook to remove Ceph
Sébastien Han [Fri, 11 Apr 2014 14:34:33 +0000 (16:34 +0200)]
Add a purge playbook to remove Ceph
The playbook removes packages, configuration files and ALL THE DATA from
Ceph.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Fri, 11 Apr 2014 12:08:52 +0000 (14:08 +0200)]
Merge pull request #67 from leseb/improve-ceph-conf
Improve ceph.conf
Sébastien Han [Fri, 11 Apr 2014 12:07:37 +0000 (14:07 +0200)]
Improve ceph.conf
Add tunables for:
* recovery
* performance
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Wed, 9 Apr 2014 15:53:55 +0000 (17:53 +0200)]
Merge pull request #66 from leseb/wip-maintenance
Add maintenance playbook
Sébastien Han [Tue, 25 Mar 2014 16:53:15 +0000 (17:53 +0100)]
Add maintenance playbook
This playbook was made to automate Ceph servers maintenance
Typical use case: hardware change
By running this playbook you will set the 'noout' flag on your
cluster, which means that OSD **can't** be marked as out
of the CRUSH map, but they will be marked as down.
Basically we tell the cluster to don't move any data since
the operation won't last for too long.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Mon, 7 Apr 2014 11:29:17 +0000 (13:29 +0200)]
Merge pull request #65 from guestisp/add-support-for-subdomains-calls
Add support for subdomain calls in S3 api.
Alessandro Corbelli [Mon, 7 Apr 2014 10:06:24 +0000 (12:06 +0200)]
Add support for subdomain calls in S3 api. See http://ceph.com/docs/master/radosgw/config/#enabling-subdomain-s3-calls
Leseb [Wed, 26 Mar 2014 10:35:54 +0000 (11:35 +0100)]
Merge pull request #61 from guestisp/remove-default-users-from-rgw
Remove RadosGW default users
Alessandro Corbelli [Wed, 26 Mar 2014 10:34:05 +0000 (11:34 +0100)]
Totally removed test users creation from RGW tasks
Leseb [Wed, 26 Mar 2014 10:27:32 +0000 (11:27 +0100)]
Merge pull request #62 from guestisp/fix-54-use-hostname-in-ceph-conf
Use ansible_hostname in RGW config inside ceph.conf
Alessandro Corbelli [Wed, 26 Mar 2014 10:26:18 +0000 (11:26 +0100)]
Use ansible_hostname in RGW config inside ceph.conf
Fix #54
Alessandro Corbelli [Wed, 26 Mar 2014 10:22:37 +0000 (11:22 +0100)]
Remove RadosGW default users
Leseb [Sun, 23 Mar 2014 22:20:09 +0000 (23:20 +0100)]
Update README
Leseb [Sun, 23 Mar 2014 10:59:58 +0000 (11:59 +0100)]
Merge pull request #59 from jcftang/wip-motd-mon
Drop a update-motd.d fragment for monitors so when the user/admin logs i...
Jimmy Tang [Sat, 22 Mar 2014 14:57:22 +0000 (14:57 +0000)]
Drop a update-motd.d fragment for monitors so when the user/admin logs into a monitor there is a small notice for the user about the state of the ceph cluster. Works only on ubuntu for now
Leseb [Thu, 20 Mar 2014 12:42:09 +0000 (13:42 +0100)]
Merge pull request #58 from guestisp/skip-missing-hosts-from-rgw-conf
Skip missing RGW hosts in ceph.conf
Alessandro Corbelli [Thu, 20 Mar 2014 11:14:43 +0000 (12:14 +0100)]
Skip missing RGW hosts in ceph.conf
Leseb [Wed, 19 Mar 2014 18:40:42 +0000 (19:40 +0100)]
Fix README typo
Fix typo
Leseb [Wed, 19 Mar 2014 18:38:04 +0000 (19:38 +0100)]
Merge pull request #57 from leseb/fix-readme
Highlight important facts (fsid and monitor key)
Sébastien Han [Wed, 19 Mar 2014 18:36:22 +0000 (19:36 +0100)]
Highlight important facts (fsid and monitor key)
Update the README with an entry for the fsid and monitor keys. We
dediced to remove every key from the default variables.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Wed, 19 Mar 2014 15:37:00 +0000 (16:37 +0100)]
Merge pull request #56 from leseb/fix-rgw-key-conf
Fix rgw section name
Sébastien Han [Wed, 19 Mar 2014 15:29:57 +0000 (16:29 +0100)]
Fix section name
All the rgw instance should share the same key so the section name
should not be changed.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Wed, 19 Mar 2014 15:28:20 +0000 (16:28 +0100)]
Merge pull request #55 from leseb/fix-rgw-pkg
Remove hardcoded distro
Sébastien Han [Wed, 19 Mar 2014 15:27:24 +0000 (16:27 +0100)]
Remove hardcoded distro
This was a leftover from some previous tests.
Closes: #53
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Wed, 19 Mar 2014 15:15:18 +0000 (16:15 +0100)]
Merge pull request #50 from guestisp/fix-49
Fix #49
Leseb [Wed, 19 Mar 2014 13:23:47 +0000 (14:23 +0100)]
Merge pull request #52 from guestisp/fix-51
Fix #51
Leseb [Wed, 19 Mar 2014 13:13:12 +0000 (14:13 +0100)]
Merge pull request #48 from guestisp/remove-https-for-ceph-key
Remove HTTPS from ceph key url
Alessandro Corbelli [Wed, 19 Mar 2014 11:45:56 +0000 (12:45 +0100)]
Fix #51
Alessandro Corbelli [Wed, 19 Mar 2014 11:40:02 +0000 (12:40 +0100)]
Fix #49
Alessandro Corbelli [Wed, 19 Mar 2014 10:56:31 +0000 (11:56 +0100)]
Remove HTTPS from ceph key url. It's a public key, no need for HTTPS as it trigger "failed to validate the SSL certificate for ceph.com:443" errors
Leseb [Mon, 17 Mar 2014 12:58:21 +0000 (13:58 +0100)]
Merge pull request #47 from jcftang/wip-haproxy
Example HAProxy configuration
Jimmy Tang [Mon, 17 Mar 2014 10:27:49 +0000 (10:27 +0000)]
re-tab/indent configuration
Jimmy Tang [Mon, 17 Mar 2014 09:17:14 +0000 (09:17 +0000)]
Fix typo in handler
Jimmy Tang [Mon, 17 Mar 2014 07:17:27 +0000 (07:17 +0000)]
note what the radosgw_interface variable means
Jimmy Tang [Sun, 16 Mar 2014 09:43:49 +0000 (09:43 +0000)]
make radosgw_interface configurable for haproxy
Jimmy Tang [Sun, 16 Mar 2014 09:40:27 +0000 (09:40 +0000)]
microformat variable to be in line with rest of roles and playbooks
Jimmy Tang [Sun, 16 Mar 2014 09:39:33 +0000 (09:39 +0000)]
fix up indentation in configuration file
Jimmy Tang [Sat, 15 Mar 2014 12:10:59 +0000 (12:10 +0000)]
more meaningful backend name
Jimmy Tang [Fri, 14 Mar 2014 09:50:31 +0000 (09:50 +0000)]
Initial add of haproxy example for fronting a number of radosgw's
Leseb [Fri, 14 Mar 2014 16:13:15 +0000 (17:13 +0100)]
Merge pull request #46 from leseb/fix-mon-initial-members
Remove mon_initial_members option
Sébastien Han [Fri, 14 Mar 2014 16:11:27 +0000 (17:11 +0100)]
Remove mon_initial_members option
The mon_initial_members is not used since we declare the mon section in
the ceph.conf file. Later, we could reduce the ceph.conf file by only
using the mon_host flag instead of all the mon sections.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Fri, 14 Mar 2014 15:48:18 +0000 (16:48 +0100)]
Merge pull request #45 from leseb/improve-read
Improve README file
Sébastien Han [Fri, 14 Mar 2014 15:32:00 +0000 (16:32 +0100)]
Improve README file
Clarify the support section, add OS support.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Fri, 14 Mar 2014 09:48:35 +0000 (10:48 +0100)]
Merge pull request #43 from leseb/feature-reduce-vm-memory
Reduce VM memory consumption
Sébastien Han [Fri, 14 Mar 2014 09:47:06 +0000 (10:47 +0100)]
Reduce VM memory consumption
We now have 7 virtual machines in the Vagrant sandbox. This takes a lot
of memory. This might be a problem for some laptop and whon't help
testings.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Thu, 13 Mar 2014 22:49:12 +0000 (23:49 +0100)]
Merge pull request #7 from leseb/wip-add-rgw
Add RADOS Gateway support
Sébastien Han [Thu, 6 Mar 2014 12:54:37 +0000 (13:54 +0100)]
Add RADOS Gateway support
This is a wip branch.
This works on Ubuntu precise, Debian Wheezy and CentOS 6.4.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Wed, 12 Mar 2014 15:14:17 +0000 (16:14 +0100)]
Merge pull request #40 from guestisp/fix-ceph-conf-if-mds-are-not-reachable
Fix ceph.conf if MDSs are not reachable
Alessandro Corbelli [Wed, 12 Mar 2014 15:13:06 +0000 (16:13 +0100)]
Fix ceph.conf if MDSs are not reachable
Leseb [Wed, 12 Mar 2014 15:00:18 +0000 (16:00 +0100)]
Merge pull request #39 from guestisp/fix-ceph-conf-if-mons-unreachable
Fix ceph.conf if mons are unreachable
Leseb [Wed, 12 Mar 2014 14:58:33 +0000 (15:58 +0100)]
Merge pull request #36 from guestisp/fix-fqdn-when-copying-keys-to-osd
Use ansible_fqdn when copying keys to OSDs
Leseb [Wed, 12 Mar 2014 14:35:27 +0000 (15:35 +0100)]
Merge pull request #38 from guestisp/fix-mons-yaml
Fix typo in group_vars/mons and remove "-C foo"
Alessandro Corbelli [Wed, 12 Mar 2014 14:34:54 +0000 (15:34 +0100)]
Fix ceph.conf if mons are unreachable
Alessandro Corbelli [Wed, 12 Mar 2014 14:32:09 +0000 (15:32 +0100)]
Fix typo in group_vars/mons and remove "-C foo"
Alessandro Corbelli [Wed, 12 Mar 2014 14:05:42 +0000 (15:05 +0100)]
Use ansible_fqdn when copying from mons
Alessandro Corbelli [Wed, 12 Mar 2014 14:01:46 +0000 (15:01 +0100)]
Use ansible_fqnd also in mds role
Alessandro Corbelli [Wed, 12 Mar 2014 13:46:40 +0000 (14:46 +0100)]
Use ansible_fqdn when copying keys to OSDs
Leseb [Wed, 12 Mar 2014 12:33:57 +0000 (13:33 +0100)]
Merge pull request #34 from leseb/fix-revert-autogeneration-keys
Revert
37882255d6ac5d15b7725df6a2c15a2c0c22928f (auto key generation)
Sébastien Han [Wed, 12 Mar 2014 12:31:22 +0000 (13:31 +0100)]
Revert
37882255d6ac5d15b7725df6a2c15a2c0c22928f (auto key generation)
We introduced a key generation mechanism that aimed to ease deployment.
In the end, it brought more complexity to the playbook and doesn't
scale.
Reverting the auto generation commit and instructing users to generate
their own keys.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Wed, 12 Mar 2014 09:49:10 +0000 (10:49 +0100)]
Merge pull request #33 from leseb/feature-reduce-verbose
Ability to reduce verbosity
Sébastien Han [Wed, 12 Mar 2014 09:46:41 +0000 (10:46 +0100)]
Ability to reduce verbosity
If you don't want to see the output of all the skipped host, you can set
this value to 'False'.
This will dramatically reduce the output of the run.
Closes: #32
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Tue, 11 Mar 2014 14:25:41 +0000 (15:25 +0100)]
Merge pull request #31 from leseb/fix-cosmetic
Cosmetic on apt module
Sébastien Han [Tue, 11 Mar 2014 14:24:51 +0000 (15:24 +0100)]
Cosmetic on apt module
Use 'present' instead of 'installed'.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Tue, 11 Mar 2014 13:23:21 +0000 (14:23 +0100)]
Merge pull request #30 from leseb/fix-vagrant-osd-disk
Add more tiny disks
Sébastien Han [Tue, 11 Mar 2014 13:22:11 +0000 (14:22 +0100)]
Add more tiny disks
Instead of having 5GB disks now we have 6 disks of 1GB per virtual
machine, which is more than enough for testing.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Tue, 11 Mar 2014 13:19:17 +0000 (14:19 +0100)]
Merge pull request #29 from leseb/fix-ansible-setup2
Remove log section
Sébastien Han [Tue, 11 Mar 2014 13:18:19 +0000 (14:18 +0100)]
Remove log section
While running Ansible for the first time, this won't affec the local
git.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Tue, 11 Mar 2014 13:16:13 +0000 (14:16 +0100)]
Merge pull request #28 from leseb/fix-ansible-setup
Improve vagrant testing and ansible setup
Sébastien Han [Tue, 11 Mar 2014 13:14:37 +0000 (14:14 +0100)]
Improve vagrant testing and ansible setup
It is easier now to git clone the repository and to run your first
environment without extra configuration.
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Tue, 11 Mar 2014 11:05:30 +0000 (12:05 +0100)]
Merge pull request #12 from jcftang/wip-cleanup
Wip cleanup
Jimmy Tang [Thu, 6 Mar 2014 07:49:46 +0000 (07:49 +0000)]
break out MDS roles into its own group, shouldn't be installing the MDS on every OSD host in this example
Jimmy Tang [Thu, 6 Mar 2014 07:22:58 +0000 (07:22 +0000)]
Ignore keyrings, they should never be commited to git
Jimmy Tang [Thu, 6 Mar 2014 06:13:08 +0000 (06:13 +0000)]
Ignore some temporary files
Leseb [Tue, 11 Mar 2014 09:37:32 +0000 (10:37 +0100)]
Merge pull request #26 from leseb/wip-separate-osd-scenarios
Wip separate osd scenarios
Sébastien Han [Mon, 10 Mar 2014 15:52:31 +0000 (16:52 +0100)]
Separate OSD scenarios to multiple files
Currently everything lives in main.yml, the file has become difficult to
read at some point and can be a real mess since we keep adding new
scenarios.
I think we should separate the scenarios into dedicated files and just
do includes in the main.yml file.
Closes: #16
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Leseb [Mon, 10 Mar 2014 23:55:54 +0000 (00:55 +0100)]
Merge pull request #25 from leseb/fix-gen-keys
Autogenerate initial mon key
Sébastien Han [Mon, 10 Mar 2014 23:47:24 +0000 (00:47 +0100)]
Autogenerate initial mon key
As mentionned in the issue 24 it's not really safe to store a default
fsid nor a monitor key. Thus the commit brings the auto-generation of
the initial monitor key. However it is quite complex to do the same for
the fsid, so I leave this to the person in charge of the deployment to
generate one and edit group_vars/all accordingly. The default fsid has
been removed as well.
Close: #24
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Alfredo Deza [Mon, 10 Mar 2014 17:06:00 +0000 (13:06 -0400)]
Merge pull request #18 from leseb/fix-rm-dev
Use full device name
Alfredo Deza [Mon, 10 Mar 2014 17:05:12 +0000 (13:05 -0400)]
Merge pull request #22 from leseb/fix-allow-disable-mds
Allow to disable MDS
Alfredo Deza [Mon, 10 Mar 2014 16:49:01 +0000 (12:49 -0400)]
Merge pull request #23 from leseb/fix-hardcoded-eth
Remove hardcoded interface name
Sébastien Han [Mon, 10 Mar 2014 16:23:41 +0000 (17:23 +0100)]
Remove hardcoded interface name
Add the ability to select a binding interface for the monitors.
Closes: #20
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>
Sébastien Han [Mon, 10 Mar 2014 16:14:11 +0000 (17:14 +0100)]
Allow to disable MDS
Even if MDS are not configured in site.yml the playbook has a
dependancy on the ceph.conf template.
This disables the mds section from the ceph.conf file.
Closes: #21
Signed-off-by: Sébastien Han <sebastien.han@enovance.com>