Commit Graph

3589 Commits (59551351a9fe44b81f58542e2f5824023cba532e)
 

Author SHA1 Message Date
Daniel Marks 404ad8f484 Better --check compatibility for ceph-rbd-mirror role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-27 15:00:10 +01:00
Daniel Marks ba0f16f485 Better --check compatibility for ceph-osd role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-27 15:00:10 +01:00
Daniel Marks 444c8fad40 Better --check compatibility for ceph-nfs role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-26 21:31:56 +01:00
Daniel Marks 9337b19ac0 Better --check compatibility for ceph-mon role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-26 21:05:32 +01:00
Daniel Marks 7b1dbbbdf8 Better --check compatibility for ceph-mds role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-26 20:52:24 +01:00
Daniel Marks 7371e9a668 Better --check compatibility for ceph-fetch-keys role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-26 20:42:08 +01:00
Daniel Marks c561d2635e Better --check compatibility for ceph-common role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-26 20:39:12 +01:00
Daniel Marks 2e50e5ba88 Better --check compatibility for ceph-common-coreos role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-26 17:48:06 +01:00
Daniel Marks 767c5c1c86 Better --check compatibility for ceph-agent role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-26 17:35:57 +01:00
Sébastien Han 0753e6e3bf Merge pull request #1102 from Logan2211/fix-default-syntax
Fix jinja defaulting syntax
2016-11-25 16:07:34 +01:00
Sébastien Han 69a2ed4847 Merge pull request #1118 from guits/master
[shrink-mon]: force playbook to fail if there is only one mon
2016-11-25 11:21:51 +01:00
Guillaume Abrioux b2b7222b3a [shrink-mon]: force playbook to fail if there is only one mon
The playbook will fail if only 1 mon is in the cluster
and advise to use the `purge-cluster` playbook instead.

Fix #1083
2016-11-25 11:20:11 +01:00
Logan V 684a513258 Fix jinja defaulting syntax
a397922 introduced a syntax error by attempting to default an unquoted
string, which causes execution failures on some ansible versions with:

Failed to template {{ ceph_rhcs_mount_path }}: Failed to template {{ ceph_stable_rh_storage_mount_path | default(/tmp/rh-storage-mount) }}: template error while templating string: unexpected '/'
2016-11-24 17:47:49 -06:00
Sébastien Han 78b6ae7dae Merge pull request #1117 from D3N14L/master
Improve --check compatibility for ceph-common, ceph-mon and ceph-rgw roles
2016-11-24 16:43:15 +01:00
Sébastien Han 25b04feff8 Merge pull request #1113 from guits/master
All `include_vars` need to have `*.yml`, `*.yaml` or `*.json` extension.
2016-11-24 15:19:29 +01:00
Guillaume Abrioux a680707f6f All `include_vars` need to have `*.yml`, `*.yaml` or `*.json` extension.
As introduced in the following PR:
- https://github.com/ansible/ansible/pull/17207
we need to refactor our code.
2016-11-24 14:03:49 +01:00
Daniel Marks 624ed92ee2 Better --check compatibility for ceph-rgw role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-24 11:54:43 +01:00
Daniel Marks 320dba7490 Better --check compatibility for ceph-mon role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-24 11:54:10 +01:00
Daniel Marks 3eef386532 Better --check compatibility for ceph-common role
Carefully chosen "always_run: true" parameters for read-only tasks that
register variables. This enables --check runs (at least on deployed
clusters).
2016-11-24 11:53:29 +01:00
Sébastien Han 90a28b8b41 contrib: add a new script to update remote repos
This script allows us to split ceph-ansible role directory into
repositories that can be pushed.
This script is an enhanced version of splitup.yml that was using git
subtree. I haven't found any straighforward way to push tag using git
subtree where git filter-branch seems easier and more practical.

Now we can not only push commits from master but also all the relative
tags. It is nice for projects that want to consume our roles from the
Ansible Galaxy, pointing to a specific tags looking for a particular
release.

For backward compatobility, I pushed a branch called
"master-history-bkp" on all the github.com/ceph/ansbible-ceph-*
repositories. So current projects using a SHA to retrieve the roles
won't be impacted by this change.

At the time of the commit, running this script took  104.60s to
complete.

Signed-off-by: Sébastien Han <seb@redhat.com>
Co-Authored-By: Erwan Velu <erwan@redhat.com>
Co-Authored-By: Logan Vig <logan2211@gmail.com>
2016-11-23 17:15:34 +01:00
Sébastien Han a5e00fea86 contrib: add a new script to update remote repos
This script allows us to split ceph-ansible role directory into
repositories that can be pushed.
This script is an enhanced version of splitup.yml that was using git
subtree. I haven't found any straighforward way to push tag using git
subtree where git filter-branch seems easier and more practical.

Now we can not only push commits from master but also all the relative
tags. It is nice for projects that want to consume our roles from the
Ansible Galaxy, pointing to a specific tags looking for a particular
release.

For backward compatobility, I pushed a branch called
"master-history-bkp" on all the github.com/ceph/ansbible-ceph-*
repositories. So current projects using a SHA to retrieve the roles
won't be impacted by this change.

At the time of the commit, running this script took  104.60s to
complete.

Signed-off-by: Sébastien Han <seb@redhat.com>
Co-Authored-By: Erwan Velu <erwan@redhat.com>
Co-Authored-By: Logan Vig <logan2211@gmail.com>
2016-11-23 14:51:39 +01:00
Daniel Marks 9cfd0f4118 Make NTP checks usable with '--check'
As those task are only reading the installed packages and their output
is required in later tasks it is safe to run them also in check mode.
2016-11-23 13:52:56 +01:00
Sébastien Han c8c927faeb Merge pull request #1112 from zhsj/fix-py3
fix python3 compatibility in jinja2 template.
2016-11-22 18:07:09 +01:00
Alfredo Deza dcd4e432c3 Merge pull request #1101 from ceph/ci-testing
testing: changes to allow for concurrent functional tests
2016-11-22 11:41:05 -05:00
Sébastien Han 829e2b6598 Merge pull request #1077 from font/rolling_update
Support containerized rolling update
2016-11-22 16:56:46 +01:00
Sébastien Han d6bfacdbf7 Merge pull request #1109 from ceph/bz-1396742
rolling_update: clarify "serial" usage
2016-11-22 16:10:17 +01:00
Shengjing Zhu 3e06e39550 fix python3 compatibility in jinja2 template.
rename iterkeys to keys
2016-11-22 21:52:52 +08:00
Andrew Schoen 20ac2cad9d testing: each scenario should use it's own fetch_directory
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2016-11-21 17:08:39 -06:00
Sébastien Han 38e846e542 rolling_update: clarify "serial" usage
Prior to this commit the serial variable was poorly documented. Now we
are making clear that this value should be left untouched as the rolling
update mechanism should happen serially.

Solves: bz-1396742
Signed-off-by: Sébastien Han <seb@redhat.com>
2016-11-21 14:42:46 +01:00
Sébastien Han 25d8ce958f Merge pull request #1104 from ktdreyer/rm-apache
remove apache support for RGW
2016-11-21 11:13:19 +01:00
Andrew Schoen 0570d2911b testing: remove the timeout for the ansible-playbook call
This isn't available on OSX and we'll end up hitting this timeout when
we don't want to, let's just remove it.

Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2016-11-18 14:15:34 -06:00
Andrew Schoen 0535b328a2 tests: update the subnet in the ceph.conf tests
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2016-11-18 14:15:34 -06:00
Andrew Schoen f7067e25d4 tests: add a UUID to the disk name when using libvirt
This will allow each testing scenario to have a unique names
for it's disks so there will not be conflicts when running tests
in parallel.

Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2016-11-18 14:15:08 -06:00
Ken Dreyer adfdf6871e remove apache support for RGW
libfcgi is dead upstream (http://tracker.ceph.com/issues/16784)

The RGW developers intend to remove libfcgi support entirely before the
Luminous release.

Since libfcgi gets little-to-no developer attention or testing, remove
it entirely from ceph-ansible.
2016-11-18 13:13:12 -07:00
Andrew Schoen 901f4cdfd2 tests: fix monitor_address in ubuntu/16.04/mon/hosts
We changed the subnet in vagrant_variables.yml so this needs to change
here as well.

Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2016-11-18 13:27:02 -06:00
Alfredo Deza 016a6624a7 testing: use the same sequence in the group_vars for every test scenario
Signed-off-by: Alfredo Deza <adeza@redhat.com>
2016-11-18 13:20:59 -06:00
Alfredo Deza b2af99f31b testing: use a sequence in the subnets for every vagrant_variables.yml to avoid collisions
Signed-off-by: Alfredo Deza <adeza@redhat.com>
2016-11-18 13:20:59 -06:00
Alfredo Deza f531e4d0f8 testing: vagranfile should use random hostnames to prevent collisions
Signed-off-by: Alfredo Deza <adeza@redhat.com>
2016-11-18 13:20:59 -06:00
Sébastien Han 68668333eb Merge pull request #1107 from font/osd_keyring_err
Fix error caused by missing OSD cluster keyring
2016-11-18 15:18:01 +01:00
Sébastien Han 02766cde10 Merge pull request #1106 from ktdreyer/rpm-packaging
add RPM packaging for CI
2016-11-18 15:16:14 +01:00
Ivan Font 92ffc5307d Fix error caused by missing OSD cluster keyring
Ansible task was not properly fetching OSD cluster keyring causing
the keyring to be missing when we needed to authenticate. Similarly, we
were not properly waiting on the OSD keyring to be available before
continuing.

Signed-off-by: Ivan Font <ifont@redhat.com>
2016-11-17 23:03:11 -08:00
Ken Dreyer a8557b0a83 add RPM packaging for CI
"make rpm" will build a ceph-ansible RPM and place it in the current
working directory.

This will allow us to run this command in Jenkins for every branch.
2016-11-17 15:23:44 -07:00
Alfredo Deza 5b8b9534c0 Merge pull request #1105 from ceph/libvirt-domain-conflicts
tests: sets random_hostname to true for libvirt nodes
2016-11-17 16:27:33 -05:00
Andrew Schoen 63e28248d4 tests: sets random_hostname to true for libvirt nodes
This will keep us from having name conflicts between our different
testing scenarios.

Signed-off-by: Andrew Schoen <aschoen@redhat.com>
2016-11-17 15:25:50 -06:00
Ivan Font 255e816e28 Rolling update changes for containerized deployments
Separate out systemd restart tasks for containerized and
non-containerized deployments

Signed-off-by: Ivan Font <ifont@redhat.com>
2016-11-17 11:25:25 -08:00
Ivan Font e72f08080d Warn user when upgrading cluster with only one mon
Signed-off-by: Ivan Font <ifont@redhat.com>
2016-11-17 11:25:25 -08:00
Ivan Font 3ff17f1c8f Support containerized rolling update
- Update rolling update playbook to support containerized deployments
  for mons, osds, mdss, and rgws
- Skip checking if existing cluster is running when performing a rolling
  update
- Fixed bug where we were failing to start the mds container because it
  was missing the admin keyring. The admin keyring was missing because
  it was not being pushed from the mon host to the ansible host due to
  the keyring not being available before running the copy_configs.yml
  task include file. Now we forcefully wait for the admin keyring to be
  generated before continuing with the copy_configs.yml task include file
- Skip pre_requisite.yml when running on atomic host. This technically
  no longer requires specifying to skip tasks containing the with_pkg tag
- Add missing variables to all.docker.sample
- Misc. cleanup

Signed-off-by: Ivan Font <ifont@redhat.com>
2016-11-17 11:25:25 -08:00
Andrew Schoen 5ccbd31d5f Merge pull request #1103 from ceph/gatherkeys-timeout
testing: add a timeout per ansible call to workaround possible gatherkeys issues
2016-11-17 11:49:55 -06:00
Alfredo Deza 2e9d2f872c testing: add a timeout per ansible call to workaround possible gatherkeys issues
Signed-off-by: Alfredo Deza <adeza@redhat.com>
2016-11-17 12:48:25 -05:00
Sébastien Han aa0f2d4112 Merge pull request #1099 from jsaintrocc/checkmode-fix
Replaced Ansible 2.2 specific `check_mode: no`
2016-11-17 17:02:22 +01:00