Commit Graph

4656 Commits (77d24203fa61f16c5e70a5d7b6a9b4225f76356d)
 

Author SHA1 Message Date
Guillaume Abrioux 5c3ce4ca77 mon: fetch initial keyring even when running rolling_update
otherwise, the task to copy mgr keyring fails during the rolling_update.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-25 16:02:56 -04:00
Guillaume Abrioux f0e616962d tests: split tox configuration into multiple pieces
This file is becoming too big, let's isolate the update related code in
a dedicated tox configuration file.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-25 16:02:56 -04:00
Guillaume Abrioux f31d6d9485 update: enable new nautilus-only functionality
once the cluster is upgraded to nautilus, we can complete the process by
disallowing pre-nautilus OSDs and enabling all new nautilus-only functionality

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-25 16:02:56 -04:00
Guillaume Abrioux afdaa70a63 update: enable msgr2 protocol
This commit enable the msgr2 protocol when the cluster is fully upgraded
to nautilus

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-25 16:02:56 -04:00
Guillaume Abrioux ef096dd021 update: ensure mgrs are upgraded after ALL monitors
As of 1c760904b0, ceph-ansible implicitly
bootstrap managers on monitors.
mgrs must be upgraded only after all monitors, therefore, this commit
refact the way mgrs are upgraded to be sure we don't upgrade a mgr
during the monitors upgrade.

This commit also ensure we handle the case were we split managers on
dedicated nodes.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-25 16:02:56 -04:00
Guillaume Abrioux 7fa2434f0f update: ensure /var/lib/ceph/bootstrap-rbd-mirror is present
This directory is created by ceph-config node by node.
In the upgrade context we need it to be created on ALL monitors as soon
as the first iteration because of the task right after which creates and sends
the keyrings on all monitors.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-25 16:02:56 -04:00
Guillaume Abrioux 82764afe8d update: mask systemd service units during upgrade
This prevents the packaging from restarting services before we do need
to restart them in the rolling update sequence.
We want to handle services restart at rolling_update playbook.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-25 16:02:56 -04:00
Guillaume Abrioux 8add55451c update: set osd flags only once
There is no need to set osd flags (noout, norebalance) each time we
upgrade a mon.

This commit moves up those tasks (before stopping the mon) so we don't need
to delegate them.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-25 16:02:56 -04:00
Guillaume Abrioux f7c6f4e0b6 update: fix tasks waiting for the node to join the quorum
We actually want to ensure the node being upgraded is joining the quorum
instead of the monitor picked up earlier.

Indeed, the `mon_host`is used only in `delegate_to:` so we can still run ceph
commands while the monitor being upgraded is stopped.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-25 16:02:56 -04:00
Guillaume Abrioux 32569b79e2 update: remove an old parameter in ceph_key module call
the `containerized` parameter in ceph_key module doesn't exist anymore.
This was making the module failing but was hidden because of the
`ignore_errors: True`.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-25 16:02:56 -04:00
Guillaume Abrioux b4f14aba8e ceph_key: `lookup_ceph_initial_entities` shouldn't fail on update
As of nautilus, the initial keyrings list has changed, it means when
upgrading from Luminous or Mimic, it is expected there's a mismatch
between what is found on the cluster and the expected initial keyring
list hardcoded in ceph_key module. We shouldn't fail when upgrading to
nautilus.

str_to_bool() took from ceph-volume.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
Co-Authored-by: Alfredo Deza <adeza@redhat.com>
2019-03-25 16:02:56 -04:00
Guillaume Abrioux e99305c684 handlers: do not trigger handlers on rolling_update
rolling_update playbook already takes care of stopping/starting services
during the sequence. There's no need to trigger potential unwanted
services restart.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-25 16:02:56 -04:00
Dimitri Savineau 179fdfbc19 ceph-osd: Ensure lvm2 is installed
When using osd_scenario lvm, we never check if the lvm2 package is
present on the host.
When using containerized deployment and docker on CentOS/RedHat this
package will be automatically installed as a dependency but not for
Ubuntu distribution.
OSD deployed via ceph-volume require the lvmetad.socket to be active
and running.

Resolves: #3728

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-20 22:26:45 +00:00
Bruceforce 6d506dba1a ceph_crush: fix rstrip for python 3
Removing bytes literals since rstrip only supports type String or None.

Please backport to stable-3.2

Signed-off-by: Bruceforce <markus.greis@gmx.de>
2019-03-19 23:57:22 +00:00
Phuong Nguyen 3305309e87 Remove trailing forward slash in ceph_docker_registry variable from group_vars/rhcs.yml.sample file.
Also fixed rhcs_edits.txt for variable ceph_docker_registry.

Moved namespace to ceph_docker_image variable.

Signed-off-by: Phuong Nguyen <pnguyen@redhat.com>
2019-03-19 13:15:59 +00:00
Guillaume Abrioux 987bdac963 osd: backward compatibility with old disk_list.sh location
Since all files in container image have moved to `/opt/ceph-container`
this check must look for new AND the old path so it's backward
compatible. Otherwise it could end up by templating an inconsistent
`ceph-osd-run.sh`.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-18 17:25:51 +00:00
Dimitri Savineau 5c39735be5 ceph-validate: fail if there's no ipaddr available in monitor_address_block subnet
When using monitor_address_block to determine the ip address of the
monitor node, we need an ip address available in that cidr to be
present in the ansible facts (ansible_all_ipv[46]_addresses).
Currently we don't check if there's an ip address available during
the ceph-validate role.
As a result, the ceph-config role fails due to an empty list during
ceph.conf template creation but the error isn't explicit.

TASK [ceph-config : generate ceph.conf configuration file] *****
fatal: [0]: FAILED! => {"msg": "No first item, sequence was empty."}

With this patch we will fail before the ceph deployment with an
explicit failure message.

Resolves: rhbz#1673687

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-18 16:35:36 +00:00
Dimitri Savineau a7b1e35a16 ceph-common: Install yum plugin priorities
When using community repository we need to set the priority on the
ceph repositories because we could have some conflict with EPEL
packages.
In order to set the priority on the ceph repositories, we need to
install the yum-plugin-priorities package.

http://docs.ceph.com/docs/master/install/get-packages/#rpm-packages

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-16 06:24:55 +00:00
Guillaume Abrioux 655bdb189c Revert "site.yml: run ceph-validate before facts/defaults roles"
This commit wasn't making any sense and should have never got merged.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-15 16:17:15 +00:00
Rishabh Dave f7b20dbb48 don't append path components while calling os.path.join()
This creates a confusion whether directory/file names are being
formed by appendng strings or path components are being appended.
Since latter should never be done manually, get rid of the statements
creating confusion.

Signed-off-by: Rishabh Dave <ridave@redhat.com>
2019-03-14 22:35:12 +00:00
Rishabh Dave ba949acab7 don't use os.path.join() on a single path component
Signed-off-by: Rishabh Dave <ridave@redhat.com>
2019-03-14 22:35:12 +00:00
Rishabh Dave 62abe7068a use os.path.join() correctly
os.path.join adds the separator (i.e. '/') between the provided path
components only if needed. Providing a single path component doesn't
lead to any checks.

Signed-off-by: Rishabh Dave <ridave@redhat.com>
2019-03-14 22:35:12 +00:00
wumingqiao 31617afca9 ceph-mgr: run mgr_modules.yml only on the first mgr host
the task will be delegated to mons[0] for all mgr hosts, so we can just run it on the first host and have the same effect.

Signed-off-by: wumingqiao <wumingqiao@beyondcent.com>
2019-03-14 20:16:33 +00:00
Dimitri Savineau d8538ad4e1 Set the default crush rule in ceph.conf
Currently the default crush rule value is added to the ceph config
on the mon nodes as an extra configuration applied after the template
generation via the ansible ini module.

This implies two behaviors:

1/ On each ceph-ansible run, the ceph.conf will be regenerated via
ceph-config+template and then ceph-mon+ini_file. This leads to a
non necessary daemons restart.

2/ When other ceph daemons are collocated on the monitor nodes
(like mgr or rgw), the default crush rule value will be erased by
the ceph.conf template (mon -> mgr -> rgw).

This patch adds the osd_pool_default_crush_rule config to the ceph
template and only for the monitor nodes (like crush_rules.yml).
The default crush rule id is read (if exist) from the current ceph
configuration.
The default configuration is -1 (ceph default).

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1638092

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-14 08:56:52 +00:00
Dimitri Savineau b7f4e3e7c7 ceph-osd: Install numactl package when needed
With 3e32dce we can run OSD containers with numactl support.
When using numactl command in a containerized deployment we need to
be sure that the corresponding package is installed on the host.
The package installation is only executed when the
ceph_osd_numactl_opts variable isn't empty.

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-12 07:43:06 +00:00
Guillaume Abrioux b89a1d5c84 samples: resync sample files
I suspect `./generate_group_vars_sample.sh` wasn't used in
b8d580b3f4 because it introduced a typo in
`group_vars/all.yml.sample` and `group_vars/clients.yml.sample`.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-11 10:14:50 +01:00
Guillaume Abrioux b3eb9206fa osd: support numactl options on OSD activate
This commit adds OSD containers activate with numactl support.

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1684146

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-11 10:14:50 +01:00
Dimitri Savineau b23c05ae52 add-osd.yml: Add become flag for ceph-validate
The check_devices task fails if the ceph-validate role isn't executed
as a privileged user (Permission denied).

failed: [osd0] (item=/dev/sdb) => {"changed": false, "err": "Error:
Error opening /dev/sdb: Permission denied\n", "item": "/dev/sdb",
"msg": "Error while getting device information with parted script:
'/sbin/parted -s -m /dev/sdb -- unit 'MiB' print'", "out": "", "rc": 1}

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-09 05:54:46 +00:00
Dimitri Savineau a089e1ec23 systemd/service: Set docker.service conditionally
We don't need to set After=docker.service when the container_binary
variable isn't set to docker.
It doesn't break anything currently but it could be confusing when
using podman.

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-07 20:56:11 +00:00
Dimitri Savineau d6e71d769c common: Use rhsm_repository module for RHCS
Instead of using subscription-manager with command module we can use
the rhsm_repository ansible module.
This module already uses repos list feature to determine if a
repository is enabled or not. That way this module is idempotent so
we don't need changed_when: false anymore.

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-07 19:15:42 +00:00
Dimitri Savineau 5da9a7dec5 ceph_key: Use client name to build key path
Because the client name is part of the client key path we can reuse
the user variable to build this path.
Also remove a duplicate user variable declaration.

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-07 08:59:04 +00:00
Dimitri Savineau 676b4c979b travis: Add python 2.7
Because we're still using Linux distributions with python 2.7 (like
CentOS/RHEL 7) it could be useful to run travis tests against python
2.7 even if the support will be ended in 2020.

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-06 02:25:55 +00:00
Dimitri Savineau 53514a5b50 common: Add noarch to community repository
The ceph stable community repository only enables the basearch
packages url.
Adding the noarch url because starting with nautilus release, some
packages are added there and useful for mgr or grafana.

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-06 00:25:11 +00:00
Dimitri Savineau 4d32ecc980 Force osd pool min_size value to integer
After b8d580b and e9e5d5a we could have either item.min_size or
osd_pool_default_min_size using string instead of int causing the
condition to be true when it's false.
As a result, the task could try to set the pool min_size value to
0 which leads to:

Error EINVAL: pool min_size must be between 1 and 1

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-05 19:48:09 +00:00
Dimitri Savineau cb381b41fe Add CONTAINER_IMAGE env var to ceph daemons
Ceph daemons will set the CONTAINER_IMAGE environment variable value
in the daemon metadata.

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-05 15:07:05 +00:00
Guillaume Abrioux e9e5d5a39a fix pool min_size customization
b8d580b3f4 introduced a bug when
`min_size` isn't set (default to 0).

Typical error:

```
Error EINVAL: pool min_size must be between 1 and 1
```

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-05 13:29:34 +00:00
Radu Toader b8d580b3f4 Customize pools min_size
Signed-off-by: Radu Toader <radu.m.toader@gmail.com>
2019-03-05 10:57:15 +00:00
Radu Toader 2048255f61 When creating pool, read pool.application and make the call to ceph osd pool enable application
Signed-off-by: Radu Toader <radu.m.toader@gmail.com>
2019-03-05 09:16:03 +00:00
Kevin Coakley b11dc13476 Updated 7 ansible-lint issues in the ceph-mon, ceph-osd, and ceph-rgw roles
The following lint issues have been resolved:

[301] Commands should not change things if nothing needs doing
/home/travis/build/ceph/ceph-ansible/roles/ceph-mon/tasks/ceph_keys.yml:2

[305] Use shell only when shell functionality is required
/home/travis/build/ceph/ceph-ansible/roles/ceph-osd/tasks/start_osds.yml:47

[301] Commands should not change things if nothing needs doing
/home/travis/build/ceph/ceph-ansible/roles/ceph-rgw/tasks/multisite/destroy.yml:2

[301] Commands should not change things if nothing needs doing
/home/travis/build/ceph/ceph-ansible/roles/ceph-rgw/tasks/multisite/destroy.yml:7

[301] Commands should not change things if nothing needs doing
/home/travis/build/ceph/ceph-ansible/roles/ceph-rgw/tasks/multisite/destroy.yml:14

[301] Commands should not change things if nothing needs doing
/home/travis/build/ceph/ceph-ansible/roles/ceph-rgw/tasks/multisite/destroy.yml:19

[301] Commands should not change things if nothing needs doing
/home/travis/build/ceph/ceph-ansible/roles/ceph-rgw/tasks/multisite/destroy.yml:24

Signed-off-by: Kevin Coakley <kcoakley@sdsc.edu>
2019-03-04 22:25:35 +00:00
Guillaume Abrioux 359f8a9a4a nfs: fix systemd template service for ubuntu
`mkdir` is located in `/bin` on Ubuntu.
Let's use some jinja to support Ubuntu.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-04 19:54:25 +00:00
Guillaume Abrioux 3d3eee8f38 tests: add symlink for ubuntu hosts inventory
otherwise a bunch of jobs will fail like following:

```
 [WARNING]: Unable to parse /home/jenkins-build/build/workspace/ceph-ansible-nightly-luminous-ubuntu-container-stable-3.2-bluestore_lvm_osds/tests/functional/bs-lvm-osds/container/hosts-ubuntu as an inventory source
 [WARNING]: No inventory was parsed, only implicit localhost is available
 [WARNING]: provided hosts list is empty, only localhost is available. Note that the implicit localhost does not match 'all'
```

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-04 18:05:26 +01:00
Guillaume Abrioux b42250332a tests: pin testinfra version
As of testinfra 2.0.0, the binary name is `py.test`.

But let's pin the version to 1.19.0.
Indeed, migrating to 2.0.0 requires our current testing to be reworked a bit.
Since we don't have the bandwidth ATM for this, it's better to simply
keep testing with testinfra 1.19.0.

Note that I've replaced all `testinfra` occurences by `py.test` anyway.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-04 14:44:27 +01:00
Guillaume Abrioux a440878533 add-osd: gather facts in second part of playbook
otherwise, it will end up with error like following:

```
FAILED! => {"msg": "'ansible.vars.hostvars.HostVarsVars object' has no attribute 'ansible_hostname'"}
```

because facts won't have been gathered.

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1670663

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-04 14:44:27 +01:00
Guillaume Abrioux 47ebef374f purge: fix rbd-mirror group name
the default is rbdmirrors in ceph-defaults

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-01 20:31:14 +00:00
Guillaume Abrioux a915308477 purge: fix rbd mirror purge
as of b70d54ac80 the service launched isn't
ceph-rbd-mirror@admin.service.

it's now `ceph-rbd-mirror@rbd-mirror.{{ ansible_hostname }}`

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-01 20:31:14 +00:00
Guillaume Abrioux 3849f30f58 purge: do not remove /var/lib/apt/lists/*
removing the content of this directory seems a bit agressive and cause a
redeployment to fail after a purge on debian based distrubition.

Typical error:
```
fatal: [mon0]: FAILED! => changed=false
  attempts: 3
  msg: No package matching 'ceph' is available
```

The following task will consider the cache is still valid, so apt
doesn't refresh it:
```
- name: update apt cache if cache_valid_time has expired
  apt:
    update_cache: yes
    cache_valid_time: 3600
  register: result
  until: result is succeeded
```

since the task installing ceph packages has a `update_cache: no` it
fails:

```
- name: install ceph for debian
  apt:
    name: "{{ debian_ceph_pkgs | unique }}"
    update_cache: no
    state: "{{ (upgrade_ceph_packages|bool) | ternary('latest','present') }}"
    default_release: "{{ ceph_stable_release_uca | default('') }}{{ ansible_distribution_release ~ '-backports' if ceph_origin == 'distro' and ceph_use_distro_backports else '' }}"
  register: result
  until: result is succeeded
```

/tmp/* isn't specific to ceph as well, so we shouldn't remove everything
in this directory.

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-01 20:31:14 +00:00
Guillaume Abrioux 89f77589fa purge: fix purge of lvm devices
using `shell` module seems to be the only way to make this task working
on rhel based distribution AND debian based distributions.

on ubuntu, using `command` ansible module fails like following
(not due to `sudo` usage or not):
```
ok: [osd1] => changed=false
  cmd: command -v ceph-volume
  failed_when_result: false
  msg: '[Errno 2] No such file or directory: ''command'': ''command'''
  rc: 2
```

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1653307

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
2019-03-01 20:31:14 +00:00
Dimitri Savineau 45a7082712 lint: Fix spaces before and after variables
ansible-lint reports:

[206] Variables should have spaces after {{ and before }}

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
2019-03-01 17:22:24 +00:00
VasishtaShastry 34c25ef49b Extends check_devices tasks to non-collocated an lvm-batch scenarios
Tuned name of a task and error message to make it more user understandable

Fixes BZ 1648168 - ceph-validate : devices are not validated in non-collocated and lvm_batch scenario

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1648168

Signed-off-by: VasishtaShastry <vipin.indiasmg@gmail.com>
2019-03-01 02:13:51 +00:00
Kevin Coakley 038401fef2 Add changed_when: false to the "get osd ids" statement
The "get osd ids" statement only registers the osd_ids_non_container variable. Running "ls /var/lib/ceph/osd/ | sed 's/.*-//'" should never produce a change on the system. Adding changed_when: false prevents irrelevant change messages from Ansible.

Signed-off-by: Kevin Coakley <kcoakley@sdsc.edu>
2019-02-28 22:46:19 +00:00