GitLab

The latest grafana container tag is using grafana 6.x release which could
cause issue with the ceph dashboard integration.
Considering that the grafana container in RHCS 3 is based on 5.x then we
should use the same version.

$ docker run --rm rhceph/rhceph-3-dashboard-rhel7:3 -v
Version 5.2.4 (commit: unknown-dev)
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 2037fb87)

Even if this improves ceph-disk/ceph-volume performances then it also
impact the ceph-osd process.
The ceph-osd process shouldn't use 1024:4096 value for the max open
files.
Removing the ulimit option from the container engine and doing this kind
of change on the container side [1].

[1] https://github.com/ceph/ceph-container/pull/1497

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1702285

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 9a996aef)

This change adds two tasks to set grafana-api user and password
that are required to inject dashboard layouts to the external
grafana instance.
Without these two parameters the ceph-ansible playbook fails
showing an authorization error (HTTPError: 401 Client Error:
Unauthorized").

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1767365

Signed-off-by: fmount <fpantano@redhat.com>
(cherry picked from commit 41b8c173)

Signed-off-by: Mihai Plasoianu <m.plasoianu@vertical.de>
(cherry picked from commit d3f67d63)

This commit adds a default value in the `with_dict` because when using
python 2.7, if a task using a `with_dict` has a condition, it is
evaluated anyway whereas in python 3 it isn't.

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1766499

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit e9823f31)

There's no need to use the default filter on active/standby groups
because if the group doesn't exist then the play is just skipped.

Currently this generates warnings like:

[WARNING]: Could not match supplied host pattern, ignoring: |
[WARNING]: Could not match supplied host pattern, ignoring: default([])
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 2ca79fcc)

The active mds host should be based on the inventory hostname and not on
the ansible hostname.
The value returns under the mdsmap structure is based on the OS hostname
so we need to find the right node in the inventory with this value when
doing operation on inventory nodes.

Othewise we could see error like:

The task includes an option with an undefined variable. The error was:
"hostvars[foobar]" is undefined
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit f1f2352c)

pycodestyle returns:

 E111 indentation is not a multiple of four
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 8a0a13f6)

To avoid unnecessary ansible warnings during playbook execution we can
move the library and plugins test files under a different directory.

[WARNING]: Skipping plugin (plugins/filter/test_ipaddrs_in_ranges.py) as
it seems to be invalid:
cannot import name 'ipaddrs_in_ranges'

Closes: #4656
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 6ce4fde8)

mon_host should use the inventory hostname and not the node hostname.
Fix creates an issue when the inventory and node hostname are different.

Closes: #4670
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 650bc0c3)

Without the become flag set to true, we can't executed the roles
successfully.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 77b21283)

This must be consistent with what is used in `name` parameter.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit d06057eb)

otherwise a single MDS would never be updated.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit d8ab11d2f8e08c28731f783d592507dd198b5742)

Let's skip this part of the code if there's no mds node in the
inventory.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 5ec906c3)

The ceph-container-engine role is missing from both playbooks so the
container engine (docker, podman) isn't install resulting in a failure
on the added nodes.

fatal: [xxxxx]: FAILED! => changed=false
  cmd: docker --version
  msg: '[Errno 2] No such file or directory'
  rc: 2

Closes: #4634
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit bfb1d6be)

Add ceph_docker_registry_username and ceph_docker_registry_password
variables in ceph-defaults role so they will be present in the group_vars
samples but commented.

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1763139

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit b33c476f)

on master, it doesn't make sense anymore to use device name, we should
use osd id instead.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit b5a61fe2)

This commit removes the backslash in allow command parameter, this was
needed before the ceph_key module integration.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 384161ed)

It doesn't make sense to test the old 3.0.x container images with
nautilus+ ceph releases.
Also disable the dashboard deployment and switch to bluestore backend.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 3c2840da)

When deploying with ceph-iscsi nodes and dashboard enabled, we need to
add the ceph iscsi gateway endpoints to the dashboard configuration and
add the mgr ip address in the trusted list in the iscsi gateway
configuration file.

Closes: #4638
Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1764173

https://docs.ceph.com/docs/master/mgr/dashboard/#enabling-iscsi-management

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit d050391c)

This commit adds the support of the ceph-iscsi stable repository when
use ceph_repository community instead of always using the devel
repositories.
We're still using the devel repositories for rtslib and tcmu-runner in
both cases (dev and community).
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit f2cb9371)

We don't have to use ceph_iscsi_config_dev (default true) on RHCS
because all iscsi packages are already included in the RHCS
repositories.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 82495eaf)

We don't need this since [1]. Also this was only working for python2 and
not supporting python3.

[1] https://github.com/ceph/ceph-iscsi/commit/00f198a

This reverts commit 167737dd

.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit fd8d47da)

When deploying with packages then the ceph-container-common role isn't
executed so the registry authentication task is ignored.

Closes: #4636
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 9ad00061)

If ansible-lint reports an error then it's skipped. We should fail in
this case.

This patch also fixes the pipefail lint in the rbd mirror role

[306] Shells that use pipes should set the pipefail option
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 3969470f)

 [303] mktemp used in place of tempfile module
 [602] Don't compare to empty string
 [701] No 'galaxy_info' found
 [702] Use 'galaxy_tags' rather than 'categories'

This patch also changes the ansible log_path value via the
ANSIBLE_LOG_PATH environment variable in the travis configuration to
avoid warnings.

[WARNING]: log file at /home/travis/ansible/ansible.log is not writeable
and we cannot create it, aborting
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit f7fd0b6d)

This task is failing when `ceph_docker_registry_auth` is enabled and
`ceph_docker_registry_username` is undefined with an ansible error
instead of the expected message.

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1763139

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit da4215e9)

Add missing quote in order to keep consistency.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 8d72ff8e)

We should show the ceph cluster status as the last task of the playbook.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 47ce9c0d)

The [group|host]_vars directories are ignored for the dashboard playbook
when the inventory file directory doesn't contain those directories.

Closes: #4601
Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1761612

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 84268562)

The commit replaces the pv/vg/lv commands used with the ansible command
module by the lvg and lvol modules.
This also fixes the size of the second data LV because we were only using
50% of the remaining space instead of 100%.

With a 50G device, the result was:
  - data-lv1 was 25G
  - data-lv2 was 12.5G
Instead of:
  - data-lv1 was 25G
  - data-lv2 was 25G
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 2c03c6fc)

This commit makes the all_daemons scenario deploying 3 mds in order to
cover the multimds case.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 25b98b2c)

This commit fixes the standby_mdss group creation by using `{{ item }}`.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit c4fc8cc8)

We don't need to have high handler delay in the CI so reducing to
10 seconds.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 04ec1ad3)

Otherwise it fails like following:

```
TASK [ceph-mds : allow multimds] **************************************************************************************************************************************************
Monday 22 July 2019  16:37:38 +0800 (0:00:03.269)       0:13:25.651 ***********
fatal: [rhel7u6clone1]: FAILED! => {"msg": "The conditional check 'ceph_release_num[ceph_release] == ceph_release_num.luminous' failed. The error was: error while evaluating conditional (ceph_release_num[ceph_release] == ceph_release_num.luminous): 'dict object' has no attribute u'dummy'\n\nThe error appears to have been in '/usr/share/ceph-ansible/roles/ceph-mds/tasks/create_mds_filesystems.yml': line 43, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n\n- name: allow multimds\n  ^ here\n"}
```

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1645379

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 4e9504c9)

this task is a leftover and no longer needed.
It even causes bug when collocating nfs with mon.

Closes: #4609
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit b63bd130)

When using python3 the name of the rtslib rpm is python3-rtslib. The
packages that use rtslib already have code that detects the python
version and distro deps, so drop it from the ceph iscsi gw task list and
let the ceph-iscsi rpm dependency handle it.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1760930

Signed-off-by: Mike Christie <mchristi@redhat.com>
(cherry picked from commit ba141298)

Due the 'failed_when: false' statement present in the peer task then
the playbook continues to ran even if the peer task was failing (like
incorrect remote peer format.

"stderr": "rbd: invalid spec 'admin@cluster1'"

This patch adds a task to list the peer present and add the peer only if
it's not already added. With this we don't need the failed_when statement
anymore.

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1665877

Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 0b1e9c07)

Refact the mds cluster upgrade code in order to follow the documented
recommandation.
See: https://github.com/ceph/ceph/blob/master/doc/cephfs/upgrading.rst

Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1569689

Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 71cebf80)

When the iscsi gateway or the ceph configuration file change then we
need to notify the rbd target api/gw services to be restarted.
This patch also merges the rbd-target-api and rbd-target-gw handler
into a single file and listen.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit bc701860)