In case of failure, the systemd ExecStop isn't executed so the container
isn't removed. After a reboot of a failed node, the container doesn't
start because the old container is still present in created state.
We should always try to remove the container in ExecStartPre for this
situation.
A normal reboot doesn't trigger this issue and this also doesn't affect
nodes running containers via docker.
This behaviour was introduced by d43769d.
Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1858865
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 47b7c00287)
The mds_name fact always gets the ansible_hostname value so we don't
need to have a dedicated fact for this and use the ansible_hostname fact
instead.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 4e84b4beed)
The ceph-crash tasks present in the ceph-handler role don't need to be
executed on all nodes.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 18e3c7a0a2)
We should ensure the container is removed in `ExecPreStart` even when
`{{ container_binary }}` is docker.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 39bb279a53)
This commit introduces a new role `ceph-crash` in order to deploy
everything needed for the ceph-crash daemon.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 9d2f2108e1)
This commit adds crlf between each task.
It makes the playbook more readable.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 8ef9fb68bc)
when rerunning lvm_setup.yml on existing cluster with OSDs already
deployed, it fails like following:
```
fatal: [osd0]: FAILED! => changed=false
msg: Sorry, no shrinking of data-lv2 to 0 permitted.
```
because we are asking `lvol` module to create a volume on an empty VG
with size extents = `100%FREE`.
The default behavior of `lvol` is to shrink the volume if the LV's current
size is greater than the requested size.
Given the requested size is calculated like this:
`size_requested = size_percent * this_vg['free'] / 100`
in our case, it is similar to:
`size_requested = 100 * 0 / 100` which basically means `0`
So the current LV size is well greater than the requested size which
leads the module to attempt to shrink it to 0 which isn't obviously now
allowed.
Adding `shrink: false` to the module calls fixes this issue.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 218f4ae361)
This commit fixes these tasks when --limit is used.
It makes sure the fact is set on right nodes even when the playbook is
run with `--limit`
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit f8a951f50c)
The ceph-dashboard role is executed on the mgr nodes so the TLS cert/key
files are copied to those nodes.
But we are running importing the cert/key files into the ceph
configuration on the monitor.
Closes: #5557
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 2b8ebf1457)
Set the cephadm cmd as a fact instead of rewriting the same command
over and over.
This also fix an issue when using docker as container engine because
the --docker cephadm parameter should be use before the subcommand
not after.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 5ef965c4dc)
This adds a new playbook for deploying ceph via cephadm.
This also adds a new dedicated tox file for CI purpose.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 957903d561)
This is a partial revert of b38019e because we don't want to execute
the whole play on the monitor otherwise if we have some empty group
like rgws or mdss then the orchestrator commands will still be
executed.
Instead we should keep the real target group name at play level and
delegate the orchestator commands to the monitor. The whole play
will be skipped is the group is empty.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 9596494911)
Print a message at the end of the playbook to inform users that they
don't have to user ceph-ansible playbooks anymore as everything else
need to be done via cephadm (day 2 operation).
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 75ae1b7e90)
When reporting the orchestrator service/daemon list at the end of the
playbook, we can use the --refresh option otherwise we could have
an outdated output.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 7164426456)
After adopting a monitor we need to wait that monitor to join back
the quorum before moving to the next node.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 0c3a2b72ff)
Like rolling_update or switch2container playbooks, we need to set/unset
some osd flags before and after the OSD daemons adoption.
This also adds a task for waiting for clean pgs at then of an OSd node.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit d3b3c8948e)
At the end of the process when don't need the cephadm script.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit c3bbc6b13c)
At the end of the playbook we can show the orchestrator status like
we do with the ceph status in initial deployment.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 381201a394)
It's better to use the --placement parameter when using ceph orch apply
commands to avoid confusion in the parameters.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 91a6c79e41)
cephadm uses default value for dashboard container images which need to
be customized by ansible for upstream or downstream purpose.
This feature wasn't present when cephadm-adopt.yml has been designed.
Also set the container_image_base variable for upgrade purpose.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit f2d997396e)
It looks like we can't run the ceph orch apply commands on nodes other
than monitors even if it used to work in the past.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit b38019e3ca)
If the systemd service exists successfully then we don't need to reset
the failed state.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 27efcbc0e5)
The ceph config assimilate-conf command requires the client.admin
keyring which isn't present on all nodes most of the time.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit fd36433826)
This adds an optional cephadm_adopt scenario which is based on
all_daemons.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 14eed63921)
This commit makes all jobs authenticating to docker hub in order to
avoid the rate limit.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 40307f810c)
This commit adds the missing stable-5.0 details about what it is
supported in this branch regarding ceph/ansible.
Fixes: #5519
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
The download.nfs-ganesha.org source for nfs-ganesha on CentOS isn't
available anymore.
Let's switch back to shaman since we have builds available now.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 1438ca0120)
We don't need to override this variable in the group_vars but use the
default value instead.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit fc599ed9f5)
not really needed in master, commit intended to be backported in octopus
branch.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 5b6f5486f7)
This commit adds a note about `stable-3.0` `stable-3.1` branches which
are deprecated and not maintained anymore.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit bbe30bcc69)
This commit updates the documentation to add a note about containerized
deployments.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit e61488507b)
nfs-ganesha 3.3 is the latest 3.x release available for octopus so we
should update to this version.
https://download.ceph.com/nfs-ganesha/rpm-V3.3-stable/octopus
This will also match the version used in RHCS 5.
Ceph container already uses that version too.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 93754bd70c)
CentOS 8.2 (2004) has been relesed so we should switch to this image
when using vagrant.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 72293b6614)
This variable isn't consumed by the container so we can remove it.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 1361e84a4e)
When rgw and osd are collocated, the current workflow prevents from
scaling out the radosgw_num_instances parameter when rerunning the
playbook.
The environment file used in the rgw systemd template is rendered when
executing the `ceph-rgw` role but during a new run of the playbook (in
order to scale out rgw instances), handlers are triggered from `ceph-osd`
role which is run before `ceph-rgw`, therefore it tries to start the new
rgw daemon whereas its corresponding environment file hasn't been
rendered yet and fails like following:
```
ceph-radosgw@rgw.ceph4osd3.rgw1.service failed to run 'start-pre' task: No such file or directory
```
This commit moves the tasks generating this file in `ceph-config` role
so it is generated early.
Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1851906
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 7dd68b9ac1)
By default, ansible gathers facts from facter and ohai if installed on
the remote nodes, given we don't need them, let's exclude these facts
from our facts gathering
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit c95adc564b)
All EL8 dependencies are now present on EPEL 8 so we don't need the
additional repositories that were only a temporary solution.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 3592ba1d61)
If a failure occurs in ceph-validate, the upgrade playbook keeps running
where we expect it to fail.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
(cherry picked from commit 8f9cdf4b10)
We need to set the mgr dashboard server ip address before restarting the
dashboard module otherwise we can try to bind the dashboard module on an
already used address.
We already do this configuration for the dashboard port value and ssl
setup so we should do the same for server address too.
Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1851455
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 03cd75845f)
The commit adds a new playbook for converting an existing ceph cluster
deployed by ceph-ansible to the cephadm orchestrator.
Signed-off-by: Dimitri Savineau <dsavinea@redhat.com>
(cherry picked from commit 548ff26256)
This fixes a long standing fail in ceph-volumes lvm test suite.
Otherwise the default behaviour should not change.
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
(cherry picked from commit 1fe8e819f9)
The use of "| success" and "| changed" are not valid syntax for modern
ansible releases.
Signed-off-by: Jonathan Rosser <jonathan.rosser@rd.bbc.co.uk>
(cherry picked from commit 42884e8175)
Dimitri Savineau
Guillaume Abrioux
Jan Fajerski
George Shuklin
Jonathan Rosser