`bluestore_purge_osd_non_container` scenario is failing because it
keeps old osd_uuid information on devices and cause the `ceph-disk activate`
to fail when trying to redeploy a new cluster after a purge.
typical error seen :
```
2017-12-13 14:29:48.021288 7f6620651d00 -1
bluestore(/var/lib/ceph/tmp/mnt.2_3gh6/block) _check_or_set_bdev_label
bdev /var/lib/ceph/tmp/mnt.2_3gh6/block fsid
770080e2-20db-450f-bc17-81b55f167982 does not match our fsid
f33efff0-2f07-4203-ad8d-8a0844d6bda0
```
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
this task hangs because `{{ inventory_hostname }}` doesn't resolv to an
actual ip address.
Using `hostvars[inventory_hostname]['ansible_default_ipv4']['address']`
should fix this because it will reach the node with its actual IP
address.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
Rebooting servers is really intrusive and perhaps this is not what the
operator wants. So we disable the reboot by default now. Note that the
reboot might not happen all the time.
It can be enabled by default by running the purge playbook with -e
reboot_osd_node=True
Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1505011
Signed-off-by: Sébastien Han <seb@redhat.com>
Using a condition when osd_scenario == 'non-collocated' was wrong since
these partitions can be collocated on a single device also. Removing the
check makes the purge of these partitions.
Closes: https://bugzilla.redhat.com/show_bug.cgi?id=1499871
Signed-off-by: Sébastien Han <seb@redhat.com>
The current inclusion of purge-iscsi-gateways.yml in purge-cluster.yml
is not working well and blocking the CI too. So removing it from
purge-cluster.yml and re-add the original purge-iscsi-gateways.yml.
Signed-off-by: Sébastien Han <seb@redhat.com>
The shell wildcard expansion of non-existing paths fails on zsh making
the whole script fail. We can use file module with with_fileglob to
alleviate the problem instead.
Signed-off-by: Boris Ranto <branto@redhat.com>
The systemd can't stop services if the unit files were removed before
the cluster was purged. We should just ignore these.
Signed-off-by: Boris Ranto <branto@redhat.com>
unti now, mgr nodes are not managed by purge-cluster.yml, therefore it
breaks scenario like purge_cluster.
Signed-off-by: Guillaume Abrioux <gabrioux@redhat.com>
Prior command was avoiding the lockbox mountpoint and the playbook was
failing with:
rmtree failed: [Errno 30] Read-only file system:
'/var/lib/ceph/osd-lockbox/4e9d8052-87c2-4fde-a56c-b8c108a3eefc/key-management-mode'
Signed-off-by: Sébastien Han <seb@redhat.com>
There is only two main scenarios now:
* collocated: everything remains on the same device:
- data, db, wal for bluestore
- data and journal for filestore
* non-collocated: dedicated device for some of the component
Signed-off-by: Sébastien Han <seb@redhat.com>
Problem: too many different commands to do the same thing. The 'cut'
command on infrastructure-playbooks/purge-cluster.yml was also wrong.
This sed command from osixia in ceph-docker
https://github.com/ceph/ceph-docker/pull/580/ addresses all the
scenarios.
Signed-off-by: Sébastien Han <seb@redhat.com>
This has the behavior of overriding custom values set in group_vars.
I've added defaults to the rest of the group names so that if they are
not overridden in group_vars then defaults will be used.
See: https://bugzilla.redhat.com/show_bug.cgi?id=1354700
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
Some playbooks use [0-9]*, others use \d+$
The latter is more correct since cluster name may contain numbers.
Signed-off-by: Shengjing Zhu <zsj950618@gmail.com>
Including variables from role defaults or files in a group_vars
directory relative to the playbook is a bad practice. We don't want to
do this because including these defaults at the task level overrides
values that would be set in a group_vars directory relative to the
inventory file, which is the correct usage if you wish to override
those default values.
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
Doing this cause some all the daemons to go down at the same time. In a
scenario where we colocate a monitor and an osd, this osds will take
some time to go down which will make the 'umount' task fail.
Signed-off-by: Sébastien Han <seb@redhat.com>
On systems running docker there is an issue with lxfs that results in
the find command returning 1 but actually did the job.
e.g: on a system with docker runnning find /var will give us the
following error:
find:
'/var/lib/lxcfs/cgroup/devices/lxc/x1/system.slice/systemd-update-utmp.service/devices.deny':
Permission denied
find:
'/var/lib/lxcfs/cgroup/devices/lxc/x1/system.slice/dev-random.mount/devices.allow':
Permission denied
...
...
However ceph files got deleted so we ignore the error.
Signed-off-by: Sébastien Han <seb@redhat.com>
We now rely on the cli tool ceph-detect-init which will tell us the init
system in used on the distribution. We do this instead of the previous
lookup for systemd unit files to call the right task depending on the
init system.
Signed-off-by: Sébastien Han <seb@redhat.com>
with_items is evaluated before the when so in a second run where the
variable is empty if will fail with "'dict object' has no attribute
'stdout_lines'". To fix this we had a default array so with_items does
not fail and the task is skipped with the when.
Signed-off-by: Sébastien Han <seb@redhat.com>
Because the purge-cluster.yml playbook does not have access to the roles
default vars then we can be sure that raw_multi_journal is defined. For
example, if this was purging a dmcrypt journal then raw_multi_journal
might not be defined at all in group_vars/all.yml or
group_vars/osds.yml.
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
When running encrypted OSDs, an encrypted device mapper is used (because
created by the crypsetup tool). So before attempting to remove all the
partitions on a device we must delete all the encrypted device mappers,
then we can delete all the partitions.
Signed-off-by: Sébastien Han <seb@redhat.com>
Please enter the commit message for your changes. Lines starting
The name of this variable was a bit confusing since its activation will
zap all the block devices no matter which osd scenario we are using.
Removing this variable and applying a condition on the OSD scenario is
now feasible and easier since we import group_vars variable files for
OSDs.
Signed-off-by: Sébastien Han <seb@redhat.com>
When purging OSDs we do not need to include these defaults as nothing in
the following tasks uses them. Also, it has the side effect of
overwriting any variables defined in group_vars files that are relative
to the inventory you are using with the default values. That behavior
was causing the CI tests to fail.
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
In my testing zapping the osd disks deleted the journal
partitions, making the 'zap ceph journal partitions' task fail because
the partitions it found previously do not exist anymore.
This moves the task that finds the journal partitions after 'zap osd disks'
to catch any partitions ceph-disk might have missed.
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
Using failed_when will still throw an exception and stop the playbook if
the file you're trying to include doesn't exist.
Signed-off-by: Andrew Schoen <aschoen@redhat.com>
Guillaume Abrioux
Sébastien Han
Boris Ranto
zhangwentao
Andrew Schoen
Ben England
yanyx
Daniel Lupescu
Shengjing Zhu