ceph-ansible/roles/ceph-osd/tasks/raw_multi_journal.yml

35 lines
1.1 KiB
YAML

---
## SCENARIO 3: N JOURNAL DEVICES FOR N OSDS
- include: zap_devices.yml
- include: check_devices.yml
# Prepare means
# - create GPT partition for a disk, or a loop label for a partition
# - mark the partition with the ceph type uuid
# - create a file system
# - mark the fs as ready for ceph consumption
# - entire data disk is used (one big partition)
# - a new partition is added to the journal disk (so it can be easily shared)
#
# NOTE (leseb): the prepare process must be parallelized somehow...
# if you have 64 disks with 4TB each, this will take a while
# since Ansible will sequential process the loop
# NOTE (alahouze): if the device is a partition, the parted command below has
# failed, this is why we check if the device is a partition too.
- name: Prepare OSD disk(s)
command: "ceph-disk prepare {{ item.2 }} {{ item.3 }}"
when: (item.0.rc != 0 or item.1.rc != 0) and raw_multi_journal
ignore_errors: True
with_together:
- parted.results
- ispartition.results
- devices
- raw_journal_devices
changed_when: False
- include: activate_osds.yml