-
Notifications
You must be signed in to change notification settings - Fork 75
override
Default settings generated by Stage 2 and individual steps can be overridden.
Stage 2 refreshes the pillar after updating the contents. To view the current settings, run
# salt '*' pillar.items
The output for default settings for a single minion will be similar to the following:
----------
available_roles:
- admin
- mon
- storage
- mds
- igw
- rgw
- client-cephfs
- client-radosgw
- client-iscsi
- mds-nfs
- rgw-nfs
- master
cluster:
ceph
cluster_network:
172.16.22.0/24
fsid:
e08ec63c-8268-3f04-bcdb-614921e94342
master_minion:
admin.ceph
mon_host:
- 172.16.21.13
- 172.16.21.11
- 172.16.21.12
mon_initial_members:
- mon3
- mon1
- mon2
public_address:
172.16.21.11
public_network:
172.16.21.0/24
roles:
- admin
- mon
- mds
time_server:
admin.ceph
time_service:
ntp
The file /srv/pillar/ceph/stack/stack.cfg defines the directory structure of the data which implies the scope of a variable. The following files are relative to /srv/pillar/ceph/stack.
-
global.yml affects all minions in the Salt cluster
-
ceph/cluster.yml affects all minions in the the Ceph cluster named ceph.
-
ceph/roles/role.yml affects all minions assigned a role in the Ceph cluster named ceph.
-
ceph/minions/minion id.yml affects the individual minion
One current exception is the ceph_conf.yml file which contains specific variables for the Ceph configuration. This will likely become obsolete in the future.
To prevent any automation conflicting with the administrator, Stage 2 creates a parallel directory tree under /srv/pillar/ceph/stack/default. Any files in this directory tree will always be overwritten. The administrator should never make changes to files in this tree.
If any data is incorrect for your environment, override it. For instance, if the guessed cluster network is 10.0.1.0/24
, but the preferred cluster network is 172.16.22.0/24
, do the following:
-
Edit the file /srv/pillar/ceph/stack/ceph/cluster.yml
-
Add
cluster_network: 172.16.22.0/24
-
Save the file
To verify the change, run
# salt '*' saltutil.pillar_refresh
# salt '*' pillar.items
This can be repeated with any configuration data. For examples, examine any of the files under /srv/pillar/ceph/stack/default.
Many of the steps have alternate defaults. All steps have a default.sls. An alternate default state file has the prefix default-.
To select an alternate default
-
Edit the file /srv/pillar/ceph/stack/ceph/cluster.yml
-
Add
<some_init>: <some_value>
-
Save the file
To verify the change, run
# salt '*' saltutil.pillar_refresh
# salt '*' pillar.items
Note that the name of the variable to override is always defined in the init.sls.
include:
- .{{ salt['pillar.get']('<any_init>', 'default') }}
Salt will evaluate the Jinja expression and return either the value of <any_init> or default. Salt will then include a file relative to this file with an sls extension.
- Prerequisites
- Manual Installation
- Custom Profiles
- Alternate Installations
- Automated Installation
- Purging
- Reinstallation
- Replacing an OSD
- Inspecting the Configuration
- Understanding Pathnames and Arguments
- Overriding Default Settings
- Overriding Default Steps
- Man Pages
- deepsea.1
- deepsea.7
- deepsea-commands.7
- deepsea-minions.7
- deepsea-monitor.1
- deepsea-policy.cfg.5
- deepsea-stage.1
- deepsea-stage-dry-run.1
- deepsea-stage-run.1
- deepsea-stages.7
- Backporting
- Testing
- Branches & Releases