openstack/openstack-ansible
Code Issues Proposed changes
4eb3802a00
openstack-ansible/etc/openstack_deploy/conf.d/swift.yml.example
wade-holler 3a0523a0f4 DOC: Change swift mountpoint to /srv/node 
Background: Install Guide commands use
/srv/node mount point for /etc/fstab updates,
however example configuration settings for swift.yml
use /mnt. A new deployer, or one that doesn't
follow all of the instructions closely may use /mnt
for the configuration settings and /srv/node for
/etc/fstab updates. This results in an
unuseable swift subsystem.

Change: Update install guide docs, and swift.yml.example
to utilize /srv/node as the mount point example,
consistent with the install guide example /etc/fstab
updates.

Closes-Bug: 1582043

Change-Id: Ic64d36c7481fb4fdb6122d8578a0a6cf45e6b978
2016-05-16 15:26:50 +00:00

486 lines
18 KiB
Plaintext
Raw Blame History

---
# Copyright 2015, Rackspace US, Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
# Overview
# ========
#
# This file contains the configuration for the OpenStack Ansible Deployment
# (OSA) Object Storage (swift) service. Only enable these options for
# deployments that contain the Object Storage service. For more information on
# these options, see the documentation at
#
# http://docs.openstack.org/developer/swift/index.html
#
# You can customize the options in this file and copy it to
# /etc/openstack_deploy/conf.d/swift.yml or create a new
# file containing only necessary options for your environment
# before deployment.
#
# OSA implements PyYAML to parse YAML files and therefore supports structure
# and formatting options that augment traditional YAML. For example, aliases
# or references. For more information on PyYAML, see the documentation at
#
# http://pyyaml.org/wiki/PyYAMLDocumentation
#
# Configuration reference
# =======================
#
# Level: global_overrides (required)
# Contains global options that require customization for a deployment. For
# example, the ring stricture. This level also provides a mechanism to
# override other options defined in the playbook structure.
#
# Level: swift (required)
# Contains options for swift.
#
# Option: storage_network (required, string)
# Name of the storage network bridge on target hosts. Typically
# 'br-storage'.
#
# Option: repl_network (optional, string)
# Name of the replication network bridge on target hosts. Typically
# 'br-repl'. Defaults to the value of the 'storage_network' option.
#
# Option: part_power (required, integer)
# Partition power. Applies to all rings unless overridden at the 'account'
# or 'container' levels or within a policy in the 'storage_policies' level.
# Immutable without rebuilding the rings.
#
# Option: repl_number (optional, integer)
# Number of replicas for each partition. Applies to all rings unless
# overridden at the 'account' or 'container' levels or within a policy
# in the 'storage_policies' level. Defaults to 3.
#
# Option: min_part_hours (optional, integer)
# Minimum time in hours between multiple moves of the same partition.
# Applies to all rings unless overridden at the 'account' or 'container'
# levels or within a policy in the 'storage_policies' level. Defaults
# to 1.
#
# Option: region (optional, integer)
# Region of a disk. Applies to all disks in all storage hosts unless
# overridden deeper in the structure. Defaults to 1.
#
# Option: zone (optional, integer)
# Zone of a disk. Applies to all disks in all storage hosts unless
# overridden deeper in the structure. Defaults to 0.
#
# Option: weight (optional, integer)
# Weight of a disk. Applies to all disks in all storage hosts unless
# overridden deeper in the structure. Defaults to 100.
#
# Option: reclaim_age (optional, integer, default 604800)
# The amount of time in seconds before items, such as tombstones are
# reclaimed, default is 604800 (7 Days).
#
# Option: statsd_host (optional, string)
# Swift supports statsd metrics, this option sets the statsd host that will
# receive statsd metrics. Specifying this here will apply to all hosts in
# a cluster. It can be overridden or specified deeper in the structure if you
# want to only catch statsd metrics on certain hosts.
#
# Option: statsd_port (optional, integer, default 8125)
# Statsd port, requires statsd_host set.
#
# Option: statsd_metric_prefix (optional, string, default ansible_host)
# Specify a prefix that will be prepended to all metrics, this should be specified
# deeper in the configuration so different host metrics can be separated.
#
# The following statsd related options are a little more complicated and are
# used to tune how many samples are sent to statsd. If you need to tweak these
# settings then first read: http://docs.openstack.org/developer/swift/admin_guide.html
#
# Option: statsd_default_sample_rate (optional, float, default 1.0)
# Option: statsd_sample_rate_factor (optional, float, default 1.0)
#
# Example:
#
# Define a typical deployment:
#
# - Storage network that uses the 'br-storage' bridge. Proxy containers
# typically use the 'storage' IP address pool. However, storage hosts
# use bare metal and require manual configuration of the 'br-storage'
# bridge on each host.
# - Replication network that uses the 'br-repl' bridge. Only storage hosts
# contain this network. Storage hosts use bare metal and require manual
# configuration of the bridge on each host.
# - Ring configuration with partition power of 8, three replicas of each
# file, and minimum 1 hour between migrations of the same partition. All
# rings use region 1 and zone 0. All disks include a weight of 100.
#
# swift:
# storage_network: 'br-storage'
# replication_network: 'br-repl'
# part_power: 8
# repl_number: 3
# min_part_hours: 1
# region: 1
# zone: 0
# weight: 100
# statsd_host: statsd.example.lan
# statsd_port: 8125
#
# Note: Most typical deployments override the 'zone' option in the
# 'swift_vars' level to use a unique zone for each storage host.
#
# Option: mount_point (required, string)
# Top-level directory for mount points of disks. Defaults to /mnt.
# Applies to all hosts unless overridden deeper in the structure.
#
# Level: drives (required)
# Contains the mount points of disks.
# Applies to all hosts unless overridden deeper in the structure.
#
# Option: name (required, string)
# Mount point of a disk. Use one entry for each disk.
# Applies to all hosts unless overridden deeper in the structure.
#
# Example:
#
# Mount disks 'sdc', 'sdd', 'sde', and 'sdf' to the '/srv/node' directory on all
# storage hosts:
#
# mount_point: /srv/node
# drives:
# - name: sdc
# - name: sdd
# - name: sde
# - name: sdf
#
# Level: account (optional)
# Contains 'min_part_hours' and 'repl_number' options specific to the
# account ring.
#
# Level: container (optional)
# Contains 'min_part_hours' and 'repl_number' options specific to the
# container ring.
#
# Level: storage_policies (required)
# Contains storage policies. Minimum one policy. One policy must include
# the 'index: 0' and 'default: True' options.
#
# Level: policy (required)
# Contains a storage policy. Define for each policy.
#
# Option: name (required, string)
# Policy name.
#
# Option: index (required, integer)
# Policy index. One policy must include this option with a '0'
# value.
#
# Option: policy_type (optional, string)
# Defines policy as replication or erasure coding. Accepts 'replication'
# 'erasure_coding' values. Defaults to 'replication' value if omitted.
#
# Option: ec_type (conditionally required, string)
# Defines the erasure coding algorithm. Required for erasure coding
# policies.
#
# Option: ec_num_data_fragments (conditionally required, integer)
# Defines the number of object data fragments. Required for erasure
# coding policies.
#
# Option: ec_num_parity_fragments (conditionally required, integer)
# Defines the number of object parity fragments. Required for erasure
# coding policies.
#
# Option: ec_object_segment_size (conditionally required, integer)
# Defines the size of object segments in bytes. Swift sends incoming
# objects to an erasure coding policy in segments of this size.
# Required for erasure coding policies.
#
# Option: default (conditionally required, boolean)
# Defines the default policy. One policy must include this option
# with a 'True' value.
#
# Option: deprecated (optional, boolean)
# Defines a deprecated policy.
#
# Note: The following levels and options override any values higher
# in the structure and generally apply to advanced deployments.
#
# Option: repl_number (optional, integer)
# Number of replicas of each partition in this policy.
#
# Option: min_part_hours (optional, integer)
# Minimum time in hours between multiple moves of the same partition
# in this policy.
#
# Example:
#
# Define three storage policies: A default 'gold' policy, a deprecated
# 'silver' policy, and an erasure coding 'ec10-4' policy.
#
# storage_policies:
# - policy:
# name: gold
# index: 0
# default: True
# - policy:
# name: silver
# index: 1
# repl_number: 3
# deprecated: True
# - policy:
# name: ec10-4
# index: 2
# policy_type: erasure_coding
# ec_type: jerasure_rs_vand
# ec_num_data_fragments: 10
# ec_num_parity_fragments: 4
# ec_object_segment_size: 1048576
#
# --------
#
# Level: swift-proxy_hosts (required)
# List of target hosts on which to deploy the swift proxy service. Recommend
# three minimum target hosts for these services. Typically contains the same
# target hosts as the 'shared-infra_hosts' level in complete OpenStack
# deployments.
#
# Level: <value> (optional, string)
# Name of a proxy host.
#
# Option: ip (required, string)
# IP address of this target host, typically the IP address assigned to
# the management bridge.
#
# Level: container_vars (optional)
# Contains options for this target host.
#
# Level: swift_proxy_vars (optional)
# Contains swift proxy options for this target host. Typical deployments
# use this level to define read/write affinity settings for proxy hosts.
#
# Option: read_affinity (optional, string)
# Specify which region/zones the proxy server should prefer for reads
# from the account, container and object services.
# E.g. read_affinity: "r1=100" this would prefer region 1
# read_affinity: "r1z1=100, r1=200" this would prefer region 1 zone 1
# if that is unavailable region 1, otherwise any available region/zone.
# Lower number is higher priority. When this option is specified the
# sorting_method is set to 'affinity' automatically.
#
# Option: write_affinity (optional, string)
# Specify which region to prefer when object PUT requests are made.
# E.g. write_affinity: "r1" - favours region 1 for object PUTs
#
# Option: write_affinity_node_count (optional, string)
# Specify how many copies to prioritise in specified region on
# handoff nodes for Object PUT requests.
# Requires "write_affinity" to be set in order to be useful.
# This is a short term way to ensure replication happens locally,
# Swift's eventual consistency will ensure proper distribution over
# time.
# e.g. write_affinity_node_count: "2 * replicas" - this would try to
# store Object PUT replicas on up to 6 disks in region 1 assuming
# replicas is 3, and write_affinity = r1
#
# Option: statsd_host (optional, string)
# Swift supports statsd metrics, this option sets the statsd host that will
# receive statsd metrics.
#
# Option: statsd_port (optional, integer, default 8125)
# Statsd port, requires statsd_host set.
#
# Option: statsd_metric_prefix (optional, string, default ansible_host)
# Specify a prefix that will be prepended to all metrics on this host.
#
# The following statsd related options are a little more complicated and are
# used to tune how many samples are sent to statsd. If you need to tweak these
# settings then first read: http://docs.openstack.org/developer/swift/admin_guide.html
#
# Option: statsd_default_sample_rate (optional, float, default 1.0)
# Option: statsd_sample_rate_factor (optional, float, default 1.0)
#
# Example:
#
# Define three swift proxy hosts:
#
# swift-proxy_hosts:
#
# infra1:
# ip: 172.29.236.101
# container_vars:
# swift_proxy_vars:
# read_affinity: "r1=100"
# write_affinity: "r1"
# write_affinity_node_count: "2 * replicas"
# infra2:
# ip: 172.29.236.102
# container_vars:
# swift_proxy_vars:
# read_affinity: "r2=100"
# write_affinity: "r2"
# write_affinity_node_count: "2 * replicas"
# infra3:
# ip: 172.29.236.103
# container_vars:
# swift_proxy_vars:
# read_affinity: "r3=100"
# write_affinity: "r3"
# write_affinity_node_count: "2 * replicas"
#
# --------
#
# Level: swift_hosts (required)
# List of target hosts on which to deploy the swift storage services.
# Recommend three minimum target hosts for these services.
#
# Level: <value> (required, string)
# Name of a storage host.
#
# Option: ip (required, string)
# IP address of this target host, typically the IP address assigned to
# the management bridge.
#
# Note: The following levels and options override any values higher
# in the structure and generally apply to advanced deployments.
#
# Level: container_vars (optional)
# Contains options for this target host.
#
# Level: swift_vars (optional)
# Contains swift options for this target host. Typical deployments
# use this level to define a unique zone for each storage host.
#
# Option: storage_ip (optional, string)
# IP address to use for accessing the account, container, and object
# services if different than the IP address of the storage network
# bridge on the target host. Also requires manual configuration of
# the host.
#
# Option: repl_ip (optional, string)
# IP address to use for replication services if different than the IP
# address of the replication network bridge on the target host. Also
# requires manual configuration of the host.
#
# Option: region (optional, integer)
# Region of all disks.
#
# Option: zone (optional, integer)
# Zone of all disks.
#
# Option: weight (optional, integer)
# Weight of all disks.
#
# Option: statsd_host (optional, string)
# Swift supports statsd metrics, this option sets the statsd host that will
# receive statsd metrics.
#
# Option: statsd_port (optional, integer, default 8125)
# Statsd port, requires statsd_host set.
#
# Option: statsd_metric_prefix (optional, string, default ansible_host)
# Specify a prefix that will prepended all metrics on this host.
#
# The following statsd related options are a little more complicated and are
# used to tune how many samples are sent to statsd. If you need to tweak these
# settings then first read: http://docs.openstack.org/developer/swift/admin_guide.html
#
# Option: statsd_default_sample_rate (optional, float, default 1.0)
# Option: statsd_sample_rate_factor (optional, float, default 1.0)
#
# Level: groups (optional)
# List of one of more Ansible groups that apply to this host.
#
# Example:
#
# Deploy the account ring, container ring, and 'silver' policy.
#
# groups:
# - account
# - container
# - silver
#
# Level: drives (optional)
# Contains the mount points of disks specific to this host.
#
# Level or option: name (optional, string)
# Mount point of a disk specific to this host. Use one entry for
# each disk. Functions as a level for disks that contain additional
# options.
#
# Option: storage_ip (optional, string)
# IP address to use for accessing the account, container, and object
# services of a disk if different than the IP address of the storage
# network bridge on the target host. Also requires manual
# configuration of the host.
#
# Option: repl_ip (optional, string)
# IP address to use for replication services of a disk if different
# than the IP address of the replication network bridge on the target
# host. Also requires manual configuration of the host.
#
# Option: region (optional, integer)
# Region of a disk.
#
# Option: zone (optional, integer)
# Zone of a disk.
#
# Option: weight (optional, integer)
# Weight of a disk.
#
# Level: groups (optional)
# List of one or more Ansible groups that apply to this disk.
#
# Example:
#
# Define four storage hosts. The first three hosts contain typical options
# and the last host contains advanced options.
#
# swift_hosts:
# swift-node1:
# ip: 172.29.236.151
# container_vars:
# swift_vars:
# zone: 0
# swift-node2:
# ip: 172.29.236.152
# container_vars:
# swift_vars:
# zone: 1
# swift-node3:
# ip: 172.29.236.153
# container_vars:
# swift_vars:
# zone: 2
# swift-node4:
# ip: 172.29.236.154
# container_vars:
# swift_vars:
# storage_ip: 198.51.100.11
# repl_ip: 203.0.113.11
# region: 2
# zone: 0
# weight: 200
# statsd_host: statsd2.example.net
# statsd_metric_prefix: swift-node4
# groups:
# - account
# - container
# - silver
# drives:
# - name: sdc
# storage_ip: 198.51.100.21
# repl_ip: 203.0.113.21
# weight: 75
# groups:
# - gold
# - name: sdd
# - name: sde
# - name: sdf
View Git Blame Copy Permalink