From 7dbbb63739db4aac973fb6d5f3f16b5e9206ce14 Mon Sep 17 00:00:00 2001 From: joehuang Date: Tue, 7 Feb 2017 04:17:31 -0500 Subject: Update the multisite documentations to reflect the progress in D As some changes in OpenStack projects like KeyStone PKI token deprecation, L2GW moved away from Neutron stadium, Tricircle shrinked scope and became OpenStack big-tent project, and Kingbird has made great progress in feature development after the initial requirements discussion. Documents need to update to reflect these recent changes. python-kingbirdclient was introduced recently, so the usage guide is updated to use python-kingbirdclient. The new feature key pair synchronization is also included in the usage guide. Change-Id: Iad9fbd441d191defa5e8793633a626ab5a24f217 Signed-off-by: joehuang --- docs/release/configguide/index.rst | 17 + .../configguide/multisite.configuration.rst | 106 ++++++ .../multisite.kingbird.configuration.rst | 264 +++++++++++++++ docs/release/installation/index.rst | 15 + .../multisite.kingbird.installation.rst | 298 +++++++++++++++++ docs/release/overview/index.rst | 12 + docs/release/overview/multisite.release.notes.rst | 11 + docs/release/userguide/index.rst | 15 + docs/release/userguide/multisite.admin.usage.rst | 365 +++++++++++++++++++++ .../release/userguide/multisite.kingbird.usage.rst | 349 ++++++++++++++++++++ .../userguide/multisite.tricircle.usage.rst | 13 + 11 files changed, 1465 insertions(+) create mode 100644 docs/release/configguide/index.rst create mode 100644 docs/release/configguide/multisite.configuration.rst create mode 100644 docs/release/configguide/multisite.kingbird.configuration.rst create mode 100644 docs/release/installation/index.rst create mode 100644 docs/release/installation/multisite.kingbird.installation.rst create mode 100644 docs/release/overview/index.rst create mode 100644 docs/release/overview/multisite.release.notes.rst create mode 100644 docs/release/userguide/index.rst create mode 100644 docs/release/userguide/multisite.admin.usage.rst create mode 100644 docs/release/userguide/multisite.kingbird.usage.rst create mode 100644 docs/release/userguide/multisite.tricircle.usage.rst (limited to 'docs/release') diff --git a/docs/release/configguide/index.rst b/docs/release/configguide/index.rst new file mode 100644 index 0000000..2ee37cb --- /dev/null +++ b/docs/release/configguide/index.rst @@ -0,0 +1,17 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 +.. (c) Sofia Wallin Ericsson AB +.. (c) Chaoyi Huang, Huawei Technologies Co., Ltd. + +***************************** +Multisite Configuration Guide +***************************** + +.. toctree:: + :numbered: + :maxdepth: 2 + + abstract.rst + multisite.configuration.rst + multisite.kingbird.configuration.rst + diff --git a/docs/release/configguide/multisite.configuration.rst b/docs/release/configguide/multisite.configuration.rst new file mode 100644 index 0000000..0a38505 --- /dev/null +++ b/docs/release/configguide/multisite.configuration.rst @@ -0,0 +1,106 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 + +Multisite identity service management +===================================== + +Goal +---- + +A user should, using a single authentication point be able to manage virtual +resources spread over multiple OpenStack regions. + +Before you read +--------------- + +This chapter does not intend to cover all configuration of KeyStone and other +OpenStack services to work together with KeyStone. + +This chapter focuses only on the configuration part should be taken into +account in multi-site scenario. + +Please read the configuration documentation related to identity management +of OpenStack for all configuration items. + +http://docs.openstack.org/liberty/config-reference/content/ch_configuring-openstack-identity.html + +How to configure the database cluster for synchronization or asynchrounous +repliation in multi-site scenario is out of scope of this document. The only +remainder is that for the synchronization or replication, only Keystone +database is required. If you are using MySQL, you can configure like this: + +In the master: + + .. code-block:: bash + + binlog-do-db=keystone + +In the slave: + + .. code-block:: bash + + replicate-do-db=keystone + + +Deployment options +------------------ + +For each detail description of each deployment option, please refer to the +admin-user-guide. + +- Distributed KeyStone service with PKI token + + In KeyStone configuration file, PKI token format should be configured + + .. code-block:: bash + + provider = pki + + or + + .. code-block:: bash + + provider = pkiz + + In the [keystone_authtoken] section of each OpenStack service configuration + file in each site, configure the identity_url and auth_uri to the address + of KeyStone service + + .. code-block:: bash + + identity_uri = https://keystone.your.com:35357/ + auth_uri = http://keystone.your.com:5000/v2.0 + + It's better to use domain name for the KeyStone service, but not to use IP + address directly, especially if you deployed KeyStone service in at least + two sites for site level high availability. + +- Distributed KeyStone service with Fernet token +- Distributed KeyStone service with Fernet token + Async replication ( + star-mode). + + In these two deployment options, the token validation is planned to be done + in local site. + + In KeyStone configuration file, Fernet token format should be configured + + .. code-block:: bash + + provider = fernet + + In the [keystone_authtoken] section of each OpenStack service configuration + file in each site, configure the identity_url and auth_uri to the address + of local KeyStone service + + .. code-block:: bash + + identity_uri = https://local-keystone.your.com:35357/ + auth_uri = http://local-keystone.your.com:5000/v2.0 + + and especially, configure the region_name to your local region name, for + example, if you are configuring services in RegionOne, and there is local + KeyStone service in RegionOne, then + + .. code-block:: bash + + region_name = RegionOne diff --git a/docs/release/configguide/multisite.kingbird.configuration.rst b/docs/release/configguide/multisite.kingbird.configuration.rst new file mode 100644 index 0000000..7eb6106 --- /dev/null +++ b/docs/release/configguide/multisite.kingbird.configuration.rst @@ -0,0 +1,264 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 + + +Configuration of Multisite.Kingbird +=================================== + +A brief introduction to configure Multisite Kingbird service. Only the +configuration items for Kingbird will be described here. Logging, +messaging, database, keystonemiddleware etc configuration which are +generated from OpenStack OSLO libary, will not be described here, for +these configuration items are common to Nova, Cinder, Neutron. So please +refer to corresponding description from Nova or Cinder or Neutron. + + +Configuration in [DEFAULT] +-------------------------- + +configuration items for kingbird-api +"""""""""""""""""""""""""""""""""""" + +bind_host +********* +- default value: *bind_host = 0.0.0.0* +- description: The host IP to bind for kingbird-api service + +bind_port +********* +- default value: *bind_port = 8118* +- description: The port to bind for kingbird-api service + +api_workers +*********** +- default value: *api_workers = 2* +- description: Number of kingbird-api workers + +configuration items for kingbird-engine +""""""""""""""""""""""""""""""""""""""" + +host +**** +- default value: *host = localhost* +- description: The host name kingbird-engine service is running on + +workers +******* +- default value: *workers = 1* +- description: Number of kingbird-engine workers + +report_interval +*************** +- default value: *report_interval = 60* +- description: Seconds between running periodic reporting tasks to + keep the engine alive in the DB. If the engine doesn't report its + aliveness to the DB more than two intervals, then the lock accquired + by the engine will be removed by other engines. + +common configuration items for kingbird-api and kingbird-engine +""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""" + +use_default_quota_class +*********************** +- default value: *use_default_quota_class = true* +- description: Enables or disables use of default quota class with default + quota, boolean value + +Configuration in [kingbird_global_limit] +---------------------------------------- + +For quota limit, a negative value means unlimited. + +configuration items for kingbird-api and kingbird-engine +"""""""""""""""""""""""""""""""""""""""""""""""""""""""" + +quota_instances +*************** +- default value: *quota_instances = 10* +- description: Number of instances allowed per project, integer value. + +quota_cores +*********** +- default value: *quota_cores = 20* +- description: Number of instance cores allowed per project, integer value. + +quota_ram +********* +- default value: *quota_ram = 512* +- description: Megabytes of instance RAM allowed per project, integer value. + +quota_metadata_items +******************** +- default value: *quota_metadata_items = 128* +- description: Number of metadata items allowed per instance, integer value. + +quota_key_pairs +*************** +- default value: *quota_key_pairs = 10* +- description: Number of key pairs per user, integer value. + +quota_fixed_ips +*************** +- default value: *quota_fixed_ips = -1* +- description: Number of fixed IPs allowed per project, this should be at + least the number of instances allowed, integer value. + +quota_security_groups +********************* +- default value: *quota_security_groups = 10* +- description: Number of security groups per project, integer value. + +quota_floating_ips +****************** +- default value: *quota_floating_ips = 10* +- description: Number of floating IPs allowed per project, integer value. + +quota_network +*************** +- default value: *quota_network = 10* +- description: Number of networks allowed per project, integer value. + +quota_subnet +*************** +- default value: *quota_subnet = 10* +- description: Number of subnets allowed per project, integer value. + +quota_port +*************** +- default value: *quota_port = 50* +- description: Number of ports allowed per project, integer value. + +quota_security_group +******************** +- default value: *quota_security_group = 10* +- description: Number of security groups allowed per project, integer value. + +quota_security_group_rule +************************* +- default value: *quota_security_group_rule = 100* +- description: Number of security group rules allowed per project, integer + value. + +quota_router +************ +- default value: *quota_router = 10* +- description: Number of routers allowed per project, integer value. + +quota_floatingip +**************** +- default value: *quota_floatingip = 50* +- description: Number of floating IPs allowed per project, integer value. + +quota_volumes +*************** +- default value: *quota_volumes = 10* +- description: Number of volumes allowed per project, integer value. + +quota_snapshots +*************** +- default value: *quota_snapshots = 10* +- description: Number of snapshots allowed per project, integer value. + +quota_gigabytes +*************** +- default value: *quota_gigabytes = 1000* +- description: Total amount of storage, in gigabytes, allowed for volumes + and snapshots per project, integer value. + +quota_backups +************* +- default value: *quota_backups = 10* +- description: Number of volume backups allowed per project, integer value. + +quota_backup_gigabytes +********************** +- default value: *quota_backup_gigabytes = 1000* +- description: Total amount of storage, in gigabytes, allowed for volume + backups per project, integer value. + +Configuration in [cache] +---------------------------------------- + +The [cache] section is used by kingbird engine to access the quota +information for Nova, Cinder, Neutron in each region in order to reduce +the KeyStone load while retrieving the endpoint information each time. + +configuration items for kingbird-engine +""""""""""""""""""""""""""""""""""""""" + +auth_uri +*************** +- default value: +- description: Keystone authorization url, for example, http://127.0.0.1:5000/v3. + +admin_username +************** +- default value: +- description: Username of admin account, for example, admin. + +admin_password +************** +- default value: +- description: Password for admin account, for example, password. + +admin_tenant +************ +- default value: +- description: Tenant name of admin account, for example, admin. + +admin_user_domain_name +********************** +- default value: *admin_user_domain_name = Default* +- description: User domain name of admin account. + +admin_project_domain_name +************************* +- default value: *admin_project_domain_name = Default* +- description: Project domain name of admin account. + +Configuration in [scheduler] +---------------------------------------- + +The [scheduler] section is used by kingbird engine to periodically synchronize +and rebalance the quota for each project. + +configuration items for kingbird-engine +""""""""""""""""""""""""""""""""""""""" + +periodic_enable +*************** +- default value: *periodic_enable = True* +- description: Boolean value for enable/disable periodic tasks. + +periodic_interval +***************** +- default value: *periodic_interval = 900* +- description: Periodic time interval for automatic quota sync job, unit is + seconds. + +Configuration in [batch] +---------------------------------------- + +The [batch] section is used by kingbird engine to periodicly synchronize +and rebalance the quota for each project. + +batch_size +*************** +- default value: *batch_size = 3* +- description: Batch size number of projects will be synced at a time. + +Configuration in [locks] +---------------------------------------- + +The [locks] section is used by kingbird engine to periodically synchronize +and rebalance the quota for each project. + +lock_retry_times +**************** +- default value: *lock_retry_times = 3* +- description: Number of times trying to grab a lock. + +lock_retry_interval +******************* +- default value: *lock_retry_interval =10* +- description: Number of seconds between lock retries. diff --git a/docs/release/installation/index.rst b/docs/release/installation/index.rst new file mode 100644 index 0000000..0687f6c --- /dev/null +++ b/docs/release/installation/index.rst @@ -0,0 +1,15 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 +.. (c) Sofia Wallin Ericsson AB +.. (c) Chaoyi Huang, Huawei Technologies Co., Ltd. + +******************************** +Multisite Installation procedure +******************************** + +.. toctree:: + :numbered: + :maxdepth: 2 + + abstract.rst + multisite.kingbird.installation.rst diff --git a/docs/release/installation/multisite.kingbird.installation.rst b/docs/release/installation/multisite.kingbird.installation.rst new file mode 100644 index 0000000..54b622d --- /dev/null +++ b/docs/release/installation/multisite.kingbird.installation.rst @@ -0,0 +1,298 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 + +================================= +Kingbird installation instruction +================================= + +Abstract +-------- +This document will give the user instructions on how to deploy +available scenarios verified for the Colorado release of OPNFV +platform. + + +Preparing the installation +-------------------------- +Kingbird is centralized synchronization service for multi-region OpenStack +deployments. In OPNFV Colorado release, Kingbird provides centralized quota +management feature. At least two OpenStack regions with shared KeyStone should +be installed first. + +Kingbird includes kingbird-api and kingbird-engine, kingbird-api and +kingbird-engine which talk to each other through message bus, and both +services access the database. Kingbird-api receives the RESTful +API request for quota management and forward the request to kingbird-engine +to do quota synchronization etc task. + +Therefore install Kingbird on the controller nodes of one of the OpenStack +region, these two services could be deployed in same node or different node. +Both kingbird-api and kingbird-engine can run in multiple nodes with +multi-workers mode. It's up to you how many nodes you want to deploy +kingbird-api and kingbird-engine and they can work in same node or +different nodes. + +HW requirements +--------------- +No special hardware requirements + +Installation instruction +------------------------ + +In colorado release, Kingbird is recommended to be installed in a python +virtual environment. So install and activate virtualenv first. + +.. code-block:: bash + + sudo pip install virtualenv + virtualenv venv + source venv/bin/activate + +Get the latest code of Kingbird from git repository: + +.. code-block:: bash + + git clone https://github.com/openstack/kingbird.git + cd kingbird/ + pip install -e . + + +or get the stable release from PyPI repository: + +.. code-block:: bash + + pip install kingbird + +In case of the database package are not installed, you may need to install: + +.. code-block:: bash + + pip install mysql + pip install pymysql + +In the Kingbird root folder, where you can find the source code of Kingbird, +generate the configuration sample file for Kingbird: + +.. code-block:: bash + + oslo-config-generator --config-file=./tools/config-generator.conf + +prepare the folder used for cache, log and configuration for Kingbird: + +.. code-block:: bash + + sudo rm -rf /var/cache/kingbird + sudo mkdir -p /var/cache/kingbird + sudo chown `whoami` /var/cache/kingbird + sudo rm -rf /var/log/kingbird + sudo mkdir -p /var/log/kingbird + sudo chown `whoami` /var/log/kingbird + sudo rm -rf /etc/kingbird + sudo mkdir -p /etc/kingbird + sudo chown `whoami` /etc/kingbird + +Copy the sample configuration to the configuration folder /etc/kingbird: + +.. code-block:: bash + + cp etc/kingbird/kingbird.conf.sample /etc/kingbird/kingbird.conf + +Before editing the configuration file, prepare the database info for Kingbird. + +.. code-block:: bash + + mysql -uroot -e "CREATE DATABASE $kb_db CHARACTER SET utf8;" + mysql -uroot -e "GRANT ALL PRIVILEGES ON $kb_db.* TO '$kb_db_user'@'%' IDENTIFIED BY '$kb_db_pwd';" + +For example, the following command will create database "kingbird", and grant the +privilege for the db user "kingbird" with password "password": + +.. code-block:: bash + + mysql -uroot -e "CREATE DATABASE kingbird CHARACTER SET utf8;" + mysql -uroot -e "GRANT ALL PRIVILEGES ON kingbird.* TO 'kingbird'@'%' IDENTIFIED BY 'password';" + +Create the service user in OpenStack: + +.. code-block:: bash + + source openrc admin admin + openstack user create --project=service --password=$kb_svc_pwd $kb_svc_user + openstack role add --user=$kb_svc_user --project=service admin + +For example, the following command will create service user "kingbird", +and grant the user "kingbird" with password "password" the role of admin +in service project: + +.. code-block:: bash + + source openrc admin admin + openstack user create --project=service --password=password kingbird + openstack role add --user=kingbird --project=service admin + + + +Then edit the configuration file for Kingbird: + +.. code-block:: bash + + vim /etc/kingbird/kingbird.conf + +By default, the bind_host of kingbird-api is local_host(127.0.0.1), and the +port for the service is 8118, you can leave it as the default if no port +conflict happened. + +Please replace the address of Kingbird service "127.0.0.1" which is mentioned +below to the address you get from OpenStack Kingbird endpoint. + +To make the Kingbird work normally, you have to edit these configuration +items. The [cache] section is used by kingbird engine to access the quota +information of Nova, Cinder, Neutron in each region, replace the +auth_uri to the keystone service in your environment, +especially if the keystone service is not located in the same node, and +also for the account to access the Nova, Cinder, Neutron in each region, +in the following configuration, user "admin" with password "password" of +the tenant "admin" is configured to access other Nova, Cinder, Neutron in +each region: + +.. code-block:: bash + + [cache] + auth_uri = http://127.0.0.1:5000/v3 + admin_tenant = admin + admin_password = password + admin_username = admin + +Configure the database section with the service user "kingbird" and its +password, to access database "kingbird". For detailed database section +configuration, please refer to http://docs.openstack.org/developer/oslo.db/opts.html, +and change the following configuration accordingly based on your +environment. + +.. code-block:: bash + + [database] + connection = mysql+pymysql://$kb_db_user:$kb_db_pwd@127.0.0.1/$kb_db?charset=utf8 + +For example, if the database is "kingbird", and the db user "kingbird" with +password "password", then the configuration is as following: + +.. code-block:: bash + + [database] + connection = mysql+pymysql://kingbird:password@127.0.0.1/kingbird?charset=utf8 + +The [keystone_authtoken] section is used by keystonemiddleware for token +validation during the API request to the kingbird-api, please refer to +http://docs.openstack.org/developer/keystonemiddleware/middlewarearchitecture.html +on how to configure the keystone_authtoken section for the keystonemiddleware +in detail, and change the following configuration accordingly based on your +environment: + +*please specify the region_name where you want the token will be validated if the +KeyStone is deployed in multiple regions* + +.. code-block:: bash + + [keystone_authtoken] + signing_dir = /var/cache/kingbird + cafile = /opt/stack/data/ca-bundle.pem + auth_uri = http://127.0.0.1:5000/v3 + project_domain_name = Default + project_name = service + user_domain_name = Default + password = $kb_svc_pwd + username = $kb_svc_user + auth_url = http://127.0.0.1:35357/v3 + auth_type = password + region_name = RegionOne + +For example, if the service user is "kingbird, and the password for the user +is "password", then the configuration will look like this: + +.. code-block:: bash + + [keystone_authtoken] + signing_dir = /var/cache/kingbird + cafile = /opt/stack/data/ca-bundle.pem + auth_uri = http://127.0.0.1:5000/v3 + project_domain_name = Default + project_name = service + user_domain_name = Default + password = password + username = kingbird + auth_url = http://127.0.0.1:35357/v3 + auth_type = password + region_name = RegionOne + + +And also configure the message bus connection, you can refer to the message +bus configuration in Nova, Cinder, Neutron configuration file. + +.. code-block:: bash + + [DEFAULT] + transport_url = rabbit://stackrabbit:password@127.0.0.1:5672/ + +After these basic configuration items configured, now the database schema of +"kingbird" should be created: + +.. code-block:: bash + + python kingbird/cmd/manage.py --config-file=/etc/kingbird/kingbird.conf db_sync + +And create the service and endpoint for Kingbird, please change the endpoint url +according to your cloud planning: + +.. code-block:: bash + + openstack service create --name=kingbird synchronization + openstack endpoint create --region=RegionOne kingbird public http://127.0.0.1:8118/v1.0 + openstack endpoint create --region=RegionOne kingbird admin http://127.0.0.1:8118/v1.0 + openstack endpoint create --region=RegionOne kingbird internal http://127.0.0.1:8118/v1.0 + +Now it's ready to run kingbird-api and kingbird-engine: + +.. code-block:: bash + + nohup python kingbird/cmd/api.py --config-file=/etc/kingbird/kingbird.conf & + nohup python kingbird/cmd/engine.py --config-file=/etc/kingbird/kingbird.conf & + +Run the following command to check whether kingbird-api and kingbird-engine +are running: + +.. code-block:: bash + + ps aux|grep python + + +Post-installation activities +---------------------------- + +Run the following commands to check whether kingbird-api is serving, please +replace $mytoken to the token you get from "openstack token issue": + +.. code-block:: bash + + openstack token issue + curl -H "Content-Type: application/json" -H "X-Auth-Token: $mytoken" \ + http://127.0.0.1:8118/ + +If the response looks like following: {"versions": [{"status": "CURRENT", +"updated": "2016-03-07", "id": "v1.0", "links": [{"href": +"http://127.0.0.1:8118/v1.0/", "rel": "self"}]}]}, +then that means the kingbird-api is working normally. + +Run the following commands to check whether kingbird-engine is serving, please +replace $mytoken to the token you get from "openstack token issue", and the +$admin_project_id to the admin project id in your environment: + +.. code-block:: bash + + curl -H "Content-Type: application/json" -H "X-Auth-Token: $mytoken" \ + -X PUT \ + http://127.0.0.1:8118/v1.0/$admin_project_id/os-quota-sets/$admin_project_id/sync + +If the response looks like following: "triggered quota sync for +0320065092b14f388af54c5bd18ab5da", then that means the kingbird-engine +is working normally. diff --git a/docs/release/overview/index.rst b/docs/release/overview/index.rst new file mode 100644 index 0000000..716f5a0 --- /dev/null +++ b/docs/release/overview/index.rst @@ -0,0 +1,12 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 + +*********************** +Multisite Release Notes +*********************** + +.. toctree:: + :numbered: + :maxdepth: 4 + + multisite.release.notes.rst diff --git a/docs/release/overview/multisite.release.notes.rst b/docs/release/overview/multisite.release.notes.rst new file mode 100644 index 0000000..85b9561 --- /dev/null +++ b/docs/release/overview/multisite.release.notes.rst @@ -0,0 +1,11 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 + +Multisite is to identify the requirements and gaps for the VIM(OpenStack) +to support multi-site NFV cloud. + +The documentation of requirements, installation, configuration and usage +guide for multi-site and Kingbird are provided. + +For Kingbird service, known bugs are registered at +https://bugs.launchpad.net/kingbird. diff --git a/docs/release/userguide/index.rst b/docs/release/userguide/index.rst new file mode 100644 index 0000000..2726184 --- /dev/null +++ b/docs/release/userguide/index.rst @@ -0,0 +1,15 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 +.. (c) Chaoyi Huang, Huawei Technologies Co., Ltd. + +************************** +Multisite Admin User Guide +************************** + +.. toctree:: + :numbered: + :maxdepth: 4 + + multisite.admin.usage.rst + multisite.kingbird.usage.rst + multisite.tricircle.usage.rst diff --git a/docs/release/userguide/multisite.admin.usage.rst b/docs/release/userguide/multisite.admin.usage.rst new file mode 100644 index 0000000..544c9b1 --- /dev/null +++ b/docs/release/userguide/multisite.admin.usage.rst @@ -0,0 +1,365 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 + +Multisite identity service management +===================================== + +Goal +---- + +A user should, using a single authentication point be able to manage virtual +resources spread over multiple OpenStack regions. + +Token Format +------------ + +There are 3 types of token format supported by OpenStack KeyStone + + * **FERNET** + * **UUID** + * **PKI/PKIZ** + +It's very important to understand these token format before we begin the +mutltisite identity service management. Please refer to the OpenStack +official site for the identity management. +http://docs.openstack.org/admin-guide-cloud/identity_management.html + +Please note that PKI/PKIZ token format has been deprecated. + +Key consideration in multisite scenario +--------------------------------------- + +A user is provided with a single authentication URL to the Identity (Keystone) +service. Using that URL, the user authenticates with Keystone by +requesting a token typically using username/password credentials. Keystone +server validates the credentials, possibly with an external LDAP/AD server and +returns a token to the user. The user sends a request to a service in a +selected region including the token. Now the service in the region, say Nova +needs to validate the token. The service uses its configured keystone endpoint +and service credentials to request token validation from Keystone. After the +token is validated by KeyStone, the user is authorized to use the service. + +The key considerations for token validation in multisite scenario are: + * Site level failure: impact on authN and authZ shoulde be as minimal as + possible + * Scalable: as more and more sites added, no bottleneck in token validation + * Amount of inter region traffic: should be kept as little as possible + +Hence, Keystone token validation should preferably be done in the same +region as the service itself. + +The challenge to distribute KeyStone service into each region is the KeyStone +backend. Different token format has different data persisted in the backend. + +* Fernet: Tokens are non persistent cryptographic based tokens and validated + online by the Keystone service. Fernet tokens are more lightweight + than PKI tokens and have a fixed size. Fernet tokens require Keystone + deployed in a distributed manner, again to avoid inter region traffic. The + data synchronization cost for the Keystone backend is smaller due to the non- + persisted token. + +* UUID: UUID tokens have a fixed size. Tokens are persistently stored and + create a lot of database traffic, the persistence of token is for the revoke + purpose. UUID tokens are validated online by Keystone, call to service will + request keystone for token validation. Keystone can become a + bottleneck in a large system. Due to this, UUID token type is not suitable + for use in multi region clouds, no matter the Keystone database + replicates or not. + +Cryptographic tokens bring new (compared to UUID tokens) issues/use-cases +like key rotation, certificate revocation. Key management is out of scope for +this use case. + +Database deployment as the backend for KeyStone service +------------------------------------------------------ + +Database replication: + - Master/slave asynchronous: supported by the database server itself + (mysql/mariadb etc), works over WAN, it's more scalable. But only master will + provide write functionality, domain/project/role provisioning. + - Multi master synchronous: Galera(others like percona), not so scalable, + for multi-master writing, and need more parameter tunning for WAN latency.It + can provide the capability for limited multi-sites multi-write + function for distributed KeyStone service. + - Symmetrical/asymmetrical: data replicated to all regions or a subset, + in the latter case it means some regions needs to access Keystone in another + region. + +Database server sharing: +In an OpenStack controller, normally many databases from different +services are provided from the same database server instance. For HA reasons, +the database server is usually synchronously replicated to a few other nodes +(controllers) to form a cluster. Note that _all_ database are replicated in +this case, for example when Galera sync repl is used. + +Only the Keystone database can be replicated to other sites. Replicating +databases for other services will cause those services to get of out sync and +malfunction. + +Since only the Keystone database is to be replicated sync. or async. to another +region/site, it's better to deploy Keystone database into its own +database server with extra networking requirement, cluster or replication +configuration. How to support this by installer is out of scope. + +The database server can be shared when async master/slave replication is +used, if global transaction identifiers GTID is enabled. + +Deployment options +------------------ + +**Distributed KeyStone service with Fernet token** + +Fernet token is a very new format, and just introduced recently,the biggest +gain for this token format is :1) lightweight, size is small to be carried in +the API request, not like PKI token( as the sites increased, the endpoint-list +will grows and the token size is too long to carry in the API request) 2) no +token persistence, this also make the DB not changed too much and with light +weight data size (just project, Role, domain, endpoint etc). The drawback for +the Fernet token is that token has to be validated by KeyStone for each API +request. + +This makes that the DB of KeyStone can work as a cluster in multisite (for +example, using MySQL galera cluster). That means install KeyStone API server in +each site, but share the same the backend DB cluster.Because the DB cluster +will synchronize data in real time to multisite, all KeyStone server can see +the same data. + +Because each site with KeyStone installed, and all data kept same, +therefore all token validation could be done locally in the same site. + +The challenge for this solution is how many sites the DB cluster can +support. Question is aksed to MySQL galera developers, their answer is that no +number/distance/network latency limitation in the code. But in the practice, +they have seen a case to use MySQL cluster in 5 data centers, each data centers +with 3 nodes. + +This solution will be very good for limited sites which the DB cluster can +cover very well. + +**Distributed KeyStone service with Fernet token + Async replication (star-mode)** + +One master KeyStone cluster with Fernet token in one or two sites(for site +level high availability purpose), other sites will be installed with at least +2 slave nodes where the node is configured with DB async replication from the +master cluster members. The async. replication data source is better to be +from different member of the master cluster, if there are two sites for the +KeyStone cluster, it'll be better that source members for async. replication +are located in different site. + +Only the master cluster nodes are allowed to write, other slave nodes +waiting for replication from the master cluster member( very little delay). + +Pros: + * Deploy database cluster in the master sites is to provide more master + nodes, in order to provide more slaves could be done with async. replication + in parallel. Two sites for the master cluster is to provide higher + reliability (site level) for writing request, but reduce the maintaince + challenge at the same time by limiting the cluster spreading over too many + sites. + * Multi-slaves in other sites is because of the slave has no knowledge of + other slaves, so easy to manage multi-slaves in one site than a cluster, and + multi-slaves work independently but provide multi-instance redundancy(like a + cluster, but independent). + +Cons: + * Need to be aware of the chanllenge of key distribution and rotation + for Fernet token. + +Multisite VNF Geo site disaster recovery +======================================== + +Goal +---- + +A VNF (telecom application) should, be able to restore in another site for +catastrophic failures happened. + +Key consideration in multisite scenario +--------------------------------------- + +Geo site disaster recovery is to deal with more catastrophic failures +(flood, earthquake, propagating software fault), and that loss of calls, or +even temporary loss of service, is acceptable. It is also seems more common +to accept/expect manual / administrator intervene into drive the process, not +least because you don’t want to trigger the transfer by mistake. + +In terms of coordination/replication or backup/restore between geographic +sites, discussion often (but not always) seems to focus on limited application +level data/config replication, as opposed to replication backup/restore between +of cloud infrastructure between different sites. + +And finally, the lack of a requirement to do fast media transfer (without +resignalling) generally removes the need for special networking behavior, with +slower DNS-style redirection being acceptable. + +Here is more concerns about cloud infrastructure level capability to +support VNF geo site disaster recovery + +Option1, Consistency application backup +--------------------------------------- + +The disater recovery process will work like this: + +1) DR(Geo site disaster recovery )software get the volumes for each VM + in the VNF from Nova +2) DR software call Nova quiesce API to quarantee quiecing VMs in desired order +3) DR software takes snapshots of these volumes in Cinder (NOTE: Because + storage often provides fast snapshot, so the duration between quiece and + unquiece is a short interval) +4) DR software call Nova unquiece API to unquiece VMs of the VNF in reverse order +5) DR software create volumes from the snapshots just taken in Cinder +6) DR software create backup (incremental) for these volumes to remote + backup storage ( swift or ceph, or.. ) in Cinder +7) If this site failed, + 1) DR software restore these backup volumes in remote Cinder in the backup site. + 2) DR software boot VMs from bootable volumes from the remote Cinder in + the backup site and attach the regarding data volumes. + +Note: Quiesce/Unquiesce spec was approved in Mitaka, but code not get merged in +time, https://blueprints.launchpad.net/nova/+spec/expose-quiesce-unquiesce-api +The spec was rejected in Newton when it was reproposed: +https://review.openstack.org/#/c/295595/. So this option will not work any more. + +Option2, Vitrual Machine Snapshot +--------------------------------- +1) DR software create VM snapshot in Nova +2) Nova quiece the VM internally + (NOTE: The upper level application or DR software should take care of + avoiding infra level outage induced VNF outage) +3) Nova create image in Glance +4) Nova create a snapshot of the VM, including volumes +5) If the VM is volume backed VM, then create volume snapshot in Cinder +5) No image uploaded to glance, but add the snapshot in the meta data of the + image in Glance +6) DR software to get the snapshot information from the Glance +7) DR software create volumes from these snapshots +9) DR software create backup (incremental) for these volumes to backup storage + ( swift or ceph, or.. ) in Cinder +10) If this site failed, + 1) DR software restore these backup volumes to Cinder in the backup site. + 2) DR software boot vm from bootable volume from Cinder in the backup site + and attach the data volumes. + +This option only provides single VM level consistency disaster recovery. + +This feature is already available in current OPNFV release. + +Option3, Consistency volume replication +--------------------------------------- +1) DR software creates datastore (Block/Cinder, Object/Swift, App Custom + storage) with replication enabled at the relevant scope, for use to + selectively backup/replicate desire data to GR backup site +2) DR software get the reference of storage in the remote site storage +3) If primary site failed, + 1) DR software managing recovery in backup site gets references to relevant + storage and passes to new software instances + 2) Software attaches (or has attached) replicated storage, in the case of + volumes promoting to writable. + +Pros: + * Replication will be done in the storage level automatically, no need to + create backup regularly, for example, daily. + * Application selection of limited amount of data to replicate reduces + risk of replicating failed state and generates less overhear. + * Type of replication and model (active/backup, active/active, etc) can + be tailored to application needs + +Cons: + * Applications need to be designed with support in mind, including both + selection of data to be replicated and consideration of consistency + * "Standard" support in Openstack for Disaster Recovery currently fairly + limited, though active work in this area. + +Note: Volume replication v2.1 support project level replication. + + +VNF high availability across VIM +================================ + +Goal +---- + +A VNF (telecom application) should, be able to realize high availability +deloyment across OpenStack instances. + +Key consideration in multisite scenario +--------------------------------------- + +Most of telecom applications have already been designed as +Active-Standby/Active-Active/N-Way to achieve high availability +(99.999%, corresponds to 5.26 minutes of unplanned downtime in a year), +typically state replication or heart beat between +Active-Active/Active-Active/N-Way (directly or via replicated database +services, or via private designed message format) are required. + +We have to accept the currently limited availability ( 99.99%) of a +given OpenStack instance, and intend to provide the availability of the +telecom application by spreading its function across multiple OpenStack +instances.To help with this, many people appear willing to provide multiple +“independent” OpenStack instances in a single geographic site, with special +networking (L2/L3) between clouds in that physical site. + +The telecom application often has different networking plane for different +purpose: + +1) external network plane: using for communication with other telecom + application. + +2) components inter-communication plane: one VNF often consisted of several + components, this plane is designed for components inter-communication with + each other + +3) backup plane: this plane is used for the heart beat or state replication + between the component's active/standby or active/active or N-way cluster. + +4) management plane: this plane is mainly for the management purpose, like + configuration + +Generally these planes are separated with each other. And for legacy telecom +application, each internal plane will have its fixed or flexble IP addressing +plan. + +There are some interesting/hard requirements on the networking (L2/L3) +between OpenStack instances, at lease the backup plane across different +OpenStack instances: + +To make the VNF can work with HA mode across different OpenStack instances in +one site (but not limited to), need to support at lease the backup plane across +different OpenStack instances: + +1) L2 networking across OpenStack instance for heartbeat or state replication. +Overlay L2 networking or shared L2 provider networks can work as the backup +plance for heartbeat or state replication. Overlay L2 network is preferred, +the reason is: + + a. Support legacy compatibility: Some telecom app with built-in internal L2 + network, for easy to move these app to VNF, it would be better to provide + L2 network. + b. Isolated L2 network will simplify the security management between + different network planes. + c. Easy to support IP/mac floating across OpenStack. + d. Support IP overlapping: multiple VNFs may have overlaping IP address for + cross OpenStack instance networking. + +Therefore, over L2 networking across Neutron feature is required in OpenStack. + +2) L3 networking across OpenStack instance for heartbeat or state replication. +For L3 networking, we can leverage the floating IP provided in current +Neutron, or use VPN or BGPVPN(networking-bgpvpn) to setup the connection. + +L3 networking to support the VNF HA will consume more resources and need to +take more security factors into consideration, this make the networking +more complex. And L3 networking is also not possible to provide IP floating +across OpenStack instances. + +3) The IP address used for VNF to connect with other VNFs should be able to be +floating cross OpenStack instance. For example, if the master failed, the IP +address should be used in the standby which is running in another OpenStack +instance. There are some method like VRRP/GARP etc can help the movement of the +external IP, so no new feature will be added to OpenStack. + +Several projects are addressing the networking requirements, deployment should +consider the factors mentioned above. + * Tricircle: https://github.com/openstack/tricircle/ + * Networking-BGPVPN: https://github.com/openstack/networking-bgpvpn/ + * VPNaaS: https://github.com/openstack/neutron-vpnaas diff --git a/docs/release/userguide/multisite.kingbird.usage.rst b/docs/release/userguide/multisite.kingbird.usage.rst new file mode 100644 index 0000000..e9ead90 --- /dev/null +++ b/docs/release/userguide/multisite.kingbird.usage.rst @@ -0,0 +1,349 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 + +============================= +Multisite.Kingbird user guide +============================= + +Quota management for OpenStack multi-region deployments +------------------------------------------------------- +Kingbird is centralized synchronization service for multi-region OpenStack +deployments. In OPNFV Colorado release, Kingbird provides centralized quota +management feature. Administrator can set quota per project based in Kingbird +and sync the quota limit to multi-region OpenStack periodiclly or on-demand. +The tenant can check the total quota limit and usage from Kingbird for all +regions. Administrator can also manage the default quota by quota class +setting. + +Following quota items are supported to be managed in Kingbird: + +- **instances**: Number of instances allowed per project. +- **cores**: Number of instance cores allowed per project. +- **ram**: Megabytes of instance RAM allowed per project. +- **metadata_items**: Number of metadata items allowed per instance. +- **key_pairs**: Number of key pairs per user. +- **fixed_ips**: Number of fixed IPs allowed per project, + valid if Nova Network is used. +- **security_groups**: Number of security groups per project, + valid if Nova Network is used. +- **floating_ips**: Number of floating IPs allowed per project, + valid if Nova Network is used. +- **network**: Number of networks allowed per project, + valid if Neutron is used. +- **subnet**: Number of subnets allowed per project, + valid if Neutron is used. +- **port**: Number of ports allowed per project, + valid if Neutron is used. +- **security_group**: Number of security groups allowed per project, + valid if Neutron is used. +- **security_group_rule**: Number of security group rules allowed per project, + valid if Neutron is used. +- **router**: Number of routers allowed per project, + valid if Neutron is used. +- **floatingip**: Number of floating IPs allowed per project, + valid if Neutron is used. +- **volumes**: Number of volumes allowed per project. +- **snapshots**: Number of snapshots allowed per project. +- **gigabytes**: Total amount of storage, in gigabytes, allowed for volumes + and snapshots per project. +- **backups**: Number of volume backups allowed per project. +- **backup_gigabytes**: Total amount of storage, in gigabytes, allowed for volume + backups per project. + +Key pair is the only resource type supported in resource synchronization. + +Only restful APIs are provided for Kingbird in Colorado release, so curl or +other http client can be used to call Kingbird API. + +Before use the following command, get token, project id, and kingbird service +endpoint first. Use $kb_token to repesent the token, and $admin_tenant_id as +administrator project_id, and $tenant_id as the target project_id for quota +management and $kb_ip_addr for the kingbird service endpoint ip address. + +Note: +To view all tenants (projects), run: + + .. code-block:: bash + + openstack project list + +To get token, run: + + .. code-block:: bash + + openstack token issue + +To get Kingbird service endpoint, run: + + .. code-block:: bash + + openstack endpoint list + +Quota Management API +-------------------- + +1. Update global limit for a tenant + + Use python-kingbirdclient: + + .. code-block:: bash + + kingbird quota update b8eea2ceda4c47f1906fda7e7152a322 --port 10 --security_groups 10 + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + -X PUT \ + -d '{"quota_set":{"cores": 10,"ram": 51200, "metadata_items": 100,"key_pairs": 100, "network":20,"security_group": 20,"security_group_rule": 20}}' \ + http://$kb_ip_addr:8118/v1.0/$admin_tenant_id/os-quota-sets/$tenant_id + +2. Get global limit for a tenant + + Use python-kingbirdclient: + + .. code-block:: bash + + kingbird quota show --tenant $tenant_id + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + http://$kb_ip_addr:8118/v1.0/$admin_tenant_id/os-quota-sets/$tenant_id + +3. A tenant can also get the global limit by himself + + Use python-kingbirdclient: + + .. code-block:: bash + + kingbird quota show + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + http://$kb_ip_addr:8118/v1.0/$tenant_id/os-quota-sets/$tenant_id + +4. Get defaults limits + + Use python-kingbirdclient: + + .. code-block:: bash + + kingbird quota defaults + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + http://$kb_ip_addr:8118/v1.0/$admin_tenant_id/os-quota-sets/defaults + +5. Get total usage for a tenant + + Use python-kingbirdclient: + + .. code-block:: bash + + kingbird quota detail --tenant $tenant_id + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + -X GET \ + http://$kb_ip_addr:8118/v1.0/$admin_tenant_id/os-quota-sets/$tenant_id/detail + +6. A tenant can also get the total usage by himself + + Use python-kingbirdclient: + + .. code-block:: bash + + kingbird quota detail + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + -X GET \ + http://$kb_ip_addr:8118/v1.0/$tenant_id/os-quota-sets/$tenant_id/detail + +7. On demand quota sync + + Use python-kingbirdclient: + + .. code-block:: bash + + kingbird quota sync $tenant_id + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + -X PUT \ + http://$kb_ip_addr:8118/v1.0/$admin_tenant_id/os-quota-sets/$tenant_id/sync + + +8. Delete specific global limit for a tenant + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + -X DELETE \ + -d '{"quota_set": [ "cores", "ram"]}' \ + http://$kb_ip_addr:8118/v1.0/$admin_tenant_id/os-quota-sets/$tenant_id + +9. Delete all kingbird global limit for a tenant + + Use python-kingbirdclient: + + .. code-block:: bash + + kingbird quota delete $tenant_id + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + -X DELETE \ + http://$kb_ip_addr:8118/v1.0/$admin_tenant_id/os-quota-sets/$tenant_id + + +Quota Class API +--------------- + +1. Update default quota class + + Use python-kingbirdclient: + + .. code-block:: bash + + kingbird quota-class update --port 10 --security_groups 10 + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + -X PUT \ + -d '{"quota_class_set":{"cores": 100, "network":50,"security_group": 50,"security_group_rule": 50}}' \ + http://$kb_ip_addr:8118/v1.0/$admin_tenant_id/os-quota-class-sets/default + +2. Get default quota class + + Use python-kingbirdclient: + + .. code-block:: bash + + kingbird quota-class show default + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + http://$kb_ip_addr:8118/v1.0/$admin_tenant_id/os-quota-class-sets/default + +3. Delete default quota class + + Use python-kingbirdclient: + + .. code-block:: bash + + kingbird quota-class delete default + + Use curl: + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + -X DELETE \ + http://$kb_ip_addr:8118/v1.0/$admin_tenant_id/os-quota-class-sets/default + + +Resource Synchronization API +----------------------------- + +1. Create synchronization job + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + -X POST -d \ + '{"resource_set":{"resources": [""],"force":,"resource_type": "keypair","source": <"Source_Region">,"target": [<"List_of_target_regions">]}}' \ + http://$kb_ip_addr:8118/v1.0/$tenant_id/os-sync + +2. Get synchronization job + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + http://$kb_ip_addr:8118/v1.0/$tenant_id/os-sync/ + +3. Get active synchronization job + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + http://$kb_ip_addr:8118/v1.0/$tenant_id/os-sync/active + +4. Get detail information of a synchronization job + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + http://$kb_ip_addr:8118/v1.0/$tenant_id/os-sync/$job_id + +5. Delete a synchronization job + + .. code-block:: bash + + curl \ + -H "Content-Type: application/json" \ + -H "X-Auth-Token: $kb_token" \ + -X DELETE \ + http://$kb_ip_addr:8118/v1.0/$tenant_id/os-sync/job_id diff --git a/docs/release/userguide/multisite.tricircle.usage.rst b/docs/release/userguide/multisite.tricircle.usage.rst new file mode 100644 index 0000000..d42f5b0 --- /dev/null +++ b/docs/release/userguide/multisite.tricircle.usage.rst @@ -0,0 +1,13 @@ +.. This work is licensed under a Creative Commons Attribution 4.0 International License. +.. http://creativecommons.org/licenses/by/4.0 + +============================== +Multisite.Tricircle user guide +============================== + +Tricircle is one OpenStack big-tent project. All user guide related documents +could be found from OpenStack website: + * Developer Guide: http://docs.openstack.org/developer/tricircle/ + * Installation Guide: http://docs.openstack.org/developer/tricircle/installation-guide.html + * Configuration Guide: http://docs.openstack.org/developer/tricircle/configuration.html + * Networking Guide: http://docs.openstack.org/developer/tricircle/networking-guide.html -- cgit 1.2.3-korg