- OpenStack-announce - lists.openstack.org

[release][nova] python-novaclient 3.4.0 release (newton)
by no-reply＠openstack.org 07 Apr '16

07 Apr '16

We are jubilant to announce the release of: python-novaclient 3.4.0: Client library for OpenStack Compute API This release is part of the newton release series. With source available at: https://git.openstack.org/cgit/openstack/python-novaclient With package available at: https://pypi.python.org/pypi/python-novaclient Please report issues through launchpad: https://bugs.launchpad.net/python-novaclient For more details, please see below. Changes in python-novaclient 3.3.0..3.4.0 ----------------------------------------- 90fbbb2 Fix host-evacuate-live for 2.25 microversion 92bbae6 Use keystoneclient python bindings for testing 7e7e552 Removed unused Oslo Incubator code c5272b3 Adding tox support for bandit bf68a0c aggregate-details changed to aggregate-show fa377e7 Handle error response for webob>=1.6.0 f6bcac4 Validate shutdown value of --block-device 7cefdd3 Add changes-since support when list servers a7bffe8 Update reno for stable/mitaka 97d63e3 Wrap interface_list by proper resource class b80d8cb nova add-secgroup help updated with secgroup id 6803201 Remove additional 'timeout' element d8e2f0c Make it clear that host-servers-migrate is a cold migration c2ec89d Return a less dramatic message for public flavors. a571ebd Remove unused code in tests/unit/v2/fakes.py 0ee5aea Fix a typo in novaclient/v2/hosts.py 8691eec Remove an unused method in novaclient/shell.py 3a538f6 Overwrite Usage class's get() function 471e08a Remove console expectation from NMI tests 16c36ce Use novaclient/utils directly and remove openstack/common (4/4) 49410c0 Use novaclient/utils directly and remove openstack/common a787d33 Use novaclient/utils directly and remove openstack/common (2/4) 93913c8 Use novaclient/utils directly and remove openstack/common (1/4) 54ed915 The novaclient Python API doc keystoneauth example fixed 68c16f8 Prepare to move extension into core plugin 345469d Deprecate run_test.sh Diffstat (except docs and test files) ------------------------------------- novaclient/client.py | 8 +- novaclient/exceptions.py | 18 +- novaclient/openstack/__init__.py | 0 novaclient/openstack/common/__init__.py | 0 novaclient/openstack/common/cliutils.py | 18 - novaclient/shell.py | 17 +- .../functional/v2/legacy/test_flavor_access.py | 6 +- novaclient/v2/contrib/baremetal.py | 37 +- novaclient/v2/contrib/cells.py | 5 +- novaclient/v2/contrib/deferred_delete.py | 5 +- novaclient/v2/contrib/host_evacuate.py | 7 +- novaclient/v2/contrib/host_evacuate_live.py | 32 +- novaclient/v2/contrib/host_servers_migrate.py | 7 +- novaclient/v2/contrib/instance_action.py | 11 +- novaclient/v2/contrib/metadata_extensions.py | 8 +- novaclient/v2/contrib/migrations.py | 7 +- novaclient/v2/contrib/tenant_networks.py | 17 +- novaclient/v2/hosts.py | 2 +- novaclient/v2/servers.py | 11 +- novaclient/v2/shell.py | 999 +++++++++++---------- novaclient/v2/usage.py | 15 + releasenotes/source/index.rst | 1 + releasenotes/source/mitaka.rst | 6 + test-requirements.txt | 1 + tools/install_venv_common.py | 172 ---- tox.ini | 5 +- 37 files changed, 851 insertions(+), 1250 deletions(-) Requirements updates -------------------- diff --git a/test-requirements.txt b/test-requirements.txt index 6a301d0..8dd1c96 100644 --- a/test-requirements.txt +++ b/test-requirements.txt @@ -5,0 +6 @@ hacking<0.11,>=0.10.0 +bandit>=0.17.3 # Apache-2.0

1 0

OpenStack Mitaka is officially released!
by Doug Hellmann 07 Apr '16

07 Apr '16

Hello OpenStack community, I'm excited to announce the final releases for the components of OpenStack Mitaka, which conclude the 6-month Mitaka development cycle. You will find a complete list of all components, their latest versions, and links to individual project release notes documents listed on the new release site: http://releases.openstack.org/mitaka/ Congratulations to all of the teams who have contributed to this release! I also want to extend a big thank you to AJAeger and jhesketh for their help with build issues today, and to ttx and dims for their help with release work all this cycle and especially over the last couple of weeks. Thank you! Our next production cycle, Newton, has already started. We will meet in Austin April 25-29 at the Newton Design Summit to plan the work for the upcoming cycle. I hope to see you there! Doug Doug

1 0

[release] Sorry for the recent flurry of announce emails
by Doug Hellmann 07 Apr '16

07 Apr '16

Hi! We are sorry for the recent flurry of announce emails. We are preparing for the final Mitaka release and an error in the build script led to prematurely sending erroneous release announcements. While the version numbers are correct, those announcements contain only partial release notes (usually only mentioning post-RC1 changes). A more complete Mitaka announcement is going to be sent in the coming hours, once we finalize the mitaka release notes. Thanks for your understanding and patience, Doug

1 0

[release][trove] trove-dashboard 6.0.0 release (mitaka)
by no-reply＠openstack.org 07 Apr '16

07 Apr '16

We are delighted to announce the release of: trove-dashboard 6.0.0: Trove Management Dashboard This release is part of the mitaka release series. For more details, please see below. Changes in trove-dashboard 6.0.0.0rc1..6.0.0 -------------------------------------------- 55a277f Imported Translations from Zanata f02ae31 Imported Translations from Zanata 3f71ea5 [Django 1.9] Remove load url from future be04f4c Imported Translations from Zanata 46faf10 Imported Translations from Zanata 5cbff9e Update .gitreview for stable/mitaka bb9f4d3 Temporarily disable failing test Diffstat (except docs and test files) ------------------------------------- .gitreview | 1 + .../templates/database_clusters/_launch.html | 1 - .../database_clusters/_reset_password.html | 1 - trove_dashboard/locale/de/LC_MESSAGES/django.po | 14 +- trove_dashboard/locale/fr/LC_MESSAGES/django.po | 9 +- trove_dashboard/locale/ko_KR/LC_MESSAGES/django.po | 5 +- trove_dashboard/locale/ru/LC_MESSAGES/django.po | 1131 ++++++++++++++++++++ trove_dashboard/locale/zh_CN/LC_MESSAGES/django.po | 13 +- 9 files changed, 1166 insertions(+), 11 deletions(-)

1 0

[release][senlin] senlin 1.0.0 release (mitaka)
by no-reply＠openstack.org 07 Apr '16

07 Apr '16

We are delighted to announce the release of: senlin 1.0.0: OpenStack Clustering This release is part of the mitaka release series. Please report issues through launchpad: https://bugs.launchpad.net/senlin For more details, please see below. 1.0.0 ^^^^^ New Features ************ * We no longer support soft-delete in senlin database. Marking an entity as soft-deleted is causing more problems than brining any convenience. * Action list now can be filtered by its 'status' property. * Added support of multi-tenancy for actions. * Added senlin.policy.affinity-v1.0 which can be used to control how VM servers are placed based on nova servergroup settings. * New actions for checking and recovering nodes/clusters are added. * Added new APIs for cluster/node check and recover. * Added support to limit number of clusters per project. * Clusters now have a new 'RESIZING' status when its scale is being changed. * The builtin deletion policy can handle cross-region and cross-zone nodes selection. * Supporting engine status scheck, with senlin-manage command. * Event list can now be filtered by its 'level' property. * Adde support to multi-tenancy for event resources. * Profile types and policy types are explicitly versioned now. We have the version string appended to the type names for easier detection. * New health check daemon is introduced to do periodical cluster status checking. It collaborates with health policy on cluster monitoring. * Improved action scheduler so that it can pick an action that is READY to be executed from DB. * Added LBaaS health monitor support to load-balancing policy v1.0. * Added support to steal a lock from dead engine. * Added support to multi-tenancy (aka. project_safe checking) when finding resources. * Use 'sort' instead of 'sort_keys' and 'sort_dir' for object sorting. * Nova server proflie now supports block device mapping v2 (BDMv2). * Enabled update to the 'flavor' of a nova server profile. * Enabled update to the 'name' of a nova server profile. * Added profile property checking regarding whether they are updatable. * Support to 'senlin-manage purge_deleted <age> [<unit>]' is added. * New abstraction 'receiver' has been added as a generic way to notify the senlin engine that something interesting has happended. * An experimental pollicy for placing nodes across multiple regions. * Removed support to 'trigger' abstraction. * Make sure 'spec' of a profile is immutable after a profile object is created. The only way to "update" a profile is to create a new one. * Added command 'senlin-manage service list' to show the status of engine. * Added command 'senlin-manage service clean' to clean the dead engine records. * Add support to update image property of a Nova server. * Added support to updating network properties of a nova server. * Both image ID and image name are supported when creating os.nova.server profile. Upgrade Notes ************* * Database tables have to be recreated as we have removed soft- delete support from both the DB layer and the engine layer. * Senlin API has removed 'tenant_id' from its endpoint. This means users have to recreate their keystone endpoints if they have an old installation. * Webhook abstraction is removed. New usage model of webhooks is through the 'receiver' abstraction. * Node actions NODE_JOIN and NODE_LEAVE are removed from API surface. * Removed cluster policy enable/disable support. We will use more generic interface cluster policy update for these use cases. * Removed permission property from profiles. We will devise a RBAC mechanism as an alternative. * Status *DELETED* is removed from clusters and nodes. * Timestamp fields like 'created_time' and 'udpated_time' are renamed to 'created_at' and 'updated_at' respectively. * As a side-effect of the rework of action dependency, a new table has been added to the database. * Senlin binaries are all made as console script entries. Bug Fixes ********* * When referenced objects are not found in an API request, 400 is returned now. * Reworked action status check so that a cluster action will always return from WAITING status. * Fixed profile type checking error when attaching affinity policy. * Fixed parsing of default values for 'max_size' and 'min_size' properties of a cluster. * Fixed race condition in service deletion. * Fixed APIs that spawn asynchronous operations to return 202 as status code. * Fixed a bug related to setting the next_index property of a cluster after new nodes have joined or existing nodes have left. * Fixed cluster-list function so that 'global-project' can be specified. * Removed useless parameters ('user', 'project', etc.) from filtering when listing clusters. * Added parameter sanitization for cluster-policy-attach. * Fixed RC/role checking in the setup-service script. * Enforce multi-tenancy checking when a non-admin user attempting to list resources from projects other than that of the requesting user. * Fixed a bug related to using the 'name' property of a nova server profile. * Fixed parameter checking when listing resources, such as sort and filters. * Added parameter checking for policy-create API calls. * Added parameter checking for cluster-policy-detach API invocation. * Added parameter checking for cluster-policy-update API invocation. * Reworked action dependency to avoid indefinite waiting problem. * Fixed trust usage when interacting with keystone. This enables senlin to be deployed on a Juno version OpenStack. Other Notes *********** * DB isolation level defaults to READ_COMMITTED in order to solve concurrency problems encountered in action dependency checking. * Added documentation for senlin.policy.deletion-v1.0. * Added configuration option for enforcing name uniqueness. * Added documentation for lb policy, affinity policy, scaling policy, zone placement policy and region placement policy. * Senlin API documentation merged into api-site and published. * Added user documentation for 'receiver'. * Added developer documentation for 'receiver'. * Removed documentation for 'webhook'. * The property 'priority' and 'level' are removed from policy create/update. * Command *senlin-manage purge_deleted* is removed. * Ensure there is no underscores ('_') in resource names exposed through RESTful API * User documentation for events and actions have been added. * User documentation (including developer docs) are published on official site. Changes in senlin 1.0.0.0rc2..1.0.0 ----------------------------------- 4da9639 Fix event generation Diffstat (except docs and test files) ------------------------------------- senlin/engine/event.py | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)

1 0

[release][sahara] sahara-image-elements 4.0.0 release (mitaka)
by no-reply＠openstack.org 07 Apr '16

07 Apr '16

We are chuffed to announce the release of: sahara-image-elements 4.0.0: Image elements for Sahara This release is part of the mitaka release series. For more details, please see below. Changes in sahara-image-elements 4.0.0.0rc1..4.0.0 -------------------------------------------------- 9ac70dc Update .gitreview for stable/mitaka Diffstat (except docs and test files) ------------------------------------- .gitreview | 1 + 1 file changed, 1 insertion(+)

1 0

[release][murano] murano 2.0.0 release (mitaka)
by no-reply＠openstack.org 07 Apr '16

07 Apr '16

We are overjoyed to announce the release of: murano 2.0.0: Murano API This release is part of the mitaka release series. For more details, please see below. 2.0.0 ^^^^^ New Features ************ * Now all native MuranoPL methods (those that are written in Python) have "?muranoMethod" metadata key referring to MuranoMethod instance for the method. * Classes to work with Cinder volumes were added to core library. Now it is possible to create new volume from various sources or use existing volume. Also it is possible to attach volumes to instances. This doesn't include ability to boot instances from volumes. * Class configs are now also versioned. For class foo.bar version 1.2.3 the following file names will be examined: foo.bar-1.2.3.json foo.bar-1.2.3.yaml foo.bar-1.2.json foo.bar-1.2.yaml foo.bar-1.json foo.bar-1.yaml In addition for classes of version 0.x.y file name without version suffix are also examined as a last attempt so the backward compatibility is retained * New plugin 'murano_heat-translator_plugin' was added. Now it is possible to deploy applications from CSAR templates. * Enable mocks in MuranoPL tests cases. * Added >>:command:`murano-test-tunner`<< command to run murano package tests. <SYSTEM MESSAGE: <string>:17: (ERROR/3) Unknown interpreted text role "command".> * Introduced two YAQL *inject* functions to enable mocks "def inject(target, target_method, mock_object, mock_name)" and "def inject(target, target_method, yaql_expr)". * New method type: extension methods. Extension methods enable you to "add" methods to existing types without modifying the original type. Extension methods are a special kind of static method, but they are called as if they were instance methods on the extended type. Extension methods are identified by "Usage: Extension" and the type they extend is determined by their first argument contract. Thus such methods must have at lease one parameter. * New type-level keyword "Import" which can be either list or scalar that specifies type names which extensions methods should be imported into class context and thus become available to type members. * Added filter by 'Name' which only matches package name. * All HOT template outputs are put into a single dictionary property 'templateOutputs' rather than in a generated property per each output. As a result there are no more constraints on output names. * Added magnum plugin to murano, that allows creating/deleting of magnum baymodels and bays from MuranoPL * Added ability to extend MuranoPL entities with custom metadata. * For MuranoPL classes new key "Usage" was added. By default it is "Class". But it can also be "Meta" to define meta-class. Meta-class has all the capabilities of regular classes and in addition has 3 new attributes: Cardinality, Applies and Inherited. * It is possible to attach meta-class instances to packages, classes (including other meta-classes), properties, methods and method arguments. Each of them got new "Meta" key containing list (or single scalar) of meta-class instances. * Now it is possible to have several classes in one YAML file. Classes are separated using YAML document separator (3 dashes). Empty documents are skipped. If the class doesn't have Namespace section corresponding section from the previous class in the same file is used. Thus it is possible to declare namespace prefixes once at the file header. Even if there are several classes in one file all of them are still required to be declared in manifest file. * Add multiple engine workers * New operator *is* was added to MuranoPL. Now it is possible to test if MuranoPL object is of particular type. * Murano engine is now capable of caching packages on disk for reuse. This is controlled by *packages_cache* directory path and *enable_packages_cache* boolean parameter (true by default). The packages are cached in a eventlet/inter-process safe manner and are cleaned up as soon as newer version of the package becomes available (unless it's used by ongoing deployment) * Added public field to environment templates. GET method for api now displays public templates from other projects(tenants). * Added public filter to environment templates api. * Added clone action to environment templates. * Basic reflection capabilities were added to MuranoPL. Now it is possible to get type info with typeinfo() function and using it as a starting point obtain information about the class, its methods and properties as well as the package of the class. Reflected properties can be used to obtain or set its value in a given object or invoke its method. * Static methods and properties were introduced. Both properties and methods can be marked as Usage: Static Statics can be accessed using ns:Class.property / ns:Class.method(), :Class.property / :Class.method() to access class from current namespace or type('full.name').property / type('full.name').method() to use full type name. * io.murano.configuration.Linux methods are now static * Heat stacks created by murano during environment deployment now have 'murano' tag by default. This is controlled by "stack_tags" config parameter. * Murano was migrated to yaql 1.1 * New format MuranoPL/1.3 can be specified in manifest files. MuranoPL/1.3 is identical to MuranoPL/1.2 but except for the fact that MuranoPL/1.3 packages cannot be imported to earlier Murano versions. Thus applications that use new features of yaql 1.1 should use this format version. Known Issues ************ * Enabling multiple workers might break workflows under BSD and Windows systems Upgrade Notes ************* * When updating to Mitaka, the operator should update service name and type for endpoint in keystone from "application_catalog" to "application-catalog" if SQL is used for catalog back-end driver. Bug Fixes ********* * Murano is now able to work with keystone configured to use a templated catalog. * It was impossible to explicitly provide attribute owner class to getAttr/setAttr methods without using namespace prefix or if the type was not from the core library. * Murano is now able to assign correct floating IPs when using multiple external networks. It attempts to choose one that shares a router with internal network. * Avoid race condition during parallel upload of packages, when packages have same tags. * Fixed incorrect murano behaviour if deployed on devstack with keystone v3 by default. * RequestContext now serialises it's roles. This should allow murano to work correctly (and allow rules like "role:xxx" in policy.json) when using oslo.context prior to 2.2.0 and oslo.policy Changes in murano 2.0.0.0rc2..2.0.0 ----------------------------------- 579731a Destroy orphan objects f2b8c28 Don't use bash eval for Linux.runCommand Diffstat (except docs and test files) ------------------------------------- meta/io.murano/Resources/RunCommand.template | 15 ++++++----- meta/io.murano/Resources/scripts/runCommand.sh | 14 ---------- murano/common/engine.py | 7 ++++- murano/dsl/executor.py | 37 ++++++++++++++++++++------ murano/dsl/object_store.py | 6 +++++ murano/dsl/serializer.py | 5 ++-- murano/engine/system/agent.py | 12 +++++---- 8 files changed, 60 insertions(+), 38 deletions(-)

1 0

[release][murano] murano-dashboard 2.0.0 release (mitaka)
by no-reply＠openstack.org 07 Apr '16

07 Apr '16

We are eager to announce the release of: murano-dashboard 2.0.0: The Murano Dashboard This release is part of the mitaka release series. For more details, please see below. 2.0.0 ^^^^^ Murano-dashboard now supports python3 New Features ************ * Add details page for packages, The details page will show more info to user, such as FQN, Description, etc. * Requirements for the flavor field accepts 'max_vcpus' 'max_disk' and 'max_memory_mb'. Previously only minimum values can be specified in the flavor requirements. * Added :filter by 'KeyWord' in package definition can matches all the package parameters like name, tags ..etc. * When using glare the endpoint for client is now 'artifact' since glare has been moved to a separate service * Version of Dynamic UI is increased to 2.3 due to *password* field update. Now *password* supports validator overloading and control of automatic password conformation field insertion. * If "regexpValidator" is provided, default complex check for numbers, capital and small letters in the password is not performed. Also, several validators with corresponding Dynamic UI field may be used. * "confirmInput" parameter is supported now for controlling whether password field should be cloned or not. If application author decided to turn off automatic field cloning, he should set the new parameter to *false*. Upgrade Notes ************* * Parameters "GLANCE_API_INSECURE" and "GLANCE_API_URL" have been renamed to "GLARE_API_INSECURE" and "GLARE_API_URL". Bug Fixes ********* * It is now possible to use any symbols in environments name. * Filter by 'Name' in package definition only matches package name. Changes in murano-dashboard 2.0.0.0rc1..2.0.0 --------------------------------------------- 8489e43 Imported Translations from Zanata 148ef99 Imported Translations from Zanata 5c4d60a Imported Translations from Zanata afcb2ee Add i18n trans for form field label and help_text c13875d Imported Translations from Zanata f0a0f65 Update .gitreview for stable/mitaka Diffstat (except docs and test files) ------------------------------------- .gitreview | 1 + muranodashboard/categories/forms.py | 4 +- muranodashboard/environments/forms.py | 2 +- muranodashboard/images/forms.py | 4 +- muranodashboard/locale/django.pot | 48 +- muranodashboard/locale/ja/LC_MESSAGES/django.po | 11 +- muranodashboard/locale/ko_KR/LC_MESSAGES/django.po | 13 +- muranodashboard/locale/pt_BR/LC_MESSAGES/django.po | 14 +- muranodashboard/locale/ru/LC_MESSAGES/django.po | 1106 ++++++++++++++++++++ muranodashboard/locale/zh_CN/LC_MESSAGES/django.po | 58 +- muranodashboard/packages/forms.py | 6 +- 11 files changed, 1228 insertions(+), 39 deletions(-)

1 0

[release][searchlight] searchlight 0.2.0 release (mitaka)
by no-reply＠openstack.org 07 Apr '16

07 Apr '16

We are eager to announce the release of: searchlight 0.2.0: OpenStack Search Service This release is part of the mitaka release series. For more details, please see below. 0.2.0 ^^^^^ Adds cinder plugins for volumes and snapshots. Added separation of administrator-only data at an index level, which allows full faceting and searching of sensitive data by those with access to it. In order to allow policy files to more be useful in Searchlight, we now support limited control over individual resource types via oslo policy control. Remove the 'index' API. It is currently a passthrough to elasticsearch, allowing direct index access and bypassing any document transforms. As such it is dangerous, and only marginally useful. Administrators wishing to do this should do so directly in Elasticsearch. Swift plugin for searchlight Ability added to reindex data while allowing indexing of notifications, and therefore returning accurate search results during the indexing operation. A fix for a couple of parent-child plugin relationship issues related to initial indexing. Child plugins can no longer be indexed separately from their parents. Config options for searchlight-api and searchlight-listener are grouped under config groups [api] and [listener]. The default config file for searchlight was searchlight-api.conf despite the fact that this config file was used by both the API and the listener processes. Therefore, we decided to rename the default config file to searchlight.conf. This change also prompted us to to follow the convention used by many other projects and simplify searchlight-api-paste.ini to api-paste.ini. New Features ************ * Cinder plugins for volumes and snapshots. * Plugins with sensitive, administrator-only data are indexed twice; once with that data and once without. Searches are directed to the appropriate version depending on the context of the user. * Adds neutron plugins for networks, subnets, ports and routers. * Policy rules can now be specified per resource type and per operation per resource type. * Removes the /index api and related tests. If an admin needs to directly modify data in Elasticsearch they should do so in Elasticsearch. * Three resource types are introduced for swift plugin. OS::Swift:Account ->OS::Swift::Container -->OS::Swift::Object * Reindexing of data now takes place in a separate Elasticsearch index. During indexing, incoming events are processed into both the old and new indices. Once indexing is complete, an Elasticsearch alias is used to switch searches seamlessly onto the new index, at which time the old one is removed. * Plugin configuration refers to *resource_group_name* rather than *index_name* to represent groups of resources separate from the logical Elasticsearch index. This setting is used to determine index and alias names. Known Issues ************ * Backups cannot currently be implemented because the cinder API does not provide enough information for RBAC. They will be added when that is resolved. * Neutron resources do not provide dates (created_at or updated_at). created_at is left empty; updated_at is taken from the server (and will be adjusted to use the notification timestamp at a later time) * RBAC is still separate from policy restriction; this is targeted for the Newton release. * The Swift service currently doesn't send notifications. Follow the swift plugin documentation for current solutions. * Reindexing must now take place across all resource types that share the index (it is not possible to re-index a specific resource type). This is a limitation that Elasticsearch 2.x would likely have forced upon us anyway. A subsequent patch will allow more efficient reindexing for associated resource types. Upgrade Notes ************* * Options 'bind_host', 'bind_port', 'backlog', 'tcp_keepidle', 'ca_file', 'cert_file', 'key_file', 'workers', 'max_header_line', 'public_endpoint', and 'http_keepalive' are moved to [api] group. Options 'listener_workers' is renamed to 'workers' and moved to [listener] group. Options 'host' and 'http_timeout' are removed since they are not currently used anywhere. * During an upgrade to Mitaka, operators must rename the previous configured searchlight-api.conf to searchlight.conf and searchlight- api-paste.ini to api-paste.ini. Security Issues *************** * Removes the admin-only '/index' api, which allowed an unnecessary amount of access to Elasticsearch. Bug Fixes ********* * Child plugins cannot be indexed on their own, only as a result of indexing their parent. This fixes an issue whereby indexing only a parent plugin could leave orphaned children (or vice versa). * Applies mappings for parent/child plugins simultaneously. Under elasticsearch 2.x, child mappings must be created at the same time or before parents, not the other way round. * The workers for listener was always 1, so config listener_workers was not work as expected. Now operators can set 'workers' under group [listener] to running multiple listener processes. Changes in searchlight 0.2.0.0rc1..0.2.0 ---------------------------------------- 2a44b63 Backward compatibility with designate v1 api 4abf961 Disable oslo_config file discovery in tests bcc8243 Increase heap for elasticsearch in func tests 1e80553 Fix _is_multiple_alias_exception signature 574a095 Add missing zero-downtime indexing documentation bb5ff07 Re-indexing optimization for doc_type 25ca131 Add missing mitaka release notes 531447d Zero Downtime Re-indexing Error Handling. b1d339b Don't index DHCP ports a5ccab8 Update .gitreview for stable/mitaka Diffstat (except docs and test files) ------------------------------------- .gitreview | 1 + .../bp-index-role-separation-535e01ba3bb2252e.yaml | 11 + ...-zero-downtime-reindexing-a607d52b8711d845.yaml | 24 + searchlight/cmd/manage.py | 116 ++-- searchlight/elasticsearch/plugins/base.py | 5 + .../plugins/designate/notification_handlers.py | 119 +++- .../elasticsearch/plugins/designate/recordsets.py | 1 - .../elasticsearch/plugins/designate/zones.py | 2 +- .../plugins/neutron/notification_handlers.py | 12 + searchlight/elasticsearch/plugins/neutron/ports.py | 6 + searchlight/elasticsearch/plugins/utils.py | 210 +++++-- 22 files changed, 2186 insertions(+), 118 deletions(-)

1 0

[release][trove] trove 5.0.0 release (mitaka)
by no-reply＠openstack.org 07 Apr '16

07 Apr '16

We are eager to announce the release of: trove 5.0.0: OpenStack DBaaS This release is part of the mitaka release series. For more details, please see below. 5.0.0 ^^^^^ New Features ************ * Support has been added for Cassandra backup and resture using the Nodetool utility. * Implement configuration groups for Cassandra 2.1. You can now manage configuration of Cassandra datastores using the Trove configuration groups capability. * This patch set implements the following functionality for Cassandra datastore. create/delete/get user list users change password grant/revoke/list access update attributes create/delete database list databases Notes on Cassandra users In Cassandra only SUPERUSERS can create other users and grant permissions to database resources. Trove uses the 'os_admin' superuser to perform its administrative tasks. It proactively removes the built-in 'cassandra' superuser on prepare. The users it creates are all 'normal' (NOSUPERUSER) accounts. The permissions it can grant are also limited to non- superuser operations. This is to prevent anybody from creating a new superuser via the Trove API. Updatable attributes include username and password. The configuration template had to be updated to enable authentication and authorization support (original configuration allowed anonymous connections). Default implementations used are authenticator org.apache.cassandra.auth.PasswordAuthenticator authorizer org.apache.cassandra.auth.CassandraAuthorizer The superuser password is set to a random Trove password which is then stored in a Trove- read-only file in '~/.cassandra/cqlshrc' which is also the default location for client settings. Notes on Cassandra keyspaces Cassandra stores replicas on multiple nodes to ensure reliability and fault tolerance. All replicas are equally important; there is no primary or master. A replication strategy determines the nodes where replicas are placed. The total number of replicas across the cluster is referred to as the replication factor. The above 'create database' implementation uses 'SimpleStrategy' with just a single replica on the guest machine. This is a very simplistic configuration only good for the most basic applications and demonstration purposes. SimpleStrategy is for a single data center only. The following system keyspaces have been included in the default 'ignore_dbs' configuration list and therefore excluded from all database operations 'system', 'system_auth', 'system_traces' Notes on user rename Cassandra does not have a native way for renaming users. The reason why Cassandra itself does not implement rename is apparently just lack of demand for that feature. We implement it by creating a new user, transferring permissions and dropping the old one (which also removes its existing permissions). I asked about the sanity of this rename approach on the Cassandra mailing list and IRC channel and there should not be anything inherently wrong with the proposed procedure. This method, however, requires the user to always provide a password. Additional notes Trove uses the official open-source Python driver for Cassandra to connect to the database and execute queries. The connection is implemented in CassandraConnection. It is now also used to obtain the current database status as opposed to the original method of parsing output of the client tool. The 'common/operating_system' module was extended with two new functions for reading/writing ini- style and YAML configuration files to/from Python dicts. Unit tests were added to 'guestagent/test_operating_system'. The existing Manager unit tests were extended to include the added functionality. Also includes some minor improvements to comments and log messages. Used the existing operating_system interface to update file ownership. The system module was removed and its contents moved to the Application class. This is to reduce the number of files and help facilitate overriding. * Support has been added for CouchDB Backup and Restore. * Support has been added for CouchDB database and user functions. * Support has been added for DB2 Express-C Backup and Restore. * Additional Ceilometer notifications have been provided by Trove including create, end, error notifications for all state-changing API calls. * OpenStack Trove now supports clustering for Cassandra datastores. You can access clustering capabilities through the Trove cluster API. * OpenStack Trove now supports superuser access for the Cassandra datastore via the root-enable and root-disable API's. * OpenStack Trove now supports clustering for MariaDB datastores. You can access clustering capabilities through the Trove cluster API. * Implements replication based on GTIDs for MariaDB. Adds GTID replication strategy for MariaDB. Implements MariaDB specific GTID handling in guestagent. Configures MariaDB config template to support bin logging. Adds MariaDB helper overrides to eliminate configuration group tests from scenario tests. * A new feature called 'module management' has been added to Trove. Users can now create, update, list and delete modules. A module is a file that is provided to Trove, and when a database instance is launched, that file is deposited on the guest instance. This feature can be used for depositing files like, for example, license files onto guest database instances. * Support has been added for Percona XtraBackup version 2.3. * Adding the ability to root enable a pxc cluster. * The adds support for pxc to grow a cluster. * api and taskmanager support for shrinking a cluster * validate that the networks given are the same for each instance in the cluster. * make sure to add the existing networks on an instance in the cluster. * add new Error task for grow and shrink. * nova client version configuration changed to a string option rather than an int option because the nova microversions change nova api output. This was needed for the network interfaces on existing instances. * testing for grow and shrink cluster * Implemented configuration groups capability for Vertica datastores. * Implemented grow and shrink for clusters of Vertica datastore. The number of nodes in the cluster must be greater than the number required to satisfy the min_ksafety configuration setting. * Vertica comes with a User Defined Load function that takes a URL as a load source. This can be used to load files that are stored in Swift. As this is a common use case, it is valuable to enable this by default. This can be done in the post-prepare method for Vertica. A new UDL_LIBS list has been added that describes any UDLs to be loaded into the database. This change only has one entry - the curl function. Deprecation Notes ***************** * Dropping support for python 2.6 Security Issues *************** * Fixes bug 1507841, provides a configuration setting to enable Role Based Access Control (RBAC) for MongoDB clusters. If mongodb.cluster_secure is set to False (default is True) then RBAC will be disabled. Bug Fixes ********* * If given, apply the configuration overrides in prepare, just before creating initial users and/or databases. Failure to apply the given configuration should flip the instance into a failed state. Default implementation saves the overrides and restarts the database service to apply the changes. Datastores that do not require restart may potentially override the base implementation in 'apply_overrides_on_prepare()'. * The guest log code raises a non-serializable exception if the given Swift endpoint is invalid. This causes an ambiguous "Circular reference detected" error on the guest, and a timeout on the caller. This case is now caught and the correct exception raised. * Fixes bug 1526024, a failure in growing a mongodb cluster because of a problem in the way in which passwords were synchronized with new query routers. * Generate trove events for the current period, and not a future period. * Filter ignored users in the original query before the result gets paginated (like in list_databases). * Mongo cluster grow operations were not creating instances with the provided az and nic values. These should be used if the caller provided them. * Fix bug 1537986 which corrects the pagination in the mysql user list command. When internal users (ignore_users) are eliminated from the list, the pagination was not correctly handled. * Fixes bug 1558794. The 2.3 version of Percona XtraBackup performs some additional validations of the command line options passed to innobackupex. The Trove code now complies with the new validations being performed. Other Notes *********** * Added CORS support. * Refactor the datastore guest manager code. * Starting with 1.0.0 osprofiler release config options needed for its workability are consolidated inside osprofiler itself. Changes in trove 5.0.0.0rc1..5.0.0 ---------------------------------- 4e3e804 Trove's tox tests should respect upper-constraints.txt 18331be Address change in Route v2.3 f34581a Update .gitreview for stable/mitaka Diffstat (except docs and test files) ------------------------------------- .gitreview | 1 + tox.ini | 26 +++++++++++++++++++++++++- trove/common/extensions.py | 5 +++-- 3 files changed, 29 insertions(+), 3 deletions(-)

1 0