OpenStack

Current documentation for “OpenStack”

To classify a blueprint as documentation, set the Implementation status to “Informational” When the blueprint's Definition status is marked “Approved”, it will appear in this listing.

Blueprint Template for OpenStack-Ansible
Overview ######## Provide a synopsis as to why you are creating this blueprint Problem Description ------------------- Describe the problem Proposed Change --------------- Provide an overview of the changes you'd like to see implemented. Notable changes: * list out all of the notable changes Playbook Imp...
Need a driver for a non-AGPL backend that is durable and can support the entire 1.x API (feature-parity with MongoDB) Elasticsearch, LevelDB, Riak are our most likely suspects. Need to do some POCs to see which way we want to go.
This proposal describes the updated Ironic system lifecycle state machine. It includes new states for handling initial system enrollment and discovery, adds a couple of new states to make handling in-band and out-of-band system configuration easier, and adds a per-system wait flag to indicate that an external auto...
Template Blueprint for OpenStack Dashboard (Horizon)
This is a blueprint to provide a template for future blueprints. All of the below fields should be populated before a blueprint can be approved. "N/A" is an acceptable answer where applicable. Summary ======= A couple of sentences summarizing the goal of this blueprint. Motivation ======== This field should contai...
We need to basically implement HA across the board.
Alarming for Ceilometer
Build a Ceilometer Alarming service.
By default, the metrics gathering underlying ceilometer alarming can share the same infrastructure as the metering functionality. However in realistic deployments, we envisage that operators may want to reflect differing requirements by separating off the transport conduit and storage elements. Also the metering d...
API v1.1 for zaqar
API v1.1 Spec and Implementation
http://devstack.org
Best performance will come from indexing purely from notifications. Currently there's information that we can only get from the API (glance metadefs are a good and tricky example, but nova instance notifications don't necessarily have all we would like). There's also a larger conversation about potential crossover ...
We should enhance the stackube documentation, including - Architecture of stackube - Deployment requirements, steps and diagram - Developer guide - Admin guide - User guide - Troubleshooting
At the Liberty design summit in Vancouver, the Ironic team generally agreed to move the Ironic project to a feature-based release model.[0] This is an informational spec to lay out why we are doing this and how it will work.
Fencing in fuel-library manifests for Fuel for OpenStack
Provide puppet classes and corosync providers for STONITH based fencing implementation for fuel-library (see full spec link). As an option, it could be a plugin for Fuel
Essex introduced nova-rootwrap command filtering. This spec collects all the improvements we intend to push in Folsom. 1. Move filter definition to configuration files This will allow to plug new commands needed by plug-ins, and support other configuration options (like specifying a log file) 2. Support for execut...
Monitoring API for MagnetoDB
The API for monitoring of MagnetoDB. [Obsolete] https://wiki.openstack.org/wiki/MagnetoDB/specs/monitoring-api http://magnetodb-specs.readthedocs.org/en/latest/specs/kilo/approved/monitoring-api.html
PKI feature were introduced in Grizzly and new Savanna releases should support it. See more details about PKI at https://wiki.openstack.org/wiki/PKI.
Refactor chaotic test codes for OpenStack DBaaS (Trove)
Trove has a complete different test code style with other OpenStack core projects which cause the developers confused. It really block other OpenStack project developers to join Trove. Besides the author of test codes, no one could have a clear mind on these in short time. We need to refactor the test codes which h...
Reference Architectures for OpenStack Compute (nova)
OpenStack has many choices for Hypervisor, Network Model, User Model, API, Storage System, OS, Packaging, etc. To keep up with Openstack's expansion, we need to document and automatically test configurations which OpenStack supports. We need to document existing configurations (for instance a rackspace-style sys...
What does scaling mean at the tripleo level; where do we do well, where are lacking, what do we need to design? For both the deployment activity itself as well as the deployed cloud...
We need to deliver UI which tells complete overcloud installation story. It has to be slick, user friendly and functional.
Volumes, snapshots and keypairs show all objects for all tenants in the Horizon user dash. This is different from instances and security groups which only show the tenant info regardless of admin/non-admin. Related to bug: https://bugs.launchpad.net/horizon/+bug/967882
Currently once you do a deployment with TripleO, you are unable to change or rotate any of the passwords used in the overcloud. For day 2 operations we need to be able to change all passwords in the environment in a case where an Operator leaves, or as part of standard security protocol (passwords rotated every 6 m...
All Strings Localizable for OpenStack Dashboard (Horizon)
Add gettext support for all strings in Dashboard.
Goal: To gain consensus on a suitable infrastructure model in Tuskar.
On the app category index, we are removing the right side control bar. Therefore we need to add an Environment selector at the top of the page to replace the one being removed from the control bar.
On the app catalog index page, we are removing the right-side control bar and instead we want to use the standard Horizon search filter UI at the top of the app listing.
Add new gate/check jobs that will run functional tests against devstack.
Currently, there is only one Records table for all the different record types. This blueprint proposes to subdivide that tables into a Record table per record type. Having one monolithic Records table for all record types can become problematic as the size of the table grows. If any modifications need to be made, f...
XenServer Guest Requirements for OpenStack Compute (nova)
There are a number of requirements needed to support XenServer guests for Cloud Servers, etc. Most of these are being broken out into other blueprints. • Add & Remove routes. (xs-guest-routes) • Add/Remove IPs to networking config, networking protections, etc with option to actually configure the guest [config vs ...
The S3 Backend needs to use boto. While nova uses boto throughout, none of glance uses it at all. It would probably be best that people who use other backends aren't required to also import boto, if only so that we have the precedent that when someone writes JoesWeirdAssBackend clients who don't use JoesWeirdAssBack...
Define MagnetoDB ACLs that can be associated with roles defined in Keystone. Enforce these access rights in the code. For example: - create row - read row - update row - delete row - create table - update table - delete table
NVMe-of connection sharing
Provide API (HTTP, streamed) for loading large amount of data. TBD
We need to impement proper RESTful interface for big data import api. Steps for implementation: 1. Create new streaming server with gevent.pywsgi 2. Improve configs 3. Create new controller in /openstack/v1 4. Create new streaming application 5. Add URL mappings to api/wsgi Also, it would be nice to add the respon...
Sending a PKI token every request introduces huge overhead (4-16KB) in bandwith usage. It is suggested to introduce alternative lightweight mechanism based on HTTP session approach.
Adding name for audit entity would be more friendly to end users to interact with audit. This blueprint suggests to add name property to audit entity, so it will be easy for users to retrieve an andit by name.
API Microversions for Ironic
Add documentation about how Ironic and clients should interact negotiating which version of the REST API to use.
Generate 'keystone.conf.sample' dynamically for OpenStack Identity (keystone)
'keystone.conf.sample' canbe generated by the conf registed in oslo-config, so we can provide shell script to scan the source code and generate the sample config file instead of maintain this file in the trunk, the benifit for keystone project is the single truth of source and can generate the file dynamically. The...
Some logging does not include the project ID for the request, which is useful context for debugging issues. Let's find a DRY approach to this - can we use the oslo context thing?
Add Trove panels unit-tests in Horizon for OpenStack Dashboard (Horizon)
Adding Trove(DBaaS) panels unit tests via horizon testing framework for validation of trove dashboard integration.
I'd like to raise this blueprint again, Kibana UI is a good option for filtering logs in RT, and Elasticsearch is a good backend as a search-engine for Logstash logging engine to replace(evntually) rsyslog as well. We should also reuse Openstack Infra experience as well, see http://logstash.openstack.org as a good ...
The backend can or cannot have support to the QoS floor feature, depending on the ONTAP storage. The administrator may want to create a volumes on backends that have that support, leading to the requirement of a new driver specific capability to filter QoS min support backends. This bluenprint addresses it by adding...
Admin API to create and restore backups for OpenStack DBaaS (Trove)
Create an Admin API that allows an admin user to create a backup on behalf of a customer. Also to restore that backup onto a new instance that is created on the customer's behalf. This will support future use cases with automated backups, and potentially automated failover, where an admin user will be required to ...
We should add сonfiguring data consistency to MagnetoDB. Steps to implement: 1. Add the possibility of adding consistency level in the request via our RESTful API. 2. Add transmitting the consistency level value to storage API and after that - to backend (at the moment - in Cassandra)
Add metadata in docs for openstack-manuals
In the generated documentation, add information about where the documentation comes from. In particular, what the associated Launchpad project is (e.g., openstack-manuals) and what the corresponding git repository is.
Hyper-V Spawn on Neutron Event for OpenStack Compute (nova)
Under large scale deployments, Hyper-V Neutron port binding can take a longer time. Starting the instance before the port was setup, can cause the instance's DHCP requests to timeout and thus, become unreachable. This problem can be solved by waiting for the Neutron port bound event before starting the VM. The conf...
Use itstool for docs translation for openstack-manuals
openstack-manuals, api-site, and operations-guide currently use a custom xml2po-based script to extract POT files from DocBook and to merge the translations back to DocBook. xml2po is not designed to work like this, and the generatepot/generatedocbook scripts override xml2po internals to provide custom behavior. its...
Add the API hooks for CRUD operations relating to the kubernetes resources. Resources mean Pod, Replication Controller and Service.
remove-unittest2 for OpenStack Dashboard (Horizon)
unittest2 codes should be removed from horizon since we prefer testtools.
API definition process for OpenStack Compute (nova)
We need to discuss and possibly revise the way we define the OpenStack API. It's currently being defined by people other than the ones implementing it, and it's being defined as a massive, single feature. It seems like a much more natural approach to have an "Changes to the API" section for each proposed feature spe...
Background OpenStack Ironic Service provides support for image based deployments to baremetal servers. This blueprint proposes adding Windows Image generation capability to diskimage-builder. Windows Disk Image Builder A separate windows Disk image builder tool will be added to the Disk Image Builder proje...
Diffie-Hellman-Merkle Key Exchange Support for OpenStack Compute (nova)
Several of the communication processes between nova and guest agents require DH key exchange for security. We need to implement a basic DH module for this.
Background This blueprint proposes to extend the functionality of windows image builder [1] for Workload/Role based Windows Disk Image generation. Windows Disk Image Builder Once the basic tool [1] is in place, it will be extended to provide hooks where user can plugin custom scripts to automate tasks. Usin...
There are some circumstances where the auth wrapper isn't needed, so it would be useful to be able to turn off the middleware entirely.
Adding support for GPU accelerators, SGI Ultra Violet (UV) shared memory machines and Tilera TileEmpower 64-core tiled processor systems as alternative machine types in OpenStack.
Support for heterogeneous architectures for OpenStack Compute (nova)
Nova should have support for cpu architectures, accelerator architectures, and network interfaces and be able to route run_instances() requests to a compute node capable of running that architecture. In addition, nova should prevent a user from inadvertently specifying, for example, an x86_64 machine image to run on...
Neutron has, in DEFAULT section in neutron.conf file, a key dns_domain which is a domain fallback. Basically, if we create port with neutron, the default dns assigment will be myport.defaultzone.com. With Designate Integration, that will create DNS record myport.defaultzone.com. A portIP The problem is, defaultz...
Boot whole disk images for OpenStack Compute (nova)
Boot whole disk images without separate kernel/root disk
This blueprint proposes configurable pool backend via API.
Since MangetoDB works with some kind of NoSQL backend. When authorized user whants to write/read data backend should already be provisioned. Main ideas are:  - implement management service which would work with different deployment systems (each deployment system relays on Magnetodb management driver); - provide pl...
As currently, there are less number of translated documents in local languages, there is just one landing page per language in the link format as http://docs.openstack.org/[language]/. But when end users try to visit http://docs.openstack.org//[release]/[language] for local documentations as English version http://d...
Snapshot, Clone and Boot from volumes for OpenStack Compute (nova)
This session will look into three different subjects: - Snapshot volumes - Clone volumes - Boot from volumes
This blueprint sets out a common method using TripleO Image Elements to install and configure a simple monitoring solution for TripleO-deployed Openstack installations and is based on Icinga and Check_MK. The system should automatically configure itself from heat and nova and checks should be automatically added fr...
The deficient of the current way of injecting network configuration to VM are: 1. the configuration file is always injected to the first partition. It will be injected wrongly if one use native ubuntu livecd installation, the file will go to /boot 2. it is unable to recognize the layout of the partitions, such as Li...
Rick Clark asked that I create a blueprint for getting cloud-init into the rackspace images.
lessons learned in scaling swift clusters for OpenStack Object Storage (swift)
lessons the swift team at rackspace has learned in managing and growing large swift deployments
Obsolete, implemented as part of other bps
planned and unplanned features for swift for OpenStack Object Storage (swift)
A discussion around the features that swift developers have identified as interesting for swift. Some are planned, some are not.
The VM image is checked by fdisk before mounting, if the windows partition is found a batch file for changing network settings by the command netsh.exe is injected to the VM. Before doing the above action the a windows VM must be installed in the way as suggested in the openstack's wiki webpage and a template of bat...
We need lightweight PutItem method at storage layer usable for bulk loads of large amounts of data. This method should perform simple transformation from dict into cql 'INSERT' statement and execute one DB access call. All aditional data (table schema etc.) should be privided through parameters. No conditionals or u...
Existing Hyper-V msi(Liberty) installation there is no option to get user interest on security groups. Because of this if user wants neutron security functionality they have change the config and need to restart the service. So its mandatory to provide such option to get user interest during installation.
Adding Cinder Image-Volume support for NetApp block drivers. This will utilize the image caching feature to create clones of images downloaded from Glance.
Open Stack Northbound REST API for OpenStack Compute (nova)
A RESTful JSON/HTTP interface & programming model for management & control plane of cloud systems built using OpenStack components. Will have RFC for appropriate Media/MIME types. To be used by cloud service consumers & operators to build higher level systems well as by the OpenStack CLI interface. Orthogonal Extens...
It is a common practice to make services managed by pacemaker in Linux HA clusters. Puppet modules for Openstack should be able to ensure both standard and pacemaker providers for Openstack services as well. In order to achieve this as a transparent solution (w/o modifications to the core puppet modules for Opensta...
At present we do not have a way to re-assert system state in the manner of other configuration management tooling to ensure that changes made outside of HEAT/Tripleo Image Elements are reset to the appropriate value. This blueprint aims to address this issue through the provision of a method for regularly re-assert...
Server Templates for Glance
A server template consists of a base image plus the definitions of configuration metadata. For example, a server template might include an Apache HTTP server; the metadata would include the server name, location of the HTML root directory, and tuning parameters. Glance stores the template in its registry; Nova, when...
Designate needs to provide a way to manipulate multiple records/domains with transaction support built into it. For example, a tenant who owns multiple domains should be able to add records in one domain while deleting in another domain. All of this manipulation should be either successful or rolled back together. A...
Reorganize and update the Architecture Design Guide. This specification is deprecated and a new specification will be created for the Newton release.
Suggested Features & Enhancements for OpenStack Dashboard (Horizon)
This blueprint exists to capture suggested features and enhancements that are without code. There is value in capturing these suggestions, but it should not obstruct our organisation of bugs or blueprints. Any bugs or blueprints that suggest new enhancements should be marked as 'Opinion' (bugs) or 'Obsolete' (bluep...
As discussed here https://etherpad.openstack.org/p/161115-kolla-kubernetes-cn-discussion we will create init container able to handle dependencies and deploy OpenStack using it
We should implement an IP discovery mechanism that will provide us an ability to determine the following 3 IPs per each instance of Hadoop cluster: * management_ip - it'll be used by Savanna to access Hadoop cluster's instances; * internal_ip (or it should be named more clearer) - it'll be added to /etc/hosts on no...
This is essential for scaling and operating a large deployment. The goals for this blueprint are: * find an approach that works * prototype it * demonstrate it Some requirements: - distribute load reasonably well amongst partitions - handle migrations transparently Some nice-to-haves: - statelessness: might requ...
Simplify conductor methods logic for OpenStack DBaaS (Trove)
Method takes to much parameters which are examined on non-equality - def update_backup(self, context, instance_id, backup_id, **backup_fields): #some logic skipped if backup_id != backup.id: LOG.error("Backup IDs mismatch! Expected %s, found %s" % ...
testing blueprint searching for OpenStack DBaaS (Trove)
Not sure why our searches for blue prints are not showing up. Do we need to change a setting?
Ocata Milestone 1A for tripleo
This blueprint will be used to track user stories and defects for openstack mistral and TripleO-Common.

85 blueprint(s) listed.