CHANGES
=======

* Remove openstack-doc-tools
* --stop-on-error flag added to watching reports
* Refresh transaction after each orchestration report tick
* Minimal solar VM packer template
* Add helper to preserve atomicity for database group operations
* Refactor error wrappers in CLI
* Use master branch of solar-resources
* Possibility to disconnect single input in resource
* Allow to use parent input more than once in CR
* Add *egg/ directories to .gitignore
* Compare names
* Don't install vagrant-vbguest when using other providers
* Rework staging to avoid unnecessary children log items creation
* Return paths to renedered files
* Execute update action in case resource was already created
* Implement traversal based on number of childs
* Refactoring of graph.py and usage of it in scheduler
* Use WeakValueDictionary instead of WeakSet in DBModelProxy
* Fix typo in Resource._pre_from_dict_check
* Fix repo.iter_contents after recent changes in Repository module
* Rework staging procedure to support both implicit and explicit stages
* Add vbox/qemu packer build for centos7
* Improve version parsing in Repository module
* Implement better and more solar specific pool for PG
* Reset times in tasks on graph reset
* Added naive method to check if object already exists
* Re-visit visited predecessors for react_on events
* Import repo to temp dir
* Add default solar.yaml config
* Added systemd solar-worker.service
* Solar orch report doesn't fail for empty graph
* Add /etc/solar/config.yaml as first file in config
* Insert missing events during build of staged graph
* Change PENDING_RETRY to ERROR_RETRY in soft_stop
* Create remote paths before syncing
* Separate vagrantfile for the docker provider
* Add troubleshooting to the development guide
* Make SolarTransportResult output contain both outs
* Add centos7 docker support
* Add packer build for docker container
* Fix riak container nested docker mounts
* solar.spec install path fixed
* Describe how to create/use the builder container
* Timestamp comparison in calculating time fixed
* Update rpm spec dependency
* Pass all args to resource and allow dumping as a yaml
* Replace Bunch with Munch package
* Replace python-pyzmq by python-zmq in solar.spec
* Fabric sync transport does not exit anymore
* Workaround for TypeError in zerorpc Events.__del__
* Fixed alignment issues in the text, corrected length
* Add rpm spec file for solar
* Fix parsing list of dicts
* Make zerorpc working with pyzmq > 13.0.2
* Move from semver to semantic_version package
* Raise an error for empty plan
* No arguments should be accepted by solar-worker
* Fix no attribute: result has no s
* Fixed non working test on riak backend
* Align requirements with global-requirements.txt
* Remove unused libraries
* Change sudo to become in bootstrap playbooks
* Added solar.version
* Add more stuff to exclude when syncing tree in tests
* temporary skip 2 problematic tests on postgresql
* Some speedups after adding PG
* Rework conftests + sql tables are cleared on success
* Use psycogreen for PG backend
* Fix dirs sync rsync type for Vagrantfile
* Fix TypeError in scheduler
* Removed old help message in solar orch cli
* Update requirements:click
* Hide errors from stevedore in transports
* Api in networkx 1.11 changed, updating
* PG support in Vagrant and in test env
* Support postgresql in dblayer
* Use openstack/solar-resources not Mirantis one
* Start 0.3.0 cycle

0.2.1
-----

* Fix incorrect version handling in Vagrantfile

0.2.0
-----

* Fix paths in tutorial
* Set versions in vagrantfile and playbook for 0.2.0
* Small fix in cli resource prefetch
* Increase disk size
* Improved ansible error reporting
* Exit with exit_code from deployement
* Add solar description
* Delete not needed travis.yml file
* Better information about resource repo in resource documentation
* Fix handling solar resource action run errors
* Ansible returns error on stdout
* Add information about libvirt provider
* Clarify docs about ansible handlers
* Fixes issue with types in composer
* Create separate page for development docs
* Added waiters instead of sleep in Lock
* Fix Vagrant version requirement in docs
* Restructure ansible handlers
* Change owner only if we are not in vagrant env
* Small cleanups
* Improve fuel-devops sctipt:
* Allow to set headless mode via env variable
* Torrent jenkins script
* Remove sleep 30 from run.sh
* Add UserKnownHostsFile=/dev/null to jenkins ssh
* Ip addressess fixes
* Use uppercase 'S' in word "OpenStack"
* Fix misspellings
* Do not interfere with vagrant network
* Use extra-vars in all ansible handlers
* Create tmp dir for overlay images
* Torrent transport can use corret solar_torrent.py
* Fix response handling in handlers/base.py
* Introduce documentation related to new worker
* Disable host key checking for whole 10.0.0.0 net
* Set correct user for /vagrant/solar-resources dir
* Ansible 2.0 support in ansible handlers
* Do not use archive mode, it copies owner and perms
* Use overlay images, it speedups tests
* Preinstall more stuff on an image
* Set StrictHostKeyChecking=no during jenkins tests
* Fixes for 'tox -e docs' in some dev environments
* Removed all __init__ files from pep8 ignore
* Added info that our examples require nodes
* Plugable constructors, runners and hooks
* Removed deprecated ansible handler documentation
* Allow to use any IPs with fuel-devops
* Fixed missing texts in docs
* Allow to load only some resources
* Use yaml to load data in solar create
* Hardcode python 2 usage
* Removed ssh_ prefix from transports resources
* Increase riak test timeout to 600
* Clean repository after removing celery
* Extensions mechanism for orchestration components
* Add support for configurable executor
* Fix simple statistics of a plan execution
* Removed global ansible library
* Added internal architecture diagram and fixed some doc errors
* Fix incorrect ansible_library path
* Timeout support for scheduled tasks
* Remove duplicated code in orchestration/scheduler
* Improve fixtures for functional tests
* Support configurable retries on orchestration layer
* Bump version to 0.2.0
* Implement timelimit mechanism based on tracking of tasks
* Add implementation of puller with pool of threads and tracking of tasks
* Re-start sqlite session before retry and after succesful acquire
* Add upstart config for solar-worker
* Zerorpc worker for orchestration modules
* Version 0.1.0
* Update usage docs
* Fix locking problems
* Update installation instructions
* More clear info about key/pw change in FAQ
* Added solar-celery script to setup.cfg
* Moved library to solar/ansible_library
* Fix session in CLI on Debian systems
* Fix celery start script
* use /usr/bin/env in celery script
* Improved celery worker section in FAQ
* Added simple FAQ to our docs
* Remove last `/` from urls
* Unify returned transport results
* Added EGroup click group which can handle errors
* Set correct classifier
* Unified exception in repo create
* Add tox checkniceness,checksyntax for docs
* Use exported env name, change default name
* Use isolated filebased sqlite for each test process
* Update Wordpress tutorial
* Increase riak timeout - time depends on download speed
* Fix 80 chars wraps for docs
* Move all text from README to documentation
* Fixing concurrency problems in history counter
* Hardcode ansible version. We are not ready for 2.0
* Remove ansible.cfg, we use .ssh/config now
* CRDTish lock to avoid concurrent update/delete
* Add test for wordpress example
* Set concurrency=1 for system log and scheduler queues
* Update path in tests
* Include ansible config when syncing repo
* Set ansible<2.0 in requirements
* Conditional imports in locking (riak or peewee)
* Allow to modify computable inputs in Composer files
* Remove redis and celery container, use init script for celery
* Improved inputs documentation
* init script for solar-celery
* Implement db based lock mechanism for orchestration
* Removed wrong git option from solar-resources clone
* Added possibility to change computable input func
* Moved examples, resources and templates
* Change celery config to use sqlite backend messaging and results
* Fix gevent cache, by default it shouldn't be shared
* Set vagrant as a owner of /var/lib/solar/repositories
* Use stevedore for handlers
* Fixed torrent transport
* Preserve original order of keys in CreatedResources.like
* Implements ansible_playbook_local handler
* Ansible Playbook uses now solar_transports
* Raise better error when requested transport is not found
* Standarized props names in ssh transports
* Use stevedore for transports management
* Fixed OSError during concurrent TempFileHandler init
* Running examples using fuel-devops
* Remove redundant filtering (old way) in riaks.py
* Use resources.like filter in openstack example
* Reduce image size and keep qemu image
* Switch to nodes.yaml composer file
* Install sshpass. Required for testing on Jenkins
* Simple sanitization for shell args in raw_ssh
* Fixed incorrect key file usage in fabric ssh
* Added create_all command to riak example
* Use stevedore for Computable Inputs
* Added passwords to schemas in transports
* Skip only lua CI related tests when no lupa
* Remove jenkins-config.yaml
* Removed old puppet_module field from resource
* composer.create returns CreatedResources object
* Added ansible_template_local handler
* Move bootstrap commands to ansible playbook
* Possibility to use ssh transports with password
* Possible to use relative paths in CR
* Removed resource compile_all action from CLI
* Refactor VirtualResources
* Add input operations (add / remove)
* Removed old config ideas
* Raise error for unknown inputs update / set action
* Replace all templates/(.*)\.yaml with templates/\1
* Use yaml to load data in CLI resource update
* Added solar repository create <name>
* Better scans source path in repository
* Add `tox -e docs` command
*  Update wp tutorial after recent changes
* remove python 2.6 trove classifier
* Mount correct dir for ssh configuration
* Fix add to VR repository from cli
* Put PYTHONWARNINGS in .bashrc before others
* Repository support for VR
* Changed {{ }} to #{ }# in VR templates
* Fix playbooks after recent changes:
* Changed block tokens in jinja for yaml compat
* Do not remove the kernel src for packer build
* Require semver in main requirements.txt
* Resource Repository Docs added
* Updated repo link, basic info in docs
* Use RFC 1738 format for redis db connection
* Change resources structure to match repository one
* Introduced Resource Repositories
* Write puppet log to /var/log/puppet/{resource.name}.log
* Advanced tags support
* SOLAR_DB from env in testr execution if defined
* Add gitreview file
* Adjusted dblayer sqlite behaviour to riak one Changed default database to sqlite Changed how solar_db config is defined Removed lupa from test-requirements Removed riak from requirements Testr uses :memory: sqlite
* Raise sync error in Executor
* Added gevent to test-requirements
* Point to custom remote in documentation
* Add doc about StrictHostKeyChecking
* Improve error handling in for ssh_raw/rsync in ansible/puppet/bash
* Add connections between nodes and hosts files
* Disable host key checking for 10.0.0.* subnet
* Fixed gevent related test failures
* Use preforked pool for celery and remove gevent from reqs
* Reuse parse_dict_input in parse_list
* Configure riak based to use multi backends
* Adjust celery configuration and fix ssh_raw to work with ENV
* Change pool to gevent from preforked for worker which performs tasks
* Convert hosts example to use VRs with all transports
* Install rsync, ssh client and mount known_hosts to container
* Add test for list of connected dicts and fix flake
* Adjust raw_ssh to plugable transport layer and use it by default
* Add rsync transport to nodes
* Allow to create list of dicts based on connections in VRs
* Added locks for python CI and for CI processors
* Removed unused imports
* Added missing files
* Pretify code
* Adjusted tests for cache
* Added `next` to proxy
* Added wrapt to requirements
* Set default cache size to 200
* Initial idea of new cache
* Fixed torrent transport + added tests
* Adjust n_val and ring_creation_size for riak
* Build riak container with solar specific commit
* Added libluajit-5.1-dev to base.yaml
* Corrected comment about sandboxing
* Set `sudo: false` in travis.yml
* Use apt addon in travis
* Add note about python_loop
* Fixed events cycles with self referenced inputs
* Moved riak_node_comp to riak_node
* Standarize API for inputs D is data with raw passed values R is make_arr result for jinja we also pass flat input as kwargs
* Automaticaly do make_arr with `full` mode
* Add tests for dicts in VR
* Extend tutorial to show how to update db pass
* Add docker_container update action with state=reloaded
* Update.yaml is update action
* Connections in dicts in VR
* Add wordpress tutorial
* Improve docker_container resource
* Added jinja2 computable input types
* Python computable inputs Simple repl like server with simple sandbox based on seccomp
* Install libluajit-5.1-dev on travis
* Added computable inputs riak example
* Fixed one more case in solar_models
* Fixed delete behaviour for single hash
* In single hash connection no tag is required (shared is used instead) Also added more tests to cover one corner case
* Dont guess mapping when emitter == receiver
* Auto wrap code with lua function declaration
* Added lupa to test-requirements, and added note in requirements
* Skip tests when lupa is not installed
* Self referenced computable inputs
* Test improvements for computable inputs
* Naming adjustments
* Added more complex case + added helpers to lua + code pretify
* Basic ops are working
* Upgrade pbr before installing solar
* Moved f2s doc, updated markdown
* Add lock on save
* Extended test for max
* First skeleton of Computable Inputs
* Fix sphinx warninngs
* Add pid to random bucket name
* Reorder gevent.monkey.patch_all()
* Docs are now generated in doc/build
* Use pip+git to install solar and solar-agent in Dockerfile
* Add f2s readme
* Use pbr to manage setup.py and setup.cfg
* Corrected documentation for style and content
* Add build utils
* Remove all f2s specific software from Dockerfile
* Add tox -e cover command
* Added some ansible handler descripton to docs
* Fixed some typos and sentences in transports docs
* Remove f2s directory
* Describe procedure of building deployment plan in docs
* Added basic transports docs
* Allow pythonhashseed to be random
* Rework tests to allow random ordering
* Link resource details in glossary
* Added more detailed resource documentation
* Fix flake8 import rules
* Fix inconsistencies after rebase
* Assign tasks based on roles generated by deployment info
* Add environment variable that will enable debug source in fsclient
* Take into account skipped (anchor-like) tasks
* Filter null values for hiera
* Generate inputs for compute and ceph-osd
* Treat NOOP or SUCCESS as visited and ERROR as blocker
* Add orch noop command to mark tasks as walked
* Add session wrapper to solar_resource task
* Add puppet_modules to every role vr
* Fix f2s/README
* Fix f2s/README
* Add README for f2s
* Add session for celery tasks
* add vms_conf val
* Disable mos_repos resource
* Change path to actions
* Fix tasks, ssh config and prefetching information from nailgun
* Add session to a client
* Fix inputs in resources
* Fixes in client and solar inputs
* Add additional packages and config for container
* Add proper client for nailgun in fsclient.py
* Use nailgun manager and solar-celery with f2s tag
* Save generated resources with inputs
* Save generated vrs based on existing roles
* Fix inconsistencies in clients
* Adjust docker to use f2s and small fixes
* Implement f2s/fsclient.py which will bootstrap solar stuff from fuel
* Add role data into events and events from prep.yaml to role_data
* Add mos repos and vr for genkeys
* Fix patches for noop tests and use new path to spec
* Add fuel entities and translate solar entities in order
* Add from_nailgun.py manager
* Add resource for keys generation and vr for copying stuff
* Generated inputs for role data
* Allow multiple managers
* Add manager globals.pp
* Add naive_sync driver which will copy resource from solar location
* Add naive manager mechanism and prefetch command
* Add values_from to vr generator
* Add values_from mechanism that will allow to guess mappings
* Save applied patches to generate inputs
* Create resoruces and events based on tasks
* Add g2vr command that will create resources and events
* Fix path to manifest
* Translate tasks into resources without inputs
* Use config.py:C in solar and fix compose file
* Run tests in parallel
* Added last info to orch docs
* Add examples docs, reformat add indexes
* Add install tag to celery playbook and use it in Dockerfile
* Install latest tox from pip
* Added tox requirement to base.yaml playbook
* Set hacking <0.11,>=0.10.0
* Run flake8 first. It will fail faster
* Mount ansible modules dir in celery container
* Move sphinx requirement from setup.py to venv in tox
* Orchestration documentation
* More flake8 fixes
* H306 doesn't support noqa, make empty lines instead
* Fix ansible import problem
* Add correct license to gevent_local
* More fixes for hacking
* License in test_execution_models
* License in test_nested
* Adjusted solar_torrent
* Added Apache 2 license to dblayer files
* Tox checks solar dir
* Slightly more Guidelines adjustments
* Flake8 fixes
* Update tox
* Add docs generated by sphinx
* Run autopep8
* Adjusted tests code to OpenStack Style Guidelines
* system_log adjusted to OpenStack Style Guidelines
* Adjusted events to OpenStack Style Guidelines
* Orchestration adjusted to OpenStack Style Guidelines
* Run autopep8
* Adjusted dblayer code for OpenStack Style Guidelines
* Solar glossary
* Ignore testrepository dir and fix coverage path
* Do not try to open directories
* Ignore python warrnings
* Fix docker scripts after moving solar and soalrd
* flake8 passes for dblayer
* pep8 for dblayer
* deleted docker compose from solar.yaml
* Remove backtrack_inputs from cli resource
* Use stopwait instead of stop in celery playbook
* Add py-test to requirements
* Use just py.test on travis
* Include README in sdist
* Update dir
* Move to tox and testrepository
* Removed solard installation from run.sh
* Install solar-agent in docker
* Fix run.sh. setup.py is in root dir now
* Fixed solar-agent transport after merge
* Install solar-agent from correct repos
* Do not run docker-compose on slaves
* custom-configs.yaml is deleted now
* solard => solar_agent in yamls
* Fix. Do not change dir
* Use docker compose to run riak, redis and celery
* Run all tests for solar
* Do not install redis. It will be run in container
* Renamed solard to solar_agent, moved transports and examples
* Removed solard from this repository
* Update paths after moving library to root directory
* Move solar lib to root directory
* Removed unused variable
* Fixed haproxy meta.yaml input type
* Added -rv (raw value) option to backtrack_inputs
* Nested keys are working properly right now
* One more test (haproxy config problem)
* Implementation + test for failing case
* Test for failing case
* Added session start / end to celery using hooks
* Install libffi-dev and libssl-dev in bootstrap
* Lookup vals only if config was found
* Fixed complex disconnects scenarios with new_db
* Remove peewee from test requirements
* Run all tests from solar dir
* Removed db configuration from dblayer conftest
* Fixed dblayer tests
* Don't add nested models into defined_models, and don't create buckets for them
* removed dblayer option from config
* Changed config keys
* Adjust travis conf
* Added exception when unsupported connection type
* Commented test for unused functionality (nested lists)
* Removed duplicated test
* More fixes in test_signals
* Set config path. Run only tests from test dir
* Restore resources fixture
* Add riak to requirements
* Allow to set major config in SOLAR_CONFIG env variable
* Make other config files optional
* Fix test_signals
* py.test handles sessions
* Rewrite config using dotted dict object and add loading from files
* Use solarproject repository
* Make redis run without password
* Fix run.sh to start celery
* Use config.py:C in solar and fix compose file
* Use docker compose
* Docker compose for development
* Basic config to provide dynamic path for resources
* Dockerfile and run.sh for solar-celery image
* Removed old db implementation
* Fix tests for load_by_tags command
* wait_finish routing will clear cache before updating state
* Remove update from yaml features
* Correct node add in controls
* Workaround for reading location_id for not existing resource in controls
* Fixed events tests
* remove event implementation
* Remove location_id/transports_id from disconnect
* Fix interface for generating pretty diff
* Fix system log api tests
* Added naive conflict resolution
* Clear cache before running fixtures for tests, not after
* After delete object is purged from cache
* Fix unit tests for VR
* Added assert for ensuring that session cache is always clear
* Added test for cache logic
* inmemory sqlite for tests
* test_resource works
* add repos configuration possibility for bareon
* The vagrant script is checking for presence of the original file so we should copy it instead of move
* Removed duplicate create_all()
* solar connections graph for new db
* Change URLs of PoC services to Mirantis forks
* connect and disconnect from CLI in new_db
* solar connections show for new_db
* Removed commented code
* Better delete_all for riak (loops and waits while)
* added backtrack value to cli
* add standalone_session_wrapper to all standalone python executions
* Created standalone_session_wrapper. Use it in standalone scripts
* Fixed solar resource show
* Building IBP images on solar 'master' instead of downloading them
* resource.to_dict could return proper inputs values. fixes solar resource show
* Don't add input index when it already exists
* Make puppet run on the new_db
* Speed improvement. One connect per resource
* Use multi_get in discard and revert
* Move args_to_update into separate function
* Fix duplicates in update filter and delete with lazy_save
* Fix revert
* Fix discard
* Removed comments
* Added index cache, fetching inputs values is faster
* Improved local cache usage (initialization), fixed for gevent case
* Don't force save events
* Load location_id instead of all args in controls.py
* Destroy input cache in disconnect
* Commented some debug stuff in singnals.py
* Filter None values in changes
* Fetch all nested childs using depth first traversal
* Add filter for Resource that will fetch only updated entities
* Added licenses to gevent stuff + added missing file
* Lowered concurrency for multiget
* Propert multi_get in system_log
* Set 10 concurrency in stage
* Fixed local usage
* Parallel input connect
* smart gevent patch import
* Reworked stage a bit (it's twice as fast right now)
* Added gevent pool that tracks parent (for cache)
* Gevent local implementation which tracks parent greenlet Slightly changed original gevent.local implementation
* Fix readme for provisioning
* Fixed hosts_file resources duplicates in riak examples
* Added disconnect possibility
* corrected usage of `as_dict` in system_log change
* Solution for hash to listhash mapping
* support list for other_name in connect
* Fixed librarian modules input schema
* Temporary added signals.connect
* Handle one more connection type (dict to hash_list)
* marked one test as xfail
* Added dict_to_list inputs connection
* Riak example works
* Added as_dict to inputs
* Fixed other value tracking
* sql_client support for delete
* Added possibility to fetch other value (required for transports and ip)
* Changed remove_all and implement it in sql_client
* removed debug print
* merging new db efforts
* Corrected method order in solar_models
* Corrected __iter__ for inputs
* Added passthrough input test
* More progress in new_db, still some minor things missing
* Removed duplicate hosts file resouce creation
* Use id instead of mac
* WIP but dshulyak needs it
* Clear database using ModelMeta from new backend and unmock events
* Mock resource connections and fix some stuff in stage/process
* Fix differences using sqlite backend
* Fix item.res in operations.py
* Use update instead of save for already created graph
* Fix several inconsistencies in interfaces
* Add CommitedResource model
* Use new LogItem interface in system_log module
* Add history field and preserve order by this field
* Add test for preserving order
* Add negative counter wrapper
* Auto-reindex composite field in case of changes
* Change bucket for every test
* Add *.svg to gitignore
* Add model for CompositeIndex and LogItems
* Add field for events
* Fix defaults for graphviz
* Conver save_graph/get_graph to riak client
* Implement childs/parents tasks
* Add Tasks model for riak db backend
* as_list for inputs
* Added comment that connections requires fixes
* Added save_lazy to resource.py
* connection graph / edges (not working because of networknx)
* Do not set duplicate tags
* load_all
* Resource show, resource show by tag, adding tags works
* Fixed inputs list tracking
* Resource creation works
* Added logic for nullable fields (transforms None => NONE)
* When field is str make it basestring
* Simple input connect update works
* dict list inputs are working
* Fixed typo
* Removed old code
* dict input with tags
* Invalid unpacking fixed
* simple to dict input works
* Run services under supervisor
* list to list + dict to dict inputs
* simple to list inputs are working
* dict_to_dict input works
* Add nat rules for solar-dev to provide internet connectivity for slaves
* Add workaround for bareon-api installation since pbr does not work with git sources
* Redesigned inputs again, this approach should be clean enough
* Allow comparsion StrInt with hex value of it
* Sync data in brareon-api before starting for provisioning
* Allow comparsion StrInt with hex value of it
* Unpin PyYAML version in requirements
* More inputs behaviour refactoring
* Some Resource integration with new backend
* Redesigned inputs flow
* Use new discovery scan service
* Enable logging for fuel-agent
* Added multi_get
* Fix minimize script
* Get partitioning information from bareon-api
* Use provisionin.json as template
* Use BatTransport instead of ssh to get discovery info
* Remove commented line from provision.sh
* Remove new provisioning.json template for not_provisioned_node resource
* Use "new" partition format during provisioning
* Get node data from ohai and put it into discovery
* removed debug print
* Naive tracking of index changes
* Improved tests
* Added save_lazy + changed how cache works
* Improved tests
* Added save_lazy + changed how cache works
* Wrapped json.dumps in sqlclient
* Wrapped json.dumps in sqlclient
* load + corrected imports
* resource.py updates
* Added to_dict to model
* Added actions to resource model
* has_own_input speedup
* Added test_nested file
* Added riak storage_backend as input (also riak.conf is jinja processed now)
* Fix virtualbox packer, add libvirt packer
* Use instance._get_index everywhere in solar_models
* Added some optimization comments
* Added updated field to Resource (using IndexedField)
* Added IndexedField
* Fix typo in IndexField
* Added session_start/session_end + added it to conftest
* Nested improvements (required for inputs)
* Introduced nested model
* Update tests
* Better defaults and from_dict behaviour
* New db layer core idea
* Fixed acitons path in riak_node
* Create and configure environment for jinja2 once
* Add json filter for jinja2 rendering engine
* Check if trg_templates_dir is not None
* Possibility to render ./templates with jinja
* changed effective_iv to effective_input_value
* Added solar autocomplete to bashrc
* Install ujson
* Added basic command to show effective input value
* Added update action to nova_compute_libvirt_puppet
* Added parameter for hosts_mapping
* Added big_riak example, it allow to create riak cluster with given size Could be used for perf testing
* Add example to change rabbitmq user for project like nova
* Add update action for nova_compute
* Add react event for nova_puppet -> nova_compute. Move acctions to meta
* More improvements
* Defining actions is meta.yaml works again
* Run == update
* Fixed naive cache behaviour for location_id and transports
* Fixed invalid else statement
* for more safety when not cache val then not store it
* Added cache delete for more safety
* Basic speedups, json=>json + naive local cache (enable with USE_CACHE=1)
* Run and remove are the same
* Run and remove are the same
* Make openstack.py example more granular
* Expose more from events module
* Allow to generate networks for more than 2 nodes
* Remove waiting for dnsmasq to be restarted since solar doesn't run several tasks on a single node
* Insert target of child node in react event
* Do not use hardcoded db port
* Update README
* Always use virtio
* Do not use the same name for all resources. It can be overwritten
* Add and fix update actions. changing db_port should work now
* Add bareon-api installation
* Added comment about exceptions in controls
* Added comments about circular imports
* Revert default limits
* make target=location_id on graph creation
* set target in orch to location_id if not given
* Temporary fix. target_based_rule is not working
* Specify vagrant-settings.yaml file in readme to enable provisioning
* Update readme with instruction, how to run provisioning
* Implement parallel node provisioning
* Refactor not_provisioned_nodes template
* Install discovery service
* Remove Fuel specific configuration from cloud-init configs
* Add node resource which can provisioned using solar
* Adjust for librarian
* Force mac address updates for slaves
* Copy action run to update
* Use librarian in main modules, add events to preserve correct order
* Make sure to use virtio bus when using libvirt
* Adjust modules for librarian
* Fixed remote action path in shell handler
* Remote and Local temp paths are different
* Use default token backend. KVS seems to be broken
* fix typo
* system_log graph in svg format
* Split openstack.py example into smaller functions. Start moving to librarian
* Adjust neutron_puppet for librarian
* Use rabbitma in librarian example
* Add mos_repos virtual resource and use it for ceph example
* Add apt_repo resource and managed_apt resources
* removed pytest cache
* Sync in temp dirs
* Sync templates and scripts in shell handler
* Dynamic import of optional transports in BAT
* Removed outdated commented code
* Use trackers from resource transports
* Adjust keystone_puppet to use Librarian resource
* Adjust rabbitmq_service to use Librarian resource
* Use connect from resource module
* Remove Librarian part from handler
* Do not install librarian. Hiera does not use redis anymore
* Changed torrent tracker to public
* Added some info how torrent transport works
* Create torrent dir if not exists (/vagrant/torrents)
* Correct readme for torrent example
* Added python-libtorrent to bootstrap base.yaml
* Sleep only in case if some tasks are not finished
* Use modules_path name
* Fix docs for the libvirt setup case
* Do not configure default route in SDN template
* Modify and rename config_dir input to modules_dir
* Make multiple networks only when PREPROVISIONED
* Remove the default route hack
* Add default route with metric 1 for vbox
* Fix dev nodes naming indexes
* Fix libvirt integration for multi-nets
* Del junk puts command
* Parametrize networks, box and provider
* Adjust parameters for vagrantfile YAML
* Change remote_file to use transports
* Add remote file resource which will download keys
* Restore Vagrantfile
* Add librarian example
* Remove reverse option
* Librarian resource
* Make vagrant silently run pxe nodes, without checks
* Change action temp files to contain resource.name
* Cleanup example
* Added torrent transport example (hosts_file)
* run.yaml for torrent adjusted
* bind_other works for Bat + Torrent
* Packer builds for Juno image
* Add README.md for library ceph installation example
* Add option to boot slave using pxe into bootstrap image
* Improvements in torrent transport
* Undo libvirt custom change
* Fix eth0->eth3 settings for nodes
* Rename SDN network template yml to yaml
* Use PuppetV2 for ceph_mon
* Use puppet modules for ceph_mon resource
* Add ceph_keys resource
* Implement library resource
* Refactor shell handler to use transports
* Add ceph_mon resource and example
* Use git provider to copy fuel-library
* Del node_network's ssh* inputs as not required
* Notifications does not work. Removing
* Fix dup IP in sdn template
* Fix node2 connection for sdn resource
* Fix network_schemas data
* Add two more networks
* Workaround l23network module issue
* Fix node network resource
* Add composition layer for nodes networking
* Update node_network resource
* Fix static IPs for mgmt bridges
* Add nodes network template
* Add node_network puppet resource
* Send travis notifications only if something changed
* Added travis + slack notification
* torrent transport resource improvements
* Correct meta.yaml name
* First stage porting torrent transport to new transorts
* Mirantis infra tracker as default
* Copied clean files from pr 117
* Fixed invalid obj usage in operations
* Fix forgotten exception in graph.py
* List input values support in VR
* Move time.sleep to CLI
* Change summary to print STATE: count for every possible state
* Add tests for graph.wait_finish routine
* Add wait option that will wait sometime for run to finish
* Only one worker for system_log queue
* Add default event for removal
* Stop celery instad of killing it. Kill did not kill children
* Use new images
* added info about orch stuff + openstack example
* Fix hosts_file/riak and solard examples
* Run full upgrade after upating repos
* Install vim-nox and mc packages
* Update README.md
* Install vagrant-vbguest plugin automaticly
* Changed click install instructions in README
* Install solard
* Use pip install instead of setup.py develop
* Removed `split` method from SolarRunResult
* Added missing attributes to RunResult
* SolarRunResult improved, fixes issues with puppet
* Fix transport usage in puupet handler
* Fix cli example - create nodes from VR - remove old command
* Command `connections clear_all` does not exists any more
* Added information about library
* Add newest module mysql_user from 2.0 release
* Add update action for mariadb_user
* Added update option for solard
* Added solard cli for starting it + adjusted actions Added child recycler in solard
* If mapping is not provided pass None to connect
* Fixed usage of `last` when no history file
* Correct riaks.py example template
* Add ip->db_host connection
* Change example to create nodes from VR
* Add info about snapshoter requirements
* fix typo in README
* Change hosts_file example to use Resource.connect_with_defaults
* Add connect_with_events method for resource.Resource class
* Fix example in README
* Removed "Launch standard deployment:" from "setup deployment env"
* Always convert to set
* Removed gevent from requirements.txt
* Changed comment in signals.py
* Added clarifications in usage section
* Add test for load by tags and add/remove tags
* Improve interface in cli for working with tags
* Add add_tags/remove_tags to orm and resource object
* Refactor interface for using resources with tags
* Sort connections before applying diff to them
* Stop solard before start
* .get for checking if $uuid
* Everything validates now
* Possibility to add solard in riak example
* Docummented signals.py magic, adjusted behaviour in some cases
* Changed start-stop-daemon arguments
* Adjustments in transports for latest changes
* Better debug in solard transport
* Added BAT transport
* Removed gevent requirement for now
* Removed some debug statement
* Added solard example
* Fixed problem with transports connected to another transports - When transports_id value is False no magic mappings - Added is_own to location_id, when set to False location_id is not mapped
* Solard resource with run action  (start-stop-daemon)
* Added SolarRunResult to wrap all Run transports results
* Fixed small bug when sending files
* properly handle files with 0 size in solard
* solard_transport added
* missing import
* transport_class defaults to tcp
* tcp_core with consts
* Added requirements
* Removed 'use_sudo' from solard core
* sudo should work now
* Changed logger
* Added licenses
* Added naive forks
* More changes
* Speed improvements
* Small adjustments
* Really PoC
* Fixed resource validate in CLI
* Move extracting resources from tags to seperate functions
* Add test for setting a location
* id is mandatory in metadata
* Use function to get tags
* Remove `>>>>` after merge
* Update README.md
* Add information about virtualbox guest vagrant plugin
* Pass args to create function
* Fix test
* Update Virtual Resource for riak cluster
* Split riak cluster into smaller pieces
* Add support for tags in Virtual Resources - resouces are now taged propperly - with_tags selector may be used to make connections for more than one resource
* Fix using tags in show command and log in tag
* Restore tags support in resources
* Add update action for hosts_file resource
* Fix quotes
* Add node tag
* Remove autogenerated tags, add default value
* Perform connect/disconnect only for relevant connections
* Add validation for uids and short how to in hosts_file/README.md
* Add tests for discard changes
* Parametrize discard command
* Add discard routine to system log cli and api
* Raise error if provided UID is not found in history
* Make connections for transports and location optional
* VR optimalization for riak cluster
* Add location key to Virtual Resources
* Guess mapping only if it is None
* Fix riak cluster virtual resource after removing ssh_* inputs
* Fix several inconsistencies with data and receiver input in case of hash
* Make removal of staged data only after commit
* Store events separetely from resource
* Use staged log in routine for building orchestration graph
* Add test for connected resources with revert of create/update
* Add support for revert of create
* Add support for revert of the removal
* Add base_path to log_item to revert removal
* Make signals between resources reversable
* Add methods to create graph of input thats is related to a single resource
* Snapshotter show outputs ALL snapshots
* Added comment about perf concerns
* openstack example to new transports
* hosts file example ported to new transports
* Removed old commented meta part
* fix tests for transports
* rsync transport ported to transports
* Riak example ported to new transports
* Uncommented riak debug actions
* If transports and location already mapped skip autoadd
* Removed ssh_* things from meta.yaml
* Added vr for nodes with transports
* Added comments in controversial parts
* Ansible template handler adjusted to transports_as_resource
* Added ip method to resource (based on location_id)
* ssh transport adjustments for transports_as_resoruce
* Late binding / lazy val for transports
* Still wip, but connections are ok
* Still wip
* location_id introduced, transports resources added
* Add coverage dir to .gitignore
* Fix interfaces in add_dep and add_react
* Allow VirtualResource only with `updates` section
* Raise error when static data is in list
* Add fixture file for tests
* Add updates section
* Add coverage icon
* Do not create virtualenv on travis
* corrected Vagrant to VB in docs
* added Travis CI status icon
* install pip-accel in run_tests.sh
* pip accel in tests instead of plain pip
* Fix typo
* Enable cache for pip
* corrected backticks and vagrant version
* Run coveralls from solar dir
* Run py.test from solar dir
* Run coveralls from solar/solar dir
* Add debug option
* Run coveralls for coverages stats
* Add coverage stats
* Add test for revert of value update
* Setup redis
* Add travis file
* Fix unused and forgotten imports in cli/resource.py
* Add tox.ini
* Add test for removal resource with connections
* Add thoughts about recursive removal of resource
* Add support for removal of resource
* Move VR for riak to riak example dir
* Add key name to an exception
* fix typo
* Use paren/child notation
* Add tests for virtual resources
* Require pytest-mock in tests
* Add resources fixture
* Add support for revert of a change
* Move adding connections to seperate function
* Update riak cluster example after changes is HAProxy resources
* Remove unused input `ports`
* Fix build_edges routing to take into account order between events
* Make riak unit test to fail
* Use db_related_field to store events for resource
* Adopt orm interface in events/api.py module
* Add EventDB object in orm.py module
* paravirt from config for master
* ORM: fix test_orm.py
* ORM: maneuvers with proper hash replacement with tags inside a list input
* ORM, signals: more tests
* ORM: backtrack_value_emitter tests, fix hash disconnection
* CLI: print out backtracked value
* Riak: yet another naming fix
* Improve VirtualResources - add support for events - refactor code after orm changes
* Riak cluster as VirtualResource
* Improve VirtualResources - support lists - allow to use jinja functions like 'range'
* Add pydot to requirements.txt
* Remove double mock
* riaks: fix names for haproxy
* riaks: node name fix
* Riaks: fix template example, small riaks.py fixes
* CLI: make 'last' the default for orch run-once and report
* connections graph as svg
* validation: don't print debugging stuff
* fixed dot generation for restricted names
* removed duplicate name in ro_node
* fixes #171 - Correcting documentation regarding requirements
* Paravirt provider is customizable from vagrant-settings.yaml
* Check minimal Vagrant version in Vagrantfile
* Removed old stuff
* Fixed haproxy resource
* Add option set number of CPUs
* ORM: change function property 'value' to better name 'as_set'
* tasks: fix save_graph call args
* CLI: more main.py cleanups
* First stage of porting
* Change riak hostnames mapping
* Speed improvements for libvirt
* CLI: minor fixes
* CLI: fix show_emitter_connections function, remove connections clear_all
* CLI: fix changes validate
* Vagrant-settings: add missin master/slaves image defaults
* Bootstrap README: move vagrant-settings to yaml file
* Add slaves/master images to vagrant-settings.yaml
* Example-bootstrap: added readme, sample Vagrantfile
* CLI: better graph
* CLI: color graph output, show values along edges
* CLI connections graph show
* signals: change tagging separator from : to |
* signals: add optional hash tagging parameter
* Cover graph api with tests and small cleanup
* Remove unused forwarded port
* Add vagrant-settings.yaml to .gitignore
* hosts example is now using recent hash changes
* Signals: fix hash update for multiple connections syntax with list
* Remove clearing connections. db.clear does the job
* Don't use dicts as default values for args
* Fix undeploy. Use new API
* Add support for hash inputs, enable multiple inputs to connect into one hash
* Events: no need for db_convert
* DBObjectMeta should wrap functions only once
* Add spaces
* Add option to choose sync_type
* Fixed schemas in haproxy / riak
* Remove db_convert in all_events function
* Remove forgotten function in tests
* Add unit tests to cover db usage by events module
* Fix hosts_file run action
* hosts.py: remove Connections.clear()
* Renamed default vagrant settings to follow yml=>yaml changes
* changed all yml occurrences to yaml
* Renamed all yml files to yaml
* keystone_service_endpoint test fix
* more ip => host fixes
* Testing fix
* Fix update and db interface used in commited data
* Testing fixes, added option to 'solar chages test' only one resource
* Remove '.value' from inputs in resource tests
* Resolve several problems after conflicts
* hiera.yaml: no Redis, use files instead
* ORM: bring back default events adding
* Libvirt support
* Bring back 'value' in input data, fix ansible_template .value
* Graph DB: upload hiera yaml configs for puppet
* ORM: resource, actions, handlers fixes
* Orch: fix wrong merge in graph.py
* ORM: no input validation on input create, validate only on demand
* Remove unittest where it is not needed
* signals.py: small code simplification
* ORM: move all db stuff from signals.py to the ORM
* ORM: backtrack value for relations
* Tests pass now
* ORM: add_input, more tests
* ORM: db relations work now, adding is possible
* Add missing orm.py and test_orm.py files
* Graph ORM: resource save/load works
* Fix tests for diff generation
* Restore default signals in signals.connect
* Fix usage of old signals module and some data interfaces
* Raw ssh transport added
* Dynamic import of DB backend
* Fix interface for system log module
* Fix wrong variable in riaks.py
* Use prefork backend with 3 workers
* Add target to parameters so that target based limit will be used
* Up network>=1.10 and use OrderedMultiDiGraph
* Show total delta time in report
* Add S: <start_time> E: <end_time> output to solar orch report
* Change 2nd celery worker to use gevent backend and example with sleeps
* Remove old resource/signals/observers stuff
* Fix usage of db interface in events module
* Use graph db interface for orchestration
* Fixes for more syntax errors/wrong arguments passing
* Add missing licenses
* Remove unused imports
* ansible_playbook.py: fix missing log import
* Events: add event load test
* riaks.py: no Connection clearing anymore
* Undefined variable in redis_db fixed
* Fix apply_config action in haproxy service: ip -> host
* Add basic validation for orch filter action
* Preserve original states if no filters supplied
* Reset filtered tasks on each filter action
* Add comments and disable specifying tasks directly
* Add tests with riak fixture
* Move two_path.yml example to solar/test
* Add filter command that will save graph after filters are applied
* Change traverse logic so it will ignore skipped predecessors
* Add more simple filters and two_path example
* Graph filtering helpers
* Fix validation for new graph db (test_validation.py passes now)
* Fixed delete when no keys in redis_graph_db, added test for resource load
* Cleanup codebase from code that isnt used or broken
* Redis backend passes test_resource.py and test_signals.py now
* Added readme in rsync transport about host key verification
* VBox Guest Additions automatic installation
* Neo4j fixes and improvements to fit the common Graph DB interface schema
* Added hosts_file_example
* Update lxc README
* Update README.md
* Make README more user friendly
* Add `bash` to command
* Update CLI example
* Make -n required for snapshot restore
* Compiled resources README and example simplification
* Update README.md
* Graph DB: resource fixes, added redis_graph_db
* Changed exception to comment in ansible_playbook
* Fix tests and requirements.txt for neo4j
* More README fixes
* README + template.py documentation fixes
* Add README for lxc example
* Move requiring mock to requirements.txt. It is used in cli
* Add required packages for run_tests.sh
* Remove mock from requirements. It is already added to requiremenets-test
* Remove unused ansible modules. Update resources
* Removed not needed puppet handler parts
* Sync /templates and /scripts dir from resource
* Delete main.yml
* README fixes
* Delete simple-deployment.yaml
* Delete Dockerfile
* Added description about SUCCESS state in riak example
* Fixed paths in riak README
* Added riak example README
* Move all examples to examples dir
* Rename riak example files
* Move riak examples to separate dir
* Rename example-puppet.py to openstack.py
* Update README
* Fixed renderer
* Moved ansible log from /tmp/ansible.log to /var/tmp/ansible.log
* Move example-puppet.py to examples dir and add README file
* Remove old stuff
* Remove outdated examples directory and create orch_fixtures with README
* fixed missing `
* Delete TODO.md
* Ansible log stuff together in solar-master.json
* Create ansible.log file during provisioning
* Remove cinder.yml from solar_resource
* Volume group changes - fix path - remove file at the end
* Add unit tests for system log item details
* Fix ansible.cfg file sync in Vagrantfile
* Add bootstrap/packer_cache to .gitignore
* Remove unused ansible playbooks
* main.yaml: subversion is in bootstrap already
* LXC: add svn to bootstrap, add comment on how to run to example-lxc.py
* snapshotter: vboxmanage poweroff and start (headless)
* example-puppet.py: admin_role_services explicit disconnect wasn't needed
* Address comments to lxc branch and fix several bugs with example-lxc.py
* Add remove action for Volum Group resource
* Snapshotter: fixes to vboxmanage (machineoutput), stop & resume vm for restore
* Use comments as tasks names
* Snapshotter: get machinereadable output if possible
* Volum Group resource
* Remove creating cinder volume from bootstrap
* Snapshotter: make an executable
* Snapshotter: fixes to 'show' list command
* SVNProvider downloads repository in resources-directory from config
* Create multiple containers and add rabbitmq resource on idx=28
* Add example-lxc.py and several fixes
* Add container_networks, ssh_key and vxlan_mesh resources
* Finish lxc container and lxc host resources
* Add SVN provider for downloading dependencies
* Lxc container resource
* Add resource for lxc_hosts and lxc_container
* Add .solar_cli_uids to .gitignore
* Testing: fix exception catching
* example-riaks-template.py: added a warning
* Better display of 'solar changes test' results
* Fix nova conductor resource's README
* Fix nova_generic_service resource
* Add nova generic service resource
* Add admin user as admin role to service tenant
* Added description to snapshots + echo fix
* Fix endpoint names for glance and cinder
* Print <name>:[<pos>] if item is a list, and only member is changed
* Unwrap diff structure till value is found
* More template.py and example-riaks-template improvements
* template.py - custom DSL for easier resource creation
* example-riaks-template.py stub
* Changed prints to click.echo
* Added utility script snapshotter
* Added better logic when reading defaults in Vagrantfile
* Added vagrant-settings.yml to gitignore
* Added possibility to ajdust Vagrantfile settings from yml file
* Fix import typo (cherry-pick of c69a4cda63073b02a97)
* Fix system_log and add compact view for changes
* Disable host key checking for ansible
* Add several helpers for creation of events and add events in puppet-example.py
* Move changes of the graph into single module (events/controls.py)
* Fix conditions in Dependency and React controls
* Add events into orchestration builder procedure
* Add several test cases for events builder
* Add cli for working with events api
* Add api for working with events
* Add AliasedGroup for click
* Add several tokens to control events between resources
* Add a draft for events hadling
* Remove outdated TODO from orchestration module
* Added comment about wget ... | bash things
* added missing haproxy_config file
* added missing update file in riak_node resource
* renamed haproxy actions (forgotten files)
* renamed haproxy actions
* Del duplicated puppet parameter
* Add detailed output for system log items
* Fix system_log and add compact view for changes
* Packer: fix-missing is always run now
* Vagrantfile: typo fix
* Vagrantfile: fix ansible_playbook_command ruby string formatting
* Ansible playbook: call with --verbose arg
* packer: cloud_archive apt-get update fixes
* Packer.io: add juno packages, fix for ansible cfg and logging
* Vagrantfile: add slave custom setups
* Vagrantfile: add libvirt paravirtualization by default (works for VB >= 5.0)
* Fix solar-master box
* Fix ml2 agent resource and compostion
* Add tags for rabbitmq_user resource
* W/a LP1490438 for ml2 plugin resource
* Neo4j: small fixes
* Sync path in ansible_template fixed
* Adjusted defaults in actions
* basic rsync sync transport
* Fix ml2 agent resource inputs
* Remove ovs plugin and agent
* Add composition later for ML2 plugin and agent
* Add neutron agent ovs with ml2 plugin resource
* Add neutron ml2 plugin resource
* Commented status of ansible_playbook handler
* Adjust ssh transport to latest changes
* Further work on pluggable transports
* Renamed handler_run|sync to transport_run|sync
* bind transports together
* fixup! AnsibleTemplate uses now `host` instead of `ip` as ansible `-hosts`
* AnsibleTemplate uses now `host` instead of `ip` as ansible `-hosts`
* still wip
* Neo4j: performance improvements
* PROFILE option added to example-puppet.py
* Virtual resources work now
* Remove duplicated code in virtual_resource.py
* example-puppet: don't use virtual resources, since they don't work yet
* Use json-ified values for inputs to prevent Neo4j errors about dict's etc
* Neo4j storage demo -- first draft
* First working version of Transports, puppet handler uses it
* Fixed incorrect 'how to'
* Add neutron agents to the composition
* Fix puppet id and module in meta.yaml
* Fix neutron conf for compute and nova_neutron resource
* added cgenie short howto
* Fix neutron db stamp head
* Removed not needed ip calc
* Implemented uids history for solar cli commands
* Implemented uids history for solar cli commands
* riak join fixes
* One more signals fix in riak example
* extended events in riak example
* haproxy update action
* small reorder in example-riaks
* More flexibility in signals.connect
* Further work on riak_example
* Added hosts_file resource
* Riak cluster 3 nodes example with haproxy on each node
* Added plan command
* Removed not needed files
* Added riak example with events
* Fix cycle when same node visited multiple times
* Idea 1
* Fixed import typo
* Fix deployment order for example puppet
* Fix system_log and add compact view for changes
* Disable host key checking for ansible
* Add solar-dev3 node for testing
* Bootstrap resource added
* Add neutron metadata agent resource
* Add several helpers for creation of events and add events in puppet-example.py
* Use namespaces for L3 & DHCP agents
* Add neutron L3 agent resource
* Add neutron DHCP agent resource
* Fix neutron stamp/upgrade head
* Moved pre-install stuff (ansible, pip) to separate script
* Move changes of the graph into single module (events/controls.py)
* Fix conditions in Dependency and React controls
* Add events into orchestration builder procedure
* Add several test cases for events builder
* Add cli for working with events api
* Add api for working with events
* Add AliasedGroup for click
* Add several tokens to control events between resources
* Add a draft for events hadling
* Remove outdated TODO from orchestration module
* Fix resource.py when value is not specified in meta.yaml
* Fix resource.py when value is not specified in meta.yaml
* Solar bootstarpping playbooks & packer.io guide
* Fix undef default
* Fix core plugin, remove external class ref for neutron
* Fix removed sql_con
* Add composition for OVS plugin/agent
* Remove deprecated/unused sql conn inputs
* Add neutron OVS agent resource
* Fix undef inputs for neutron
* Add neutron ovs plugin resource
* Remove external class ref for neutron server resource
* Decouple neutron resource
* Fix undef inputs for apache2
* Add apache puppet resource
* main.yml cleanup
* CLI: move DryRunExecutors to separate module
* Fix undefined inputs
* Validation: allow None for not required fields
* Refactor example-puppet.py actions
* Fix glance compostion manifest
* Decompose glance_api_servicers input
* Fix cinder_glance ref
* Add cinder volume ISCSI resource
* Add cinder_glance resource
* Nova API test.py with servers, flavors, images
* Fix glance api servers for nova compute
* TODO fix glance_api_servers for compute
* Add nova_neutron resource
* Fix libvirt inject password input
* Add nova comp libvirt resource
* Add exec post config and network refresh to n-compute
* Fix example puppet for multinode
* Fix nodes template
* Add compostion layer for node2 nova compute
* Add compute node2
* Add nova compute resource
* Add nova conductor resource
* Fix remove nova api resource
* Fix schema and db-sync condition
* Add missing post-config and nova-common
* Add missed admin_password for nova api resource
* Fix typo
* Add nova api puppet resource
* Fix cinder puppet module references
* Del nova_api after moved to nova_puppet resource
* Fix schema for nova puppet
* Add missing port parameter for nova puppet
* Create nova base puppet resource
* Add generic token validation function
* Fix glance/cinder test.py input names
* Disconnect uniq inputs for g-api/g-registry
* Fix remove actions for glance puppet resources
* Fix missing collation and encoding for mariadb_db
* Add compostion layer for glance registry puppet
* Add glance registry puppet resource
* Del duplicated params
* Fix glance keystone user role to admin
* Fix glance puppet classes dependencies
* Fix glance puppet
* Align glance_puppet with cinder_puppet and upstream
* Glance-puppet, not fully working yet
* Fix db_host connections for nova resource
* Add test-requirements.txt
* Upgrade all packages on nodes
* Add compostion layer for c-volume puppet
* Remove enabled and manage_services params for c-volume
* Add cinder volume puppet resource
* Remove enabled and manage_service for c-scheduler
* Fix cinder scheduler puppet params
* Add composition layer for cinder scheduler puppet
* Add cinder scheduler puppet resource
* Add OS_VOLUME_API_VERSION 2 to openrc
* Add fixes for celery executor and unit test
* Remove enabled and manage_service for c-api
* Fix undeploy and resource_name ref at cinder api
* Replace undef to empty strings for cinder api
* Fix package cinder dependency for cinder api
* Fix indentation and parameters
* Fix cinder api resource metadata
* Add a composition layer for cinder api
* Fix cinder api resource metadata
* Fix README.md
* Add cinder api puppet resource
* Add tests for rules and filtering chain
* Introduce NOOP state with unit tests for traversal
* Remove/Update outdated tests
* Integrate traversal/limits/execution in schedule action
* Cleanup traversal module
* Move traversal to separate module
* Add chain and rules for specifying limits
* Add missing db_host input to cinder puppet
* Fix db_host connections for cinder puppet
* Handle puppet error during remove as WARNING in example-puppet.py
* Use detailed error codes to handle puppet errors
* Clean keystone resources
* Add ansible to dependencies
* Update path in tests
* Create cinde-volume on all slaves
* Fix system_log commands in README.md
* Take into account only predecessors that are changed
* Use correct port numbers
* Update virtual resource for current branch
* Fix remove and resource name ref for cinder
* Add basica documentation for orchestration/system log usage
* Fix rabbitmq connection for cinder resource
* Neutron fixes
* Use default nodename when removing
* Fix cinder endpoint
* Replace undef to empty strings for cinder
* Fix README.md
* Remove nodename
* Fix admin role for cinder in services tenant
* change path
* add tag
* nova-api resource
* Fix cinder endpoint port connection
* Fix keystone_host connection for cinder endp
* Fix rabbit connections after rebase
* Align cinder connections
* Fix undef db conneciton for cinder
* Update a composition layer for cinder puppet
* Add keystone specific metadata to cinder
* Add README for cinder puppet resource
* Fix the cinder puppet parameters schema
* Add a composition layer for cinder puppet
* Add cinder puppet resource
* Keystone fixes
* Revert "Use detailed error codes to handle puppet errors"
* Use detailed error codes to handle puppet errors
* Puppet should inheir from correct TempFileHandler
* Fix variables names in action command
* Fix resource imports after rebase
* Fix update of logitem bug in system_log
* Prevent fabric from raising SystemExit in case of errors
* Use absolute path for resources
* Limit celery workers
* Prepare orchestration plan based on precedence scheduling
* Simplify information stored for history
* Improve ordering for stage/commit log
* Add several helpers for cleanup and fixes
* Add show command for a plan
* Add plan creation based on resources topological sort
* Send events to modules from base task class
* Move system log cli to separate module
* Add discovery of tasks stored in separate modules
* Add system_log module
* Add ordered redis data structure to store system log
* Add lost handler for puppet
* Use Ubuntu python-mock
* Fix rabbitmq startup
* Use deb package
* More example-puppet simplifications
* Stop execution by marking all tasks SKIPPED
* Small resources fixes
* example-puppet.py: reformat dicts to be more readable
* Rabbitmq remove action and fixes
* Add second rabbitmq
* Fix resource name
* Rabbimtq service from puppet
* Add openrc file
* Del duplicating connections
* Fix keystone admin connection
* Use /var/tmp for librarian puppet storage
* Add error handling for ansible_playbook handler
* Unify LocalAnsible and Ansible implementations
* Add local_ansible handler that will execute ansible-playbook locally
* Del duplicating connection
* Rebase on master
* Comment out cinder WIP
* Organize the simple deploy composition layer
* Fix include path for virtual_resource
* Fix puppet owner for librarian storage
* Update librarian puppet data storage path
* Add resources_compiled.py to gitignore
* Fix bools in puppet manifests
* Puppet handler fixes
* README.md: fix deprecated rs/ dirs and cli.py
* Add some README's to couple of resources
* README update
* Add keystone endpoint for ... keystone
* Virtual resources are files
* README updated with resources compiled
* solar connections graph fix, README updated
* Puppet: return codes TODO added
* Dry run: hash startswith mapping, added possibility to read from file
* Validation: add bool to supported schemas for the custom format
* Use FACTER_resource_name in shell variables
* validation.py: docstring added
* Remove accidentall commented 'python setup.py deploy'
* Vagrant: pip, setuptools upgraded to newest versions
* Python pip install fixes in main.yml
* Run puppet in debug mode
* Fix remove
* Clean forge entry
* Add forge "https://forge.puppetlabs.com to Puppetfile
* fix
* Use json for tasks serialization and run celery with sudo
* Remove unnecessary code for now with some cleanup
* Move orchestraton cli to solar namespace
* Move everything into solar subproject
* Add celery.yml playbook to configure celery
* Add several examples and improvements to orchestration
* Add soft stop support and update for deployment plan
* Add basic workflow into cli
* Basic API for orchestration
* Add example with timeout support
* Add timelimit and target examples
* Graph orchestration with celery
* keystone_service_endpoint test improvements
* keystone_service_endpoint: working with neutron, tests added
* Separate DryRunExecutor class, used also in 'solar run' command
* Resource action dry run: hash also call number
* resource action dry-run added with possibility to modify fabric return values
* neutron_puppet works now with upstream openstack manifests
* log.debug instead of print for resource tests
* keystone_user: added test for logging with token
* Keystone_puppet test added
* Various fixes for example.py and resource testing
* Puppet: lots of fixes & improvements
* Remove obsolete keystone_puppet file
* example-puppet: added nova-network
* Puppet: add symlink for newer hiera configs
* example-compiled-resources fix for missing field in glance_api_container
* Various fixes for example.py and resource testing
* Compiled resources: various code fixes to make example work
* testing: add warning about not tested resources
* Make dedicated example-compiled-resources.py file, keep original example.py
* cli.py: validate has optional --check-missing-connections
* example.py: keystone service pulled from git, not zip
* Find missing connections: remove duplicated pairs
* Validation of resource input value duplication added
* cli.py: resource create/connect/update alternative input formats
* cli.py: better printing during resource/connection creation
* README updated, fix for printing in resource.show
* cli.py: path validation, show resource by name option added
* Operations: fix updating of resource data with stage/commit
* Replace subprocess.call with fabric.api.local
* Update README
* README: extend sample cluster setup with update showoff
* Update documentation
* fixed repo name
* Testing: autonomous resource testing draft
* Replace subprocess.call with fabric.api.local
* Remove defaults
* openrc file resource
* Fix broken python-pip package, use easy_install
* Hiera fixes for trusty
* name -> resource_name
* Update example.py for virtual resources
* Move validation to seperate function and cli command. Add support for ResourceProviders
* Fix port
* Use juno image
* Fix swapped vars
* Make resource universal
* Remove debug
* Save resource vars to inventory. Change resource name in args
* Create connections after resource is created
* Puppet: nova, cinder -- not working yet
* Rabbitmq service: fix ports to be ints
* Use Fuel packages and change box to trusty
* Neutron is working now
* Add glance
* Remove unused node
* Add admin token, fix keystone_admin_port
* Validate created resources
* Print created resources
* Fix validation
* Puppet: neutron keystone credentials added
* Controller `role`
* Openstack base virtual resource helper
* Puppet: neutron works now with very simple config
* Glance virtual resources
* Keystone virtual resources
* Puppet: adjust git resources path
* Change name and password input
* Do not connect virtual resources
* Add virtual resource
* example-puppet.py typo fix
* Replace subprocess.call with fabric.api.local
* Move Puppet resources to outside repos
* Bring back keystone_* resources here
* GitProvider: do not clone when directory is present
* RemoteZipProvider added for resources, with a sample in example.py
* Resource providers added (with GitProvider as sample)
* hiera-redis added, single-master redis cluster
* Hiera storage showoff
* Puppet: Keystone remove action is working now
* Puppet: added port to keystone
* Keystone can now be deployed using Puppet
* Puppet handler added
* Compiled resources showoff
* Tests: mark test_update_list_resource as failing
* cli.py: better, color format of resource connections
* Stage: remove mocks, fix test_update_list_resource missing connect
* Redis: bulk read for get_list
* signals.py: better variable names
* connections graph: added end_with, simplified the algorithm
* Graph connections start_with: take subgraph
* cli.py: connections graph start_with option
* Redis: bulk save added
* Create resource for tests
* Added tests for propagated data and connections
* Remove xfail from tests
* Add test for rolling back one attribute
* Add conftest and other setup for testing
* Use full resource representation in staged data
* Remove file based databases
* Add tests with some fixes for changes stage procedure
* Removed print's, added logging
* Some fixes & cleanup
* Use solar/requirements.txt and setup solar in ansible playbook
* Redis: get rid of global CLIENTS variable
* load_all fix to reconnect all resources
* Use ansible docker module to remove services
* cli.py: refactor duplicated 'run' function
* cli.py: remove duplicated connect, sort functions alphabetically
* cli.py fixes & resource show colors added
* Remove cli.py, move all to solar/solar/cli.py, migrate all to click
* Signals: fix list observer not notifying on disconnect
* Remove commented code, Resource.actions refactor
* Resource: args_dict refactor
* Redis: proxied resource
* Redis: fix tests
* Resource: load method raises error if resource not found
* Redis: fix get_obj_resource undefined method
* state.py: some code removal
* Redis: state.py db fixes
* Redis: observer notify creates resource instance on demand
* Redis: connections are now saved to Redis DB
* Redis: first version
* Glance: add DB migration, fix timeout in example.py
* Fix example.sh clean command
* Scale up only config and service
* Add clean, clean_all, clean_local, start, scaleup commands
* Glance: better requests test for registry
* Glance API works now with HAProxy
* Remove hardcoded name, for different resources different names should be used
* Fix connect and disconnect cli commands
* Fixing update procedure and timeouts for services
* Add removal of keystone service and endpoint
* Fixing rollback and other operations
* Add error passing from ansible and handling it in commit procedure
* Glance: API is now accessible via HAProxy
* example.sh fixes
* Change password to mariadb
* Added atexit.register(flush) for DB and Connections
* Add mariadb user back
* Add mariadb_db resource back
* Fix conflicts
* Fix DB init
* Fix rabbitmq resources
* Fix reading resources from storage
* Vagrantfile fix, added python-keystoneclient to main.yml
* Remove state dir from git repository
* Added rabbitmq connections
* Fix example.py
* Added prefix meaningfull prefix for each resource instance
* Add generation of keystone user, role, tenant, endpoint
* Use highlevel deployment tool in example.sh
* Revert count of nodes in Vagrantfile back to 2
* Working mariadb -> keystone -> haproxy connections
* Fix remove actions
* Glance: Keystone API endpoint added
* Validation fix
* Cached file system and Connections flush
* Json.dumps fix for Obervable values
* Glance registry finally working
* Simulate removing keystone services
* Wait until rabbitmq is ready
* Better validation by taking resource.Resource subclasses
* Increase slave RAM to 1024 MB
* Change update to use remove with context and run with new
* glance: registry & api fixes
* Rabbitmq resources
* Forked ansible rabbitmq modules to run in docker
* Add path to custom ansible modules
* glance_registry_service works
* Move glance_service to glance_api_service resource
* Glance docker container is up
* Keystone related resources
* Fix mariadb container name in resource description
* 2 Vagrant slave machines
* 2 Vagrant slave machines
* Fixes to example.py and Vagrantfile
* Fixes to example.py and Vagrantfile
* Add execution of action on commiting changes
* Fix resource assignment, pass only values in args parameter
* Add information about action to commit log
* Rollback support by uid and last one
* Several stylistic changes
* Shortcut for commiting log items
* Add staging procedure
* Add entities for commited/staging pipeline
* Fix resource __repr__ function, to_dict is a function
* Update pip before running through requirements
* Glance: more configs, keystone_user fix
* Added glance keystone user resource instance
* Keystone Glance script, should be later moved to some other resource
* Glance /v2/images GET works now with glance_admin token
* Glance: add mariadb config
* Glance resource and deployment added
* Bring back init script for slaves
* example.py: undeploy action is functional
* wrap_resource fix
* Ansible: _render_inventory refactor
* example.py: deploy/undeploy actions
* Use base path in as resources directory
* example.py validation fixes
* cli.py: add option to print out resource as JSON
* Validation fixes
* Fix example.py after refactoring of storage
* Fix path to actions in handler
* Fixed tests
* Set default for WORKSPACE in run_tests.sh
* Fix tagging for resources
* Implemented a single storage for resources
* Cleaned file_system_db from unused code
* Remove x-README.md because it's merged with README.md
* Fix nodes examples
* Remove remove node to resource convertation
* Fix config after rebase
* Renamed clients.json to connections.yaml
* Create resource from node and connect it to resource during the assignment
* Use path to resources instances from config
* Fix example.py
* Cleanup of repository
* Updated README.md with information from x/README.md
* Restore TODO.md
* Add connect action
* Fix connection order in profile
* Remove attrs key from node_list
* Added jsonschema in requirements
* Added defaults for resources
* Add connections example
* Changed path to resources
* Remove x directory
* Added tags to resources
* Remove init script for slaves and remove copy-pasted nodes creation
* example.py fixes, tests added
* Fix example
* Move haproxy.cfg to haproxy resource templates
* Move remaining x stuff to main directory
* Docs: commit-log update
* Docs: commit log
* Fix nested jsonschema construction
* Fix test imports
* Added resource validation
* Replaced && with "&" and "," operators and replaced "||" with "|"
* Fix tags concatination for resources creation
* Update readme
* Implemented resources to nodes assignment
* Added assign command and implemented resources and nodes retrival
* Implemented tags filtration language
* Remove data action
* Fix imports for s/x/solar.core/
* Restored removed files after the latest merge
* Moved all of entities from x directory to solar core
* Fix encoding system in signals.py file
* Add run_tests.sh and x configs
* README, TODO moved to x
* example.py fix
* example.py fixes
* TODO updated
* Example deployement
* Run only resource specific tasks. Use host net mode by default
* Expose IPs from keystone service. Host mode not needed anymore
* Increase memory for vms. Mariadb needs 256 MB for internal use
* Use docker-py version 1.1.0
* Install python-mysqldb on all nodes
* Remove not resource related tasks from action
* Install docker-py on all dev vms
* Remove solar-de[5,6]
* Fix haproxy config removal
* Use absolute path for resource base dir
* Fix resource error message for not existing base resource
* Remove not needed code from haproxy_config resource
* Add command to keyston service resource and run it in host network
* Add user name to maria_db_user resource and use it instead of resource name
* Add db name to maria_db resource and use it instead of resource name
* Rename mariadb resource to mariadb_service
* Keystone user resource
* Keystone config resource
* Rename keystone resource to keystone_service
* Allow connections from any host
* Stylistic changes
* haproxy-deployment.yaml fixes according to new keystone inputs
* Add resource path to args
* ListObserver fixes
* Found, fixed and tested simple circular connection
* Implemented node based scope
* Increased debugging level for ansible
* Add containers actions to profile
* Implement ensure mechanism for containers
* Unsubscribe for simple observer when reconnecting to another observer
* ListObserver refactored to really represent list, not some dict
* Connections abstraction added over CLIENTS
* Test fixes, __eq__ added to Observer
* Keystone resource and action run
* Restart haproxy after config update
* Implemented idempotent docker run and user creation for mariadb
* Small disconnect refactor
* Fix unsubscribing for ListObserver
* Disconnect fix & test
* Old tests pass now
* Use yaml based instead of template based format for objects filtering
* Mariadb resource
* Add mariadb db and user resources
* Generated detailed connection graph, also output to PNG
* Use graphviz to render graph instead of matplotlib
* HAProxy config listen_port name fix
* Add listen_port to haproxy_config
* HAProxy deployment
* Successful HAProxy deployment with config file
* First working version of haproxy Docker
* Mariadb resource
* Fix resource.py input args so that validation is performed
* TODO updated
* Some signals/connections refactoring
* README: added info about 'deploy' command
* Vagrant fixes
* Extend haproxy test case
* Remove unused code
* haproxy-deployment haproxy connection fixes
* Deployment from YAML file, added tests for haproxy
* One more connect 'guessed' fix
* Connect -- provided mapping overrides guessed
* Connect event bubbling
* Add remove actions for keystone, mariadb, haproxy
* Changed resources to use templating language and tags, without hardcoded values
* Revert change which filters resources by profile's tags
* Set emitter's values to receiver upon connection
* List input-type introduced
* Fix renderer to serialize numbers without changes
* Fix python object searilization problem in yaml
* HAProxy deployment elements
* Implemented resources iterator in render engine
* Implemented first_with_tags method call
* Implemented "this" instruction for render engine
* Implement tags dependencies retrieving from templates
* Add data graph example with mariadb, haproxy, keystone
* Add networkx
* Remove modules/playbook.py
* Add a way to execute playbook only once
* Use resource as primary citizen in inventory
* TODO updated
* Disconnect receiver inputs before reconnecting to a new one
* Basic inventory with links to another values
* Fix guess_mapping mapping value
* Move handlers to seperate files. Add Ansible handler
* Basic tagging support
* TODO updated Please enter the commit message for your changes. Lines starting
* Remove from run task containers removing
* Add simple script to check that keystone works via haproxy
* Add keystone and haproxy installation into default profile template
* Add haproxy with hardcoded keystone
* Add keystone resource
* Remove pulling of image for docker resource
* User mariadb from kolla project
* Added missing main.yml, docker.yml for Astute, fixed graph outputting
* TODO updated
* Disconnect method added, refactoring, added CLI update
* Add new item to TODO
* Disconnect functionality
* Fix up the markup in README
* Add missed items in Dockerfile and kolla.yml
* Make paths in ansible extention fully qualified
* Fix extra 's' in template-:dir so that the value can be found
* TODO: added deployment task
* Basic CLI implementation with Click
* Guessing of resource keys, automatic connection
* Translate todo to english
* Networkx added, graph created from CLIENTS
* Fixed resource saving, option to load all resources
* CLIENTS and Resources are saved and loaded
* todo
* Actions are in separate directory
* Fix paths to match the README example
* Move all hardcoded paths to one config.yml file
* Init x
* Prepare for X
* DB storage path must be absolute, not relative
* Update path to vagrant box, because previous was removed
* apt hangs when it tries to install lxc-docker from ubuntu url
* It's not required to pull specific docker image for development env
* Do not run development env configuration for all VMs
* Default tags for resources should not be node specific
* Fixed multi-node deployment
* Do not override existen nodes during the discovery
* Add node specific tag generation
* Rename execute to discover in Discovery extension
* Remove profile from extensions list in profile template
* Fix readme formatting
* Fix configuration run in readme
* If none of nodes have resources the resource should be skipped even if it defined in the profile
* Change pudb to python-pudb
* Add instruction how to use solar into README file
* Configure command accepts id of the profile instead of file
* Add profile related commands
* All files which are stored by dirdbm are yaml files
* Introduced collections for storage
* Renamed storage directory from "created" to "storage"
* Move profile example into templates directory
* Improve orchestration examples and remove outdated schema
* Execute profile actions with cli
* Allow to provide actions to configure
* Add execute methods which will work by providing path
* Add single doc to refrence all of fuel's orchestration logics
* Add dockerfile to build container
* Some code cleanup
* Fixed resources duplication
* Fix template inventory identation
* Fix nodes generation for inventory
* Set vms names and less memory for the second node
* Add tags implementation
* Added tags for profile and filtering of the nodes
* Add second VM for distributed deployment
* Implemented extensions communication mechanism
* Add profile example
* Add configuration action
* Add additional actions store and store_from_file for storage
* Add notes how to handle primary/non-primary resources
* Add helpers for yaml files in utils
* Added extensions manager in the core
* Add exaples with 3 resources converted to ansible
* Moved ansible orchestrator into extension
* Add spec on building dependencies procedure
* Renamed resource to extension, added ID and NAME for each extension
* Add nodes resources generation for discovery
* Add ipython for development
* Create tmp directory
* Set vm name to solar-dev
* Created a single db entry point
* Add discovery
* Add init
* Moved dirdbm into separate third_party directory
* Add more commands to work with resources
* Add commands for creating resources
* Fix error when tmp directory is not created
* Ansible playbook fixes
* Add development info in README
* Setup development env for solar in vm
* Use pip to install the latest ansible version
* Add commands to work with prototype using ansible
* Install and run ansible in vm
* Fixed path in template
* Create initial structure of the project
* Add working debian jessie distribution
* Change Vagrantfile to use debian jessie
* Building block of the deployment
* Second iteration of specification
* Original schema
* Initial commit
