)]}'
{"specs/kilo/api-no-pipeline.rst":[{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"1e574e84c7e6ea663efe43d78698bf7763d880d6","unresolved":false,"context_lines":[{"line_number":31,"context_line":"The use of the pipeline creates dependencies in code, testing and"},{"line_number":32,"context_line":"installation that could be avoided. In code we must include the"},{"line_number":33,"context_line":"pipeline modules and classes. In testing and installation we must"},{"line_number":34,"context_line":"have a ``pipeline.yaml``."},{"line_number":35,"context_line":""},{"line_number":36,"context_line":"In the installation case, if there are multiple API servers, the"},{"line_number":37,"context_line":"``pipeline.yaml`` must be kept in sync across the hosts on which"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a961159_15391dd8","line":34,"updated":"2015-01-05 15:51:50.000000000","message":"+1","commit_id":"9764712ea0bbc2a4eeaf323918ec0db6ac0068a6"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"1e574e84c7e6ea663efe43d78698bf7763d880d6","unresolved":false,"context_lines":[{"line_number":44,"context_line":"API server at all. Instead when new samples are posted, format them"},{"line_number":45,"context_line":"as notifications and push them onto the notification bus to be"},{"line_number":46,"context_line":"retreived by the notification agent. The pipeline within that agent"},{"line_number":47,"context_line":"will then do any expected transformations."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"Alternatives"},{"line_number":50,"context_line":"------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a961159_d55795c8","line":47,"updated":"2015-01-05 15:51:50.000000000","message":"i like this. my only suggestion is since we\u0027re creating a notification in api service. can we decide here what the event_type name will be for this and what the schema will be?\n\nthe payload could be a Sample but event_type might be better as something ceilometer specific and not the actual sample name?","commit_id":"9764712ea0bbc2a4eeaf323918ec0db6ac0068a6"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"ffd8256e1bf618acdea8eaae883815a28d82c266","unresolved":false,"context_lines":[{"line_number":44,"context_line":"API server at all. Instead when new samples are posted, format them"},{"line_number":45,"context_line":"as notifications and push them onto the notification bus to be"},{"line_number":46,"context_line":"retreived by the notification agent. The pipeline within that agent"},{"line_number":47,"context_line":"will then do any expected transformations."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"Alternatives"},{"line_number":50,"context_line":"------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a961159_df16b248","line":47,"in_reply_to":"3a961159_d55795c8","updated":"2015-01-06 13:34:42.000000000","message":"Any suggestions? Such things are a bit beyond my experience. I often think \"there must be some rules and standards for this naming stuff\" and then I discover, no, people just wing it, and I get all fussy and confused.","commit_id":"9764712ea0bbc2a4eeaf323918ec0db6ac0068a6"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"c65ad8caadc6e946e654619284e5aee9dd979122","unresolved":false,"context_lines":[{"line_number":44,"context_line":"API server at all. Instead when new samples are posted, format them"},{"line_number":45,"context_line":"as notifications and push them onto the notification bus to be"},{"line_number":46,"context_line":"retreived by the notification agent. The pipeline within that agent"},{"line_number":47,"context_line":"will then do any expected transformations."},{"line_number":48,"context_line":""},{"line_number":49,"context_line":"Alternatives"},{"line_number":50,"context_line":"------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a961159_9fad0a34","line":47,"in_reply_to":"3a961159_df16b248","updated":"2015-01-06 14:15:02.000000000","message":"hmmm. so the only precedence we have (i think) is when we publish data to MQ from agents to collector... in that case we use whatever is define in conf option[1] (possibly with meter name append to it if per meter topics is enabled)\n\nwe\u0027d have to figure out some naming logic for ceilometermiddleware as well so i\u0027m open to anything... i guess we should keep relatively the same logic as other notifications in OpenStack: \u0027telemetry.*\u0027.... maybe telemetry.\u003csource\u003e.\u003cpayload_type\u003e? so telemtry.api.sample? and middleware could be telemetry.middleware.sample?\n\nthe payload, i think i\u0027m ok with anything. if we used sample.Sample as a payload that\u0027d be fine... or if you want to use a CADF or some other schema, that\u0027s fine.\n\n[1] https://github.com/openstack/ceilometer/blob/master/ceilometer/publisher/messaging.py#L47-L51","commit_id":"9764712ea0bbc2a4eeaf323918ec0db6ac0068a6"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"1e574e84c7e6ea663efe43d78698bf7763d880d6","unresolved":false,"context_lines":[{"line_number":90,"context_line":"---------------"},{"line_number":91,"context_line":""},{"line_number":92,"context_line":"Pipeline transformations that had been happening in the api server"},{"line_number":93,"context_line":"will happen in the notification agent."},{"line_number":94,"context_line":""},{"line_number":95,"context_line":"Other end user impact"},{"line_number":96,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a961159_75450142","line":93,"updated":"2015-01-05 15:51:50.000000000","message":"+1","commit_id":"9764712ea0bbc2a4eeaf323918ec0db6ac0068a6"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"1e574e84c7e6ea663efe43d78698bf7763d880d6","unresolved":false,"context_lines":[{"line_number":102,"context_line":""},{"line_number":103,"context_line":"This will potentially introduce some latency in the time between"},{"line_number":104,"context_line":"when a sample is posted to the API and when it lands in the"},{"line_number":105,"context_line":"data store."},{"line_number":106,"context_line":""},{"line_number":107,"context_line":"Other deployer impact"},{"line_number":108,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a961159_356299d9","line":105,"updated":"2015-01-05 15:51:50.000000000","message":"i think one potential gain is also the api service memory footprint might drop because we don\u0027t need to load pipeline anymore but i think it\u0027s safe to leave it out of spec because i\u0027ve no idea if this actually translate to real world gains.","commit_id":"9764712ea0bbc2a4eeaf323918ec0db6ac0068a6"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"1e574e84c7e6ea663efe43d78698bf7763d880d6","unresolved":false,"context_lines":[{"line_number":117,"context_line":"For example the time between posting a sample and being able to"},{"line_number":118,"context_line":"retrieve that sample may become more unpredictable. As it is already"},{"line_number":119,"context_line":"unpredictable any tests which rely on immediate retrieval are bad"},{"line_number":120,"context_line":"anyway, so we should fix that."},{"line_number":121,"context_line":""},{"line_number":122,"context_line":"Implementation"},{"line_number":123,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":1,"id":"3a961159_95e8ad17","line":120,"updated":"2015-01-05 15:51:50.000000000","message":"i think this might not be as big a concern. is it safe to just test that\n\n  1) notification is sent from ceilometer api and has correct payload schema\n  2) listener works on notification agent\n  3) ignore all the async issues between step 1 and 2.\n\nwould the above be safe?","commit_id":"9764712ea0bbc2a4eeaf323918ec0db6ac0068a6"}],"specs/liberty/api-no-pipeline.rst":[{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"cf1c05794049cdd5a554c3684516fa79f0559e0e","unresolved":false,"context_lines":[{"line_number":56,"context_line":"  service would only listen for ceilometer generated measurement"},{"line_number":57,"context_line":"  notifications, transform them according to a pipeline description"},{"line_number":58,"context_line":"  and then dispatch them onward to storage or other processing. This"},{"line_number":59,"context_line":"  model would skip the notification agent entirely."},{"line_number":60,"context_line":""},{"line_number":61,"context_line":"  Pro: extends the long term plan of decomposing to smaller pieces."},{"line_number":62,"context_line":"  Con: adds to the complexity of this change, increasing time to"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa32b979_2474c89a","line":59,"updated":"2015-06-14 08:48:09.000000000","message":"this sounds exactly like what the notification agent currently does.","commit_id":"ed4c9b775284e3a0856521bbd04732a31fe3a67a"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"c295e30428de864a8e88c6cca4d3e47b0f5c1fe7","unresolved":false,"context_lines":[{"line_number":56,"context_line":"  service would only listen for ceilometer generated measurement"},{"line_number":57,"context_line":"  notifications, transform them according to a pipeline description"},{"line_number":58,"context_line":"  and then dispatch them onward to storage or other processing. This"},{"line_number":59,"context_line":"  model would skip the notification agent entirely."},{"line_number":60,"context_line":""},{"line_number":61,"context_line":"  Pro: extends the long term plan of decomposing to smaller pieces."},{"line_number":62,"context_line":"  Con: adds to the complexity of this change, increasing time to"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa32b979_bf6545e2","line":59,"in_reply_to":"fa32b979_2474c89a","updated":"2015-06-14 11:16:44.000000000","message":"The notification agent listens to nova, neutron etc and does transformations of those notifications into samples and then transformations of those samples in the pipeline.\n\nThe thing I\u0027m suggesting (and I guess didn\u0027t make clear) is that there would be an independent process which is the sample-transformation-dispatcher-pipeline-agent, which listens to:\n\n* the api-server\n* the polling-agents\n* the notification-agents\n\neach of which have no transformation pipeline.\n\nMost of the time the transformation agent would just dispatch, but some of the time it would do aggregate transformations or calculations to create new meters.\n\n(Please note I am mostly riffing on these alternatives, there may be practical concerns which prevent them from being workable. However we need to work through them when we are exploring these decomposition concepts otherwise we might decompose along the wrong lines.)","commit_id":"ed4c9b775284e3a0856521bbd04732a31fe3a67a"},{"author":{"_account_id":8290,"name":"liusheng","email":"liusheng2048@gmail.com","username":"LiuSheng"},"change_message_id":"4a3690a9ca0a656df26aa0eea0848c4a4298a916","unresolved":false,"context_lines":[{"line_number":56,"context_line":"  service would only listen for ceilometer generated measurement"},{"line_number":57,"context_line":"  notifications, transform them according to a pipeline description"},{"line_number":58,"context_line":"  and then dispatch them onward to storage or other processing. This"},{"line_number":59,"context_line":"  model would skip the notification agent entirely."},{"line_number":60,"context_line":""},{"line_number":61,"context_line":"  Pro: extends the long term plan of decomposing to smaller pieces."},{"line_number":62,"context_line":"  Con: adds to the complexity of this change, increasing time to"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa32b979_90e5cba9","line":59,"in_reply_to":"fa32b979_bf6545e2","updated":"2015-06-15 01:30:24.000000000","message":"This idea is cool, it is more like a improved collector, but this will impact the polling-agents and notification-agents, especially there is a notification-agent splitting proposal now.","commit_id":"ed4c9b775284e3a0856521bbd04732a31fe3a67a"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"cf1c05794049cdd5a554c3684516fa79f0559e0e","unresolved":false,"context_lines":[{"line_number":87,"context_line":"---------------"},{"line_number":88,"context_line":""},{"line_number":89,"context_line":"No direct impact on the resources presented by the API, however"},{"line_number":90,"context_line":"samples created via the API will take a slightly different between"},{"line_number":91,"context_line":"the API and their eventual storage."},{"line_number":92,"context_line":""},{"line_number":93,"context_line":"Security impact"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa32b979_c465dce1","line":90,"updated":"2015-06-14 08:48:09.000000000","message":"s/different between/different path between/","commit_id":"ed4c9b775284e3a0856521bbd04732a31fe3a67a"},{"author":{"_account_id":8290,"name":"liusheng","email":"liusheng2048@gmail.com","username":"LiuSheng"},"change_message_id":"4a3690a9ca0a656df26aa0eea0848c4a4298a916","unresolved":false,"context_lines":[{"line_number":89,"context_line":"No direct impact on the resources presented by the API, however"},{"line_number":90,"context_line":"samples created via the API will take a slightly different between"},{"line_number":91,"context_line":"the API and their eventual storage."},{"line_number":92,"context_line":""},{"line_number":93,"context_line":"Security impact"},{"line_number":94,"context_line":"---------------"},{"line_number":95,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa32b979_bf9305a3","line":92,"updated":"2015-06-15 01:30:24.000000000","message":"Should we also consider the CLI script[1] for sending samples ?\n\n[1] http://git.openstack.org/cgit/openstack/ceilometer/tree/setup.cfg#n333","commit_id":"ed4c9b775284e3a0856521bbd04732a31fe3a67a"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"4200d94a28a3d151c37f571517fbdb1aee92b516","unresolved":false,"context_lines":[{"line_number":89,"context_line":"No direct impact on the resources presented by the API, however"},{"line_number":90,"context_line":"samples created via the API will take a slightly different between"},{"line_number":91,"context_line":"the API and their eventual storage."},{"line_number":92,"context_line":""},{"line_number":93,"context_line":"Security impact"},{"line_number":94,"context_line":"---------------"},{"line_number":95,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa32b979_2bb9f0e9","line":92,"in_reply_to":"fa32b979_bf9305a3","updated":"2015-06-15 02:00:52.000000000","message":"i can\u0027t remember the use case for that script. i think it was for testing. we can probably leave it as is for this blueprint.","commit_id":"ed4c9b775284e3a0856521bbd04732a31fe3a67a"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"cf1c05794049cdd5a554c3684516fa79f0559e0e","unresolved":false,"context_lines":[{"line_number":111,"context_line":""},{"line_number":112,"context_line":"This will potentially introduce some latency in the time between"},{"line_number":113,"context_line":"when a sample is posted to the API and when it lands in the"},{"line_number":114,"context_line":"data store."},{"line_number":115,"context_line":""},{"line_number":116,"context_line":"Other deployer impact"},{"line_number":117,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa32b979_6462d0da","line":114,"updated":"2015-06-14 08:48:09.000000000","message":"i don\u0027t think it adds much latency, by default the samples are published to queue and processed by collector already.\n\npro: it removes load off api service","commit_id":"ed4c9b775284e3a0856521bbd04732a31fe3a67a"},{"author":{"_account_id":8290,"name":"liusheng","email":"liusheng2048@gmail.com","username":"LiuSheng"},"change_message_id":"4a3690a9ca0a656df26aa0eea0848c4a4298a916","unresolved":false,"context_lines":[{"line_number":117,"context_line":"---------------------"},{"line_number":118,"context_line":""},{"line_number":119,"context_line":"Deployers of the API server will no longer need to deploy a"},{"line_number":120,"context_line":"``pipeline.yaml`` alongside the API server."},{"line_number":121,"context_line":""},{"line_number":122,"context_line":"Developer impact"},{"line_number":123,"context_line":"----------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa32b979_1f9b11cb","line":120,"updated":"2015-06-15 01:30:24.000000000","message":"The potential impact is the post_sample API will depend on the notification agent for receiving samples. If there is deployment that don\u0027t need notification data collection (no notification-agent ). that will break this api and users may won\u0027t know. since this api mainly used for testing(not sure), maybe this is not serious.","commit_id":"ed4c9b775284e3a0856521bbd04732a31fe3a67a"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"4200d94a28a3d151c37f571517fbdb1aee92b516","unresolved":false,"context_lines":[{"line_number":117,"context_line":"---------------------"},{"line_number":118,"context_line":""},{"line_number":119,"context_line":"Deployers of the API server will no longer need to deploy a"},{"line_number":120,"context_line":"``pipeline.yaml`` alongside the API server."},{"line_number":121,"context_line":""},{"line_number":122,"context_line":"Developer impact"},{"line_number":123,"context_line":"----------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"fa32b979_6bd5b841","line":120,"in_reply_to":"fa32b979_1f9b11cb","updated":"2015-06-15 02:00:52.000000000","message":"maybe we should make a direct path (ie. the api builds the samples and writes directly to db, doesn\u0027t have pipeline).\n\nthis would make the testing easier and keep the flexibility of deploying just api.\n\ni don\u0027t think that is a big deal as most deployers didn\u0027t realise there even was a pipeline in the api.","commit_id":"ed4c9b775284e3a0856521bbd04732a31fe3a67a"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"c37a880425ea7463cc6e6a7dde5ce4b443778d48","unresolved":false,"context_lines":[{"line_number":60,"context_line":""},{"line_number":61,"context_line":"  Pro: extends the long term plan of decomposing to smaller pieces."},{"line_number":62,"context_line":"  Con: adds to the complexity of this change, increasing time to"},{"line_number":63,"context_line":"  completion and risk of error."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Other alternatives which are related but not quite the same include:"},{"line_number":66,"context_line":""}],"source_content_type":"text/x-rst","patch_set":3,"id":"fa32b979_0e672526","line":63,"updated":"2015-06-16 03:58:04.000000000","message":"tbh, i sort of think this makes the notification agent \u0027too dumb\u0027?","commit_id":"0be7813f97ce25bf049f2f83bbd3a21c3925def6"},{"author":{"_account_id":7478,"name":"Nadya Shakhat","email":"nadmi4@gmail.com","username":"nprivalova"},"change_message_id":"4c4d9d838f0898fc9c6f710637061137afddde22","unresolved":false,"context_lines":[{"line_number":63,"context_line":"  completion and risk of error."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Other alternatives which are related but not quite the same include:"},{"line_number":66,"context_line":""},{"line_number":67,"context_line":"* The duplication of ``pipeline.yaml`` across multiple servers could"},{"line_number":68,"context_line":"  be addressed by the `Configuration via Data Store`_ spec, which"},{"line_number":69,"context_line":"  proposes to use a database as a central store of all"}],"source_content_type":"text/x-rst","patch_set":3,"id":"fa32b979_1de679b4","line":66,"updated":"2015-06-16 07:38:46.000000000","message":"Do we still want to mention this approach?","commit_id":"0be7813f97ce25bf049f2f83bbd3a21c3925def6"},{"author":{"_account_id":8290,"name":"liusheng","email":"liusheng2048@gmail.com","username":"LiuSheng"},"change_message_id":"5eee34682ef2badcc11a69d53dad4dd07e2c1756","unresolved":false,"context_lines":[{"line_number":63,"context_line":"  completion and risk of error."},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Other alternatives which are related but not quite the same include:"},{"line_number":66,"context_line":""},{"line_number":67,"context_line":"* The duplication of ``pipeline.yaml`` across multiple servers could"},{"line_number":68,"context_line":"  be addressed by the `Configuration via Data Store`_ spec, which"},{"line_number":69,"context_line":"  proposes to use a database as a central store of all"}],"source_content_type":"text/x-rst","patch_set":3,"id":"fa32b979_2893f650","line":66,"in_reply_to":"fa32b979_1de679b4","updated":"2015-06-16 08:40:48.000000000","message":"Hi, Nadya, yeah, maybe this alternative can be removed, for the centralized config proposal[1] has been abandoned after discussion.\n\n[1] https://review.openstack.org/#/c/188613/7","commit_id":"0be7813f97ce25bf049f2f83bbd3a21c3925def6"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"c37a880425ea7463cc6e6a7dde5ce4b443778d48","unresolved":false,"context_lines":[{"line_number":88,"context_line":""},{"line_number":89,"context_line":"No direct impact on the resources presented by the API, however"},{"line_number":90,"context_line":"samples created via the API will take a slightly different path between"},{"line_number":91,"context_line":"the API and their eventual storage."},{"line_number":92,"context_line":""},{"line_number":93,"context_line":"Security impact"},{"line_number":94,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"fa32b979_ce47fd75","line":91,"range":{"start_line":91,"start_character":34,"end_line":91,"end_character":35},"updated":"2015-06-16 03:58:04.000000000","message":"i think i\u0027m ok with everything as is. i think the only change i\u0027d add is maybe in the api, you can specify a \u0027direct\u0027 parameter when posting. doing so will avoid the requeue to notification agent.\n\nthis direct method could be use for test cases such as gabbi.","commit_id":"0be7813f97ce25bf049f2f83bbd3a21c3925def6"},{"author":{"_account_id":8290,"name":"liusheng","email":"liusheng2048@gmail.com","username":"LiuSheng"},"change_message_id":"d4dc6ddea3dbc9bd22eaca2811106dea6dab91e9","unresolved":false,"context_lines":[{"line_number":88,"context_line":""},{"line_number":89,"context_line":"No direct impact on the resources presented by the API, however"},{"line_number":90,"context_line":"samples created via the API will take a slightly different path between"},{"line_number":91,"context_line":"the API and their eventual storage."},{"line_number":92,"context_line":""},{"line_number":93,"context_line":"Security impact"},{"line_number":94,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"fa32b979_59eaa4ad","line":91,"in_reply_to":"fa32b979_701d5c0e","updated":"2015-06-17 01:33:49.000000000","message":"thanks, I got it :)","commit_id":"0be7813f97ce25bf049f2f83bbd3a21c3925def6"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"d9c49d8f17902a5610fae21d73562d5497d60a7e","unresolved":false,"context_lines":[{"line_number":88,"context_line":""},{"line_number":89,"context_line":"No direct impact on the resources presented by the API, however"},{"line_number":90,"context_line":"samples created via the API will take a slightly different path between"},{"line_number":91,"context_line":"the API and their eventual storage."},{"line_number":92,"context_line":""},{"line_number":93,"context_line":"Security impact"},{"line_number":94,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"fa32b979_701d5c0e","line":91,"in_reply_to":"fa32b979_b16d3c06","updated":"2015-06-16 12:14:35.000000000","message":"so i was thinking more that if you did a normal post (as is) it would now do what is being proposed: requeue, notification agent, collector, db\n\nas second option (this can be done in future cycle or now to keep changes minimal), we add in a param to api which allows you to define \u0027direct\u003dTrue\u0027. when direct\u003dTrue, it will not queue to notification agent but rather it would use the connection that the api service already has and would write straight into db. (no pipeline)... this could used for gabbi tests... or by users who don\u0027t care about pipeline for their POST samples.","commit_id":"0be7813f97ce25bf049f2f83bbd3a21c3925def6"},{"author":{"_account_id":8290,"name":"liusheng","email":"liusheng2048@gmail.com","username":"LiuSheng"},"change_message_id":"b8523d55ddd46f263c40e16adeb9544030f9aeca","unresolved":false,"context_lines":[{"line_number":88,"context_line":""},{"line_number":89,"context_line":"No direct impact on the resources presented by the API, however"},{"line_number":90,"context_line":"samples created via the API will take a slightly different path between"},{"line_number":91,"context_line":"the API and their eventual storage."},{"line_number":92,"context_line":""},{"line_number":93,"context_line":"Security impact"},{"line_number":94,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":3,"id":"fa32b979_b16d3c06","line":91,"in_reply_to":"fa32b979_ce47fd75","updated":"2015-06-16 06:58:40.000000000","message":"The pipline do 3 works: transforming samples, publishing samples, checking samples supported or not. if we add \u0027direct\u0027 mechanism, we will lost the check that whether meter supported or not, which depend on the pipline configuration.\n\nI am not sure should we need to limit the meter posted from users by the definition in pipline.yaml. but just for testing, maybe it is ok.","commit_id":"0be7813f97ce25bf049f2f83bbd3a21c3925def6"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"c37a880425ea7463cc6e6a7dde5ce4b443778d48","unresolved":false,"context_lines":[{"line_number":151,"context_line":"* Replace pipeline-based publishing in the API code with a notifier."},{"line_number":152,"context_line":"* Update API tests to deal with things like mocked pipelines that no"},{"line_number":153,"context_line":"  longer exist."},{"line_number":154,"context_line":"* your items here, based on discussion"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"Future lifecycle"},{"line_number":157,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":3,"id":"fa32b979_4e544d20","line":154,"range":{"start_line":154,"start_character":28,"end_line":154,"end_character":38},"updated":"2015-06-16 03:58:04.000000000","message":"remove and add \u0027api opt to do direct write to db\u0027","commit_id":"0be7813f97ce25bf049f2f83bbd3a21c3925def6"},{"author":{"_account_id":8290,"name":"liusheng","email":"liusheng2048@gmail.com","username":"LiuSheng"},"change_message_id":"b8523d55ddd46f263c40e16adeb9544030f9aeca","unresolved":false,"context_lines":[{"line_number":151,"context_line":"* Replace pipeline-based publishing in the API code with a notifier."},{"line_number":152,"context_line":"* Update API tests to deal with things like mocked pipelines that no"},{"line_number":153,"context_line":"  longer exist."},{"line_number":154,"context_line":"* your items here, based on discussion"},{"line_number":155,"context_line":""},{"line_number":156,"context_line":"Future lifecycle"},{"line_number":157,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":3,"id":"fa32b979_3c3023e8","line":154,"in_reply_to":"fa32b979_4e544d20","updated":"2015-06-16 06:58:40.000000000","message":"Done\n\nMaybe the direct publisher[1] may be reused for this.\n\n[1]http://git.openstack.org/cgit/openstack/ceilometer/tree/ceilometer/publisher/direct.py","commit_id":"0be7813f97ce25bf049f2f83bbd3a21c3925def6"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"2688ecbe0ea9f4d71b54cfec29cffb16257c98e5","unresolved":false,"context_lines":[{"line_number":161,"context_line":""},{"line_number":162,"context_line":"An optional boolean parameter \u0027direct\u0027 will be added to the post-samples API"},{"line_number":163,"context_line":"that allow users posting samples directly to storage and avoid notification"},{"line_number":164,"context_line":"agent processing(will skip pipline), this is mainly used for testing."},{"line_number":165,"context_line":""},{"line_number":166,"context_line":"Dependencies"},{"line_number":167,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"fa32b979_b218a51f","line":164,"range":{"start_line":164,"start_character":27,"end_line":164,"end_character":34},"updated":"2015-06-17 03:19:51.000000000","message":"s/pipline/pipeline/\n\nto be honest, you might want to do this first. i think it will make handling existing test cases a lot easier.","commit_id":"1e3fd6ef50d82e51b98a196e89924f96149e55a7"},{"author":{"_account_id":8290,"name":"liusheng","email":"liusheng2048@gmail.com","username":"LiuSheng"},"change_message_id":"f25ed650d5a943417813d5634b6eaa3b1f919501","unresolved":false,"context_lines":[{"line_number":161,"context_line":""},{"line_number":162,"context_line":"An optional boolean parameter \u0027direct\u0027 will be added to the post-samples API"},{"line_number":163,"context_line":"that allow users posting samples directly to storage and avoid notification"},{"line_number":164,"context_line":"agent processing(will skip pipline), this is mainly used for testing."},{"line_number":165,"context_line":""},{"line_number":166,"context_line":"Dependencies"},{"line_number":167,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"fa32b979_f204ed4a","line":164,"in_reply_to":"fa32b979_b218a51f","updated":"2015-06-17 03:39:20.000000000","message":"Yeah, I have readed some gabbi tests just now, this should be done firstly for tests :)\nI will upload a new ps.","commit_id":"1e3fd6ef50d82e51b98a196e89924f96149e55a7"}]}