)]}'
{"specs/juno/ironic-notifications.rst":[{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"0e4e4cd78cb0fb8d43c9cc63d86109333081d6a0","unresolved":false,"context_lines":[{"line_number":8,"context_line":"Ironic Notifications"},{"line_number":9,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/ceilometer/+spec/ironic-notifcations"},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Work is `in progress`_ to get Ironic to emit notifications from data provided by"},{"line_number":14,"context_line":"IPMI sensors (such as cpu temp and voltage). Ceilometer needs to be updated to"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_b9d2d4ce","line":11,"updated":"2014-06-19 17:07:41.000000000","message":"Non-blocking nitpick (only fix if you\u0027re uploading a new patch set anyway):\n\n  s/notifcations/notifications/","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"4a5ff72ff78ea2fad346659c8d9864c9fe390245","unresolved":false,"context_lines":[{"line_number":8,"context_line":"Ironic Notifications"},{"line_number":9,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"https://blueprints.launchpad.net/ceilometer/+spec/ironic-notifcations"},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Work is `in progress`_ to get Ironic to emit notifications from data provided by"},{"line_number":14,"context_line":"IPMI sensors (such as cpu temp and voltage). Ceilometer needs to be updated to"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_147d4fb2","line":11,"in_reply_to":"1ae5cdf2_b9d2d4ce","updated":"2014-06-19 17:28:14.000000000","message":"Done","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"0e4e4cd78cb0fb8d43c9cc63d86109333081d6a0","unresolved":false,"context_lines":[{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Work is `in progress`_ to get Ironic to emit notifications from data provided by"},{"line_number":14,"context_line":"IPMI sensors (such as cpu temp and voltage). Ceilometer needs to be updated to"},{"line_number":15,"context_line":"hear, transform and record these notifications."},{"line_number":16,"context_line":""},{"line_number":17,"context_line":".. _in progress: https://blueprints.launchpad.net/ironic/+spec/send-data-to-ceilometer"},{"line_number":18,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_39dec4da","line":15,"updated":"2014-06-19 17:07:41.000000000","message":"Maybe \"consume\" as opposed to \"hear\"?","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"4a5ff72ff78ea2fad346659c8d9864c9fe390245","unresolved":false,"context_lines":[{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Work is `in progress`_ to get Ironic to emit notifications from data provided by"},{"line_number":14,"context_line":"IPMI sensors (such as cpu temp and voltage). Ceilometer needs to be updated to"},{"line_number":15,"context_line":"hear, transform and record these notifications."},{"line_number":16,"context_line":""},{"line_number":17,"context_line":".. _in progress: https://blueprints.launchpad.net/ironic/+spec/send-data-to-ceilometer"},{"line_number":18,"context_line":""}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_744f7392","line":15,"in_reply_to":"1ae5cdf2_39dec4da","updated":"2014-06-19 17:28:14.000000000","message":"Done","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"0e4e4cd78cb0fb8d43c9cc63d86109333081d6a0","unresolved":false,"context_lines":[{"line_number":21,"context_line":""},{"line_number":22,"context_line":"The Ironic project is doing work to emit notifications on the message bus"},{"line_number":23,"context_line":"containing IPMI sensor data. If Ceilometer processes these notifications other"},{"line_number":24,"context_line":"services will be able use queries and alarms to monitor, bill and scale as"},{"line_number":25,"context_line":"required."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"It is not possible to simply dump a message on the bus and for Ceilometer to"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_b9b73484","line":24,"updated":"2014-06-19 17:07:41.000000000","message":"Monitor and drive scaling logic, defintely.\n\nNot sure about billing being driven by these data though?","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"4a5ff72ff78ea2fad346659c8d9864c9fe390245","unresolved":false,"context_lines":[{"line_number":21,"context_line":""},{"line_number":22,"context_line":"The Ironic project is doing work to emit notifications on the message bus"},{"line_number":23,"context_line":"containing IPMI sensor data. If Ceilometer processes these notifications other"},{"line_number":24,"context_line":"services will be able use queries and alarms to monitor, bill and scale as"},{"line_number":25,"context_line":"required."},{"line_number":26,"context_line":""},{"line_number":27,"context_line":"It is not possible to simply dump a message on the bus and for Ceilometer to"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_5418b783","line":24,"in_reply_to":"1ae5cdf2_b9b73484","updated":"2014-06-19 17:28:14.000000000","message":"For this spec (and others too, but this one especially) I\u0027m frequently casting out words to see what catches. That is it appears one of the most efficient ways of figuring out what the spec is supposed to specify is to write something that is incorrect and then it will be winnowed to something better.","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"0e4e4cd78cb0fb8d43c9cc63d86109333081d6a0","unresolved":false,"context_lines":[{"line_number":41,"context_line":""},{"line_number":42,"context_line":"Handling notifications is the standard and preferred method for gathering data"},{"line_number":43,"context_line":"into Ceilometer. The other option is polling which has known scalability"},{"line_number":44,"context_line":"issues."},{"line_number":45,"context_line":""},{"line_number":46,"context_line":"Data model impact"},{"line_number":47,"context_line":"-----------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_94a15ff9","line":44,"updated":"2014-06-19 17:07:41.000000000","message":"You could expand there on the known issue with over-frequent polling of IPMI, see related discussion:\n\n  http://lists.openstack.org/pipermail/openstack-dev/2014-March/031101.html\n\nSince Ironic \"owns\" the polling cycle under this approach, and can throttle access to the IPMI sensors, this is less likely to occur.\n\nAlso these notifications would be available for any other pipeline to pick up instead, e.g. a nagios plugin conceivably.","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"0e4e4cd78cb0fb8d43c9cc63d86109333081d6a0","unresolved":false,"context_lines":[{"line_number":62,"context_line":"Pipeline impact"},{"line_number":63,"context_line":"---------------"},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Unknown. Additional inspection of the expected data is required."},{"line_number":66,"context_line":""},{"line_number":67,"context_line":"Other end user impact"},{"line_number":68,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_34896b6f","line":65,"updated":"2014-06-19 17:07:41.000000000","message":"Not sure what this means, additional definitions in the pipeline.yaml, or?","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"4a5ff72ff78ea2fad346659c8d9864c9fe390245","unresolved":false,"context_lines":[{"line_number":62,"context_line":"Pipeline impact"},{"line_number":63,"context_line":"---------------"},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Unknown. Additional inspection of the expected data is required."},{"line_number":66,"context_line":""},{"line_number":67,"context_line":"Other end user impact"},{"line_number":68,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_4f82e252","line":65,"in_reply_to":"1ae5cdf2_34896b6f","updated":"2014-06-19 17:28:14.000000000","message":"At the time of writing that paragraph I had no information on what the data looked like so no clue whether the pipeline.yaml would be engaged or not. In fact I\u0027m still unclear on what it does. So yeah, \"Unknown\" fits the current bill.","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"3c027042071faa8f5d07fba0714cc64f92d09938","unresolved":false,"context_lines":[{"line_number":62,"context_line":"Pipeline impact"},{"line_number":63,"context_line":"---------------"},{"line_number":64,"context_line":""},{"line_number":65,"context_line":"Unknown. Additional inspection of the expected data is required."},{"line_number":66,"context_line":""},{"line_number":67,"context_line":"Other end user impact"},{"line_number":68,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_b779f8c6","line":65,"in_reply_to":"1ae5cdf2_4f82e252","updated":"2014-06-19 20:22:41.000000000","message":"Some background on the purpose of pipeline configuration:\n\n  http://docs.openstack.org/developer/ceilometer/configuration.html#pipelines","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"0e4e4cd78cb0fb8d43c9cc63d86109333081d6a0","unresolved":false,"context_lines":[{"line_number":72,"context_line":"Performance/Scalability Impacts"},{"line_number":73,"context_line":"-------------------------------"},{"line_number":74,"context_line":""},{"line_number":75,"context_line":"This implementation will add yet more writes to an already stressed system."},{"line_number":76,"context_line":""},{"line_number":77,"context_line":"Other deployer impact"},{"line_number":78,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_f4d9c383","line":75,"updated":"2014-06-19 17:07:41.000000000","message":"Which agent will host the notification handlers?\n\nThe central agent?\n\n(which is currently a SPoF, so maybe make the point that the central agent re-partition work is crucial for this to be workable in practice)","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"3c027042071faa8f5d07fba0714cc64f92d09938","unresolved":false,"context_lines":[{"line_number":72,"context_line":"Performance/Scalability Impacts"},{"line_number":73,"context_line":"-------------------------------"},{"line_number":74,"context_line":""},{"line_number":75,"context_line":"This implementation will add yet more writes to an already stressed system."},{"line_number":76,"context_line":""},{"line_number":77,"context_line":"Other deployer impact"},{"line_number":78,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_373c08e1","line":75,"in_reply_to":"1ae5cdf2_4fe90219","updated":"2014-06-19 20:22:41.000000000","message":"A notification handler is loaded by a particular agent, as determined by the stevedore entry points config.\n\nAnd now that I type this, I recall that it would make absolutely no sense to host a notification handler in the central agent when we have a notification agent for this very purpose:\n\n  https://github.com/openstack/ceilometer/blob/master/setup.cfg#L31\n\nSo please, ignore the noise :)","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"4a5ff72ff78ea2fad346659c8d9864c9fe390245","unresolved":false,"context_lines":[{"line_number":72,"context_line":"Performance/Scalability Impacts"},{"line_number":73,"context_line":"-------------------------------"},{"line_number":74,"context_line":""},{"line_number":75,"context_line":"This implementation will add yet more writes to an already stressed system."},{"line_number":76,"context_line":""},{"line_number":77,"context_line":"Other deployer impact"},{"line_number":78,"context_line":"---------------------"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_4fe90219","line":75,"in_reply_to":"1ae5cdf2_f4d9c383","updated":"2014-06-19 17:28:14.000000000","message":"These things are beyond my ken, tell me more please.","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"0e4e4cd78cb0fb8d43c9cc63d86109333081d6a0","unresolved":false,"context_lines":[{"line_number":118,"context_line":""},{"line_number":119,"context_line":"* Create sample query tests."},{"line_number":120,"context_line":""},{"line_number":121,"context_line":"[To what extent are alarms and/or events relevant here?]"},{"line_number":122,"context_line":""},{"line_number":123,"context_line":"Future lifecycle"},{"line_number":124,"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":2,"id":"1ae5cdf2_1468ef86","line":121,"updated":"2014-06-19 17:07:41.000000000","message":"Alarms are just based on statistics, regardless of source.\n\nEvents OTOH are notifications that are *selectively* persisted, so if ceilo is configured to do so, these events could be saved in that way. But I think unlikely, as the value has already be reaped from the sample transformation.","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"4a5ff72ff78ea2fad346659c8d9864c9fe390245","unresolved":false,"context_lines":[{"line_number":118,"context_line":""},{"line_number":119,"context_line":"* Create sample query tests."},{"line_number":120,"context_line":""},{"line_number":121,"context_line":"[To what extent are alarms and/or events relevant here?]"},{"line_number":122,"context_line":""},{"line_number":123,"context_line":"Future lifecycle"},{"line_number":124,"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":2,"id":"1ae5cdf2_afcb66a2","line":121,"in_reply_to":"1ae5cdf2_1468ef86","updated":"2014-06-19 17:28:14.000000000","message":"Done","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"0e4e4cd78cb0fb8d43c9cc63d86109333081d6a0","unresolved":false,"context_lines":[{"line_number":126,"context_line":"In the future new types of notifications are expected from the Ironic"},{"line_number":127,"context_line":"controller. These will need to be handled either by additional notification"},{"line_number":128,"context_line":"plugins or (hopefully) generic notification handling. The Ceilometer team will"},{"line_number":129,"context_line":"be responsible for collaborating with the Ironic team to insure these are"},{"line_number":130,"context_line":"handled smoothly."},{"line_number":131,"context_line":""},{"line_number":132,"context_line":"Dependencies"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_34fc0bb5","line":129,"updated":"2014-06-19 17:07:41.000000000","message":"Non-blocking nitpick (only fix if you\u0027re uploading a new patch set anyway):\n\n  s/insure/ensure/","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"4a5ff72ff78ea2fad346659c8d9864c9fe390245","unresolved":false,"context_lines":[{"line_number":126,"context_line":"In the future new types of notifications are expected from the Ironic"},{"line_number":127,"context_line":"controller. These will need to be handled either by additional notification"},{"line_number":128,"context_line":"plugins or (hopefully) generic notification handling. The Ceilometer team will"},{"line_number":129,"context_line":"be responsible for collaborating with the Ironic team to insure these are"},{"line_number":130,"context_line":"handled smoothly."},{"line_number":131,"context_line":""},{"line_number":132,"context_line":"Dependencies"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_0fba9a0a","line":129,"in_reply_to":"1ae5cdf2_34fc0bb5","updated":"2014-06-19 17:28:14.000000000","message":"Clearly I\u0027ve been reading the New York Time and New Yorker too much (they use the archaic form).","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"0e4e4cd78cb0fb8d43c9cc63d86109333081d6a0","unresolved":false,"context_lines":[{"line_number":141,"context_line":"Given that this addition adds support for hearing hardware sensor data it is"},{"line_number":142,"context_line":"not clear that any testing could be done in Tempest as there is [is this true?]"},{"line_number":143,"context_line":"no physical hardware to sense. As long as the sample data used in unit tests is"},{"line_number":144,"context_line":"sufficiently representative then the tests themselves ought to be as well."},{"line_number":145,"context_line":""},{"line_number":146,"context_line":"Documentation Impact"},{"line_number":147,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_34ea4bd8","line":144,"updated":"2014-06-19 17:07:41.000000000","message":"I wonder how Ironic in general is tested?\n\nIn the nova baremetal case, there used to a bm_posuer that could masquerade a baremetal node. That concept ,ight be useful in this regard also?","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"4a5ff72ff78ea2fad346659c8d9864c9fe390245","unresolved":false,"context_lines":[{"line_number":141,"context_line":"Given that this addition adds support for hearing hardware sensor data it is"},{"line_number":142,"context_line":"not clear that any testing could be done in Tempest as there is [is this true?]"},{"line_number":143,"context_line":"no physical hardware to sense. As long as the sample data used in unit tests is"},{"line_number":144,"context_line":"sufficiently representative then the tests themselves ought to be as well."},{"line_number":145,"context_line":""},{"line_number":146,"context_line":"Documentation Impact"},{"line_number":147,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_af59a6fe","line":144,"in_reply_to":"1ae5cdf2_34ea4bd8","updated":"2014-06-19 17:28:14.000000000","message":"Yeah, still waiting on some engagement with that side. I\u0027ve jumped in the conversation (\"review in progress\" linked below) to get the relationship flowing. I\u0027m hoping the outcome of that will inform this chunk.","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"3c027042071faa8f5d07fba0714cc64f92d09938","unresolved":false,"context_lines":[{"line_number":141,"context_line":"Given that this addition adds support for hearing hardware sensor data it is"},{"line_number":142,"context_line":"not clear that any testing could be done in Tempest as there is [is this true?]"},{"line_number":143,"context_line":"no physical hardware to sense. As long as the sample data used in unit tests is"},{"line_number":144,"context_line":"sufficiently representative then the tests themselves ought to be as well."},{"line_number":145,"context_line":""},{"line_number":146,"context_line":"Documentation Impact"},{"line_number":147,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":2,"id":"1ae5cdf2_928dea7e","line":144,"in_reply_to":"1ae5cdf2_af59a6fe","updated":"2014-06-19 20:22:41.000000000","message":"Wow! My typing is atrocious today! :(\n\nI meant:\n\n \"In the nova baremetal case, there used to be a bm_posuer that could\n  masquerade as a baremetal node. Might that concept be useful in this\n  regard also?\"\n\nSo I had agreed with lgomes that he\u0027d sit with you for a while at the Paris mid-cycle and step you through an Ironic setup. I\u0027m thinking now that it would make sense to kick off that interaction a bit earlier than July 2nd in order to get this spec fleshed out in good time. I\u0027ll reach out to him and organize something.","commit_id":"04da4c0d6d56a207e8cceb0c04207ad60ed0538e"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"880e14f1dce79e38a312dca5ec84028fc4d157eb","unresolved":false,"context_lines":[{"line_number":41,"context_line":"Handling notifications is the standard and preferred method for gathering data"},{"line_number":42,"context_line":"into Ceilometer. The other option is polling which has known scalability"},{"line_number":43,"context_line":"issues, including over-frequent `polling of IPMI`_. If Ironic is"},{"line_number":44,"context_line":"sending notifications it can control the cadence with which sensors"},{"line_number":45,"context_line":"are polled. Using notifications also allows other services to"},{"line_number":46,"context_line":"consume the sensor data."},{"line_number":47,"context_line":""}],"source_content_type":"text/x-rst","patch_set":5,"id":"1ae5cdf2_e2e1cd9b","line":44,"updated":"2014-06-22 10:14:11.000000000","message":"Suggestion: mention the credentials management aspect also, e.g.:\n\n ... it can continue to own the credentials for IPMI access and also control the cadence with which sensors are polled.","commit_id":"824a94560ff6304ce62caff6df2302e5a90d59bf"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"0606ba53e206bdcb60b2596fbb730e81d602ed97","unresolved":false,"context_lines":[{"line_number":41,"context_line":"Handling notifications is the standard and preferred method for gathering data"},{"line_number":42,"context_line":"into Ceilometer. The other option is polling which has known scalability"},{"line_number":43,"context_line":"issues, including over-frequent `polling of IPMI`_. If Ironic is"},{"line_number":44,"context_line":"sending notifications it can control the cadence with which sensors"},{"line_number":45,"context_line":"are polled. Using notifications also allows other services to"},{"line_number":46,"context_line":"consume the sensor data."},{"line_number":47,"context_line":""}],"source_content_type":"text/x-rst","patch_set":5,"id":"1ae5cdf2_ba0fd82e","line":44,"in_reply_to":"1ae5cdf2_e2e1cd9b","updated":"2014-06-23 10:38:22.000000000","message":"Done","commit_id":"824a94560ff6304ce62caff6df2302e5a90d59bf"},{"author":{"_account_id":8106,"name":"Haomeng,Wang","email":"wanghaomeng@gmail.com","username":"whaom"},"change_message_id":"59cec45a1de3ce8bedfce7b4c45ada258916c5f2","unresolved":false,"context_lines":[{"line_number":49,"context_line":""},{"line_number":50,"context_line":"Data model impact"},{"line_number":51,"context_line":"-----------------"},{"line_number":52,"context_line":""},{"line_number":53,"context_line":"None."},{"line_number":54,"context_line":""},{"line_number":55,"context_line":"REST API impact"}],"source_content_type":"text/x-rst","patch_set":5,"id":"1ae5cdf2_87f9f3c9","line":52,"updated":"2014-06-23 03:30:52.000000000","message":"I understand that we have no common data model change from Ceilometer part, however for our sensor data, how to model them into our Ceilometer data model - resource-\u003emeter-\u003esamples, how do you think if we need to add some sensor data model design here clearly, that is our Ceilometer and Ironic common data model I think.\n\nI have idea about how to do the sensor data mapping with Ceilometer to support the basic IPMI sensor data including \u0027Fan\u0027, \u0027Temperature\u0027, and \u0027Voltage\u0027 as below:\n\nresource -\u003e sensor level\nmeters -\u003e \u0027Fan\u0027, \u0027Temperature\u0027, and \u0027Voltage\u0027\nsample -\u003e one sensor data\n\nAnd we can discuss with this shared data model between us to get it works with both our projects, Ironic need this as input to refine the design I think.","commit_id":"824a94560ff6304ce62caff6df2302e5a90d59bf"},{"author":{"_account_id":8106,"name":"Haomeng,Wang","email":"wanghaomeng@gmail.com","username":"whaom"},"change_message_id":"59cec45a1de3ce8bedfce7b4c45ada258916c5f2","unresolved":false,"context_lines":[{"line_number":138,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"The primary dependency is work described in an `Ironic spec`_. Once"},{"line_number":141,"context_line":"that is implemented, notifications will be present on the bus."},{"line_number":142,"context_line":""},{"line_number":143,"context_line":"Testing"},{"line_number":144,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"1ae5cdf2_e742b739","line":141,"updated":"2014-06-23 03:30:52.000000000","message":"Ironic has the solution already, which defines the sensor data notification payload structure[1], here is the whole notification including header and payload example [2] which will be sent to Ceilometer, should we take this as reference as the interface between Ceilometer and Ironic? If this data model is acceptable from Ceilometer, we can go ahead to focus on the owning codes part I think.\n\n[1] \u0027Sensor Type\u0027 -\u003e \u0027Sensor ID\u0027 -\u003e \u0027key, value\u0027\n\n{\n    \u0027Sensor Type 1\u0027: {\n      \u0027Sensor ID 1\u0027: {\n        \u0027key1\u0027: \u0027value1\u0027,\n        \u0027key2\u0027: \u0027value2\u0027\n      },\n      \u0027Sensor ID 2\u0027: {\n        \u0027key1\u0027: \u0027value1\u0027,\n        \u0027key2\u0027: \u0027value2\u0027\n      }\n    },\n    \u0027Sensor Type 2\u0027: {\n      \u0027Sensor ID 3\u0027: {\n        \u0027key1\u0027: \u0027value1\u0027,\n        \u0027key2\u0027: \u0027value2\u0027\n      },\n      \u0027Sensor ID 4\u0027: {\n        \u0027key1\u0027: \u0027value1\u0027,\n        \u0027key2\u0027: \u0027value2\u0027\n      }\n    }\n    ...\n}\n\n\n[2] http://paste.openstack.org/show/68808/","commit_id":"824a94560ff6304ce62caff6df2302e5a90d59bf"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"b244341a9fb3f0d6a38118386c5d73a5af00ccf2","unresolved":false,"context_lines":[{"line_number":138,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"The primary dependency is work described in an `Ironic spec`_. Once"},{"line_number":141,"context_line":"that is implemented, notifications will be present on the bus."},{"line_number":142,"context_line":""},{"line_number":143,"context_line":"Testing"},{"line_number":144,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"1ae5cdf2_35c2bb34","line":141,"in_reply_to":"1ae5cdf2_32a02d91","updated":"2014-06-24 18:29:12.000000000","message":"@Haomeng:\n\nCould you clarify if any of these \u0027Voltage\u0027 readings ever change?\n\nIf there\u0027s no varying of the data, I\u0027m not sure that it makes sense to record it in ceilometer ... i.e. the mere existence of a voltage \"thing\" at a certain point in time is probably not an interesting fact that someone might query, I\u0027d suspect.\n\nBut I could be wrong. Can you think of any usecases where these static-looking data might be interesting in a retrospective sense?\n\nWhat would be *really* useful would be to classify each of the sensor data in:\n\n  http://paste.openstack.org/show/68808/\n\naccording to 3 criteria:\n\n* is there some varying numeric data involved in a particulat sensor?\n* is there some varying non-numeric data (e.g. state) involved?\n* or is this sensor data static and unchanging?","commit_id":"824a94560ff6304ce62caff6df2302e5a90d59bf"},{"author":{"_account_id":8106,"name":"Haomeng,Wang","email":"wanghaomeng@gmail.com","username":"whaom"},"change_message_id":"dbd546fbe5841c99eaeec22931170cd5f2ed794a","unresolved":false,"context_lines":[{"line_number":138,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"The primary dependency is work described in an `Ironic spec`_. Once"},{"line_number":141,"context_line":"that is implemented, notifications will be present on the bus."},{"line_number":142,"context_line":""},{"line_number":143,"context_line":"Testing"},{"line_number":144,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"1ae5cdf2_6e2d4845","line":141,"in_reply_to":"1ae5cdf2_35c2bb34","updated":"2014-06-25 02:30:33.000000000","message":"@Eoghan Glynn\n\nMake sense, yes, agree with you, for the no varying data, no need to record.\n\nThe sample data is got from my physical hardware with the command [1], the server type is, the ipmitool command output is [2]. \n\nI think the \u0027voltage\u0027 value should be changed for some special case, such as critical issue, but it is very diffcult to recreate, I will try to get more samples to get the difference about these data.\n\nAnother concern here for me is, for my sample [3], there is no data \u0027Nominal Reading\u0027 and \u0027Sensor Reading\u0027 data item for the Voltage sensor type, so I will get more different server type to have try.\n\nTo answer your 3 criteria question:\n\n  1. is there some varying numeric data involved in a particulat sensor?\n  A:For the varying number data, it should have  \u0027Nominal Reading\u0027 and \u0027Sensor Reading\u0027 data items in the sensor data, so we suppose if the sensor data contains  \u0027Nominal Reading\u0027 and \u0027Sensor Reading\u0027 fields, they are varying.\n\n  2. is there some varying non-numeric data (e.g. state) involved?\n  A:Yes, for example, some Temperature sensor \u0027Status\u0027 data value will be \u0027OK\u0027 or \u0027NS\u0027\n\n\n\n  3. or is this sensor data static and unchanging?\n  A: I think yes, but it is difficult to check for each sensor, and the guideline should be that if the sensor has no such  \u0027Nominal Reading\u0027 and \u0027Sensor Reading\u0027 data item, we suppose the sensor data is configuration data, will not changed for any time.\n\n\n\nNot sure if I answer you clearly, welcome your more comments here to discuss our sensor data content and formats etc, thank you.\n\n\n\n\n  [1] ipmitool -I lanplus -H 192.168.0.228 -U root -P password sdr -v\n\n  [2] http://paste.openstack.org/show/63267/\n\n  [3] http://paste.openstack.org/show/68808/","commit_id":"824a94560ff6304ce62caff6df2302e5a90d59bf"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"4b22c9be3cd228db8c6b5124cc9a04d07fb103b6","unresolved":false,"context_lines":[{"line_number":138,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"The primary dependency is work described in an `Ironic spec`_. Once"},{"line_number":141,"context_line":"that is implemented, notifications will be present on the bus."},{"line_number":142,"context_line":""},{"line_number":143,"context_line":"Testing"},{"line_number":144,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"1ae5cdf2_f32df4fe","line":141,"in_reply_to":"1ae5cdf2_8d65831f","updated":"2014-06-23 22:33:06.000000000","message":"@Haomeng:\n\nOK, I\u0027m showing my ignorance here, but which of these values is the *actual* numerical data that varies from one sensor-probe to the next?\n\nFor example is the \"Nominal Reading\u0027: \u00275325.000\u0027 the value that we really want to capture in ceilometer as the metered datapoint, and the other data are effectively metadata that are either static or infrequently changing?\n\nNote that the static or infrequently-changing can be captured in ceilometer also as resource metadata, but there should be a single distinguished numerical value that is the varying \"essence\" that we want to capture a trend for.","commit_id":"824a94560ff6304ce62caff6df2302e5a90d59bf"},{"author":{"_account_id":8106,"name":"Haomeng,Wang","email":"wanghaomeng@gmail.com","username":"whaom"},"change_message_id":"90002ce15d13e10ca4a2a2246005c7b90920eeb9","unresolved":false,"context_lines":[{"line_number":138,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"The primary dependency is work described in an `Ironic spec`_. Once"},{"line_number":141,"context_line":"that is implemented, notifications will be present on the bus."},{"line_number":142,"context_line":""},{"line_number":143,"context_line":"Testing"},{"line_number":144,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"1ae5cdf2_8d65831f","line":141,"in_reply_to":"1ae5cdf2_c7085131","updated":"2014-06-23 13:48:58.000000000","message":"Yes, I see your comments a moment ago, and get the feedback to you, can you check again?\n\nAnd here, you means we should make the keys useful, so any suggestion about how to change the keys names? I understand for the IPMI sensor names, current, I select \u0027Sensor ID\u0027 as the key, the reason is that I found the \u0027Entity ID\u0027 value is meaningful, but it is not UNIQUE, so how do you think if we should use \u0027Entity ID\u0027+\u0027Sensor ID\u0027 as the key to get it better? \n\nExample, for the previous sample data:\n\n\n \u0027Fan\u0027: {\n\t\t\t\u0027FAN MOD 2A RPM (0x34)\u0027: {\n\t\t\t  \u0027Status\u0027: \u0027ok\u0027,\n\t\t\t  \u0027Sensor Reading\u0027: \u00278250 (+/- 75) RPM\u0027,\n\t\t\t  \u0027Entity ID\u0027: \u00277.1 (System Board)\u0027,\n\t\t\t  \u0027Assertions Enabled\u0027: \u0027lcr-\u0027,\n\t\t\t  \u0027Normal Minimum\u0027: \u002710425.000\u0027,\n\t\t\t  \u0027Positive Hysteresis\u0027: \u0027375.000\u0027,\n\t\t\t  \u0027Assertion Events\u0027: \u0027\u0027,\n\t\t\t  \u0027Event Message Control\u0027: \u0027Per-threshold\u0027,\n\t\t\t  \u0027Normal Maximum\u0027: \u002714775.000\u0027,\n\t\t\t  \u0027Deassertions Enabled\u0027: \u0027lcr-\u0027,\n\t\t\t  \u0027Sensor Type (Analog)\u0027: \u0027Fan\u0027,\n\t\t\t  \u0027Lower critical\u0027: \u00274275.000\u0027,\n\t\t\t  \u0027Negative Hysteresis\u0027: \u0027375.000\u0027,\n\t\t\t  \u0027Threshold Read Mask\u0027: \u0027lcr\u0027,\n\t\t\t  \u0027Maximum sensor range\u0027: \u0027Unspecified\u0027,\n\t\t\t  \u0027Readable Thresholds\u0027: \u0027lcr\u0027,\n\t\t\t  \u0027Sensor ID\u0027: \u0027FAN MOD 2A RPM (0x34)\u0027,\n\t\t\t  \u0027Settable Thresholds\u0027: \u0027\u0027,\n\t\t\t  \u0027Minimum sensor range\u0027: \u0027Unspecified\u0027,\n\t\t\t  \u0027Nominal Reading\u0027: \u00275325.000\u0027\n\t\t\t},\n\t\t\t\u0027FAN MOD 2B RPM (0x35)\u0027: {\n\t\t\t  \u0027Status\u0027: \u0027ok\u0027,\n\t\t\t  \u0027Sensor Reading\u0027: \u00278175 (+/- 75) RPM\u0027,\n\t\t\t  \u0027Entity ID\u0027: \u00277.1 (System Board)\u0027,\n\t\t\t  \u0027Assertions Enabled\u0027: \u0027lcr-\u0027,\n\t\t\t  \u0027Normal Minimum\u0027: \u002710425.000\u0027,\n\t\t\t  \u0027Positive Hysteresis\u0027: \u0027375.000\u0027,\n\t\t\t  \u0027Assertion Events\u0027: \u0027\u0027,\n\t\t\t  \u0027Event Message Control\u0027: \u0027Per-threshold\u0027,\n\t\t\t  \u0027Normal Maximum\u0027: \u002714775.000\u0027,\n\t\t\t  \u0027Deassertions Enabled\u0027: \u0027lcr-\u0027,\n\t\t\t  \u0027Sensor Type (Analog)\u0027: \u0027Fan\u0027,\n\t\t\t  \u0027Lower critical\u0027: \u00274275.000\u0027,\n\t\t\t  \u0027Negative Hysteresis\u0027: \u0027375.000\u0027,\n\t\t\t  \u0027Threshold Read Mask\u0027: \u0027lcr\u0027,\n\t\t\t  \u0027Maximum sensor range\u0027: \u0027Unspecified\u0027,\n\t\t\t  \u0027Readable Thresholds\u0027: \u0027lcr\u0027,\n\t\t\t  \u0027Sensor ID\u0027: \u0027FAN MOD 2B RPM (0x35)\u0027,\n\t\t\t  \u0027Settable Thresholds\u0027: \u0027\u0027,\n\t\t\t  \u0027Minimum sensor range\u0027: \u0027Unspecified\u0027,\n\t\t\t  \u0027Nominal Reading\u0027: \u00277800.000\u0027\n\t\t\t},\n\n\nHow about these new Key names?\n\n\n \u0027Fan\u0027: {\n\t\t\t\u0027System Board - FAN MOD 2A RPM (0x34)\u0027: {\n\t\t\t  \u0027Status\u0027: \u0027ok\u0027,\n\t\t\t  \u0027Sensor Reading\u0027: \u00278250 (+/- 75) RPM\u0027,\n\t\t\t  \u0027Entity ID\u0027: \u00277.1 (System Board)\u0027,\n\t\t\t  \u0027Assertions Enabled\u0027: \u0027lcr-\u0027,\n\t\t\t  \u0027Normal Minimum\u0027: \u002710425.000\u0027,\n\t\t\t  \u0027Positive Hysteresis\u0027: \u0027375.000\u0027,\n\t\t\t  \u0027Assertion Events\u0027: \u0027\u0027,\n\t\t\t  \u0027Event Message Control\u0027: \u0027Per-threshold\u0027,\n\t\t\t  \u0027Normal Maximum\u0027: \u002714775.000\u0027,\n\t\t\t  \u0027Deassertions Enabled\u0027: \u0027lcr-\u0027,\n\t\t\t  \u0027Sensor Type (Analog)\u0027: \u0027Fan\u0027,\n\t\t\t  \u0027Lower critical\u0027: \u00274275.000\u0027,\n\t\t\t  \u0027Negative Hysteresis\u0027: \u0027375.000\u0027,\n\t\t\t  \u0027Threshold Read Mask\u0027: \u0027lcr\u0027,\n\t\t\t  \u0027Maximum sensor range\u0027: \u0027Unspecified\u0027,\n\t\t\t  \u0027Readable Thresholds\u0027: \u0027lcr\u0027,\n\t\t\t  \u0027Sensor ID\u0027: \u0027FAN MOD 2A RPM (0x34)\u0027,\n\t\t\t  \u0027Settable Thresholds\u0027: \u0027\u0027,\n\t\t\t  \u0027Minimum sensor range\u0027: \u0027Unspecified\u0027,\n\t\t\t  \u0027Nominal Reading\u0027: \u00275325.000\u0027\n\t\t\t},\n\t\t\t\u0027System Board - FAN MOD 2B RPM (0x35)\u0027: {\n\t\t\t  \u0027Status\u0027: \u0027ok\u0027,\n\t\t\t  \u0027Sensor Reading\u0027: \u00278175 (+/- 75) RPM\u0027,\n\t\t\t  \u0027Entity ID\u0027: \u00277.1 (System Board)\u0027,\n\t\t\t  \u0027Assertions Enabled\u0027: \u0027lcr-\u0027,\n\t\t\t  \u0027Normal Minimum\u0027: \u002710425.000\u0027,\n\t\t\t  \u0027Positive Hysteresis\u0027: \u0027375.000\u0027,\n\t\t\t  \u0027Assertion Events\u0027: \u0027\u0027,\n\t\t\t  \u0027Event Message Control\u0027: \u0027Per-threshold\u0027,\n\t\t\t  \u0027Normal Maximum\u0027: \u002714775.000\u0027,\n\t\t\t  \u0027Deassertions Enabled\u0027: \u0027lcr-\u0027,\n\t\t\t  \u0027Sensor Type (Analog)\u0027: \u0027Fan\u0027,\n\t\t\t  \u0027Lower critical\u0027: \u00274275.000\u0027,\n\t\t\t  \u0027Negative Hysteresis\u0027: \u0027375.000\u0027,\n\t\t\t  \u0027Threshold Read Mask\u0027: \u0027lcr\u0027,\n\t\t\t  \u0027Maximum sensor range\u0027: \u0027Unspecified\u0027,\n\t\t\t  \u0027Readable Thresholds\u0027: \u0027lcr\u0027,\n\t\t\t  \u0027Sensor ID\u0027: \u0027FAN MOD 2B RPM (0x35)\u0027,\n\t\t\t  \u0027Settable Thresholds\u0027: \u0027\u0027,\n\t\t\t  \u0027Minimum sensor range\u0027: \u0027Unspecified\u0027,\n\t\t\t  \u0027Nominal Reading\u0027: \u00277800.000\u0027\n\t\t\t},","commit_id":"824a94560ff6304ce62caff6df2302e5a90d59bf"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"0606ba53e206bdcb60b2596fbb730e81d602ed97","unresolved":false,"context_lines":[{"line_number":138,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"The primary dependency is work described in an `Ironic spec`_. Once"},{"line_number":141,"context_line":"that is implemented, notifications will be present on the bus."},{"line_number":142,"context_line":""},{"line_number":143,"context_line":"Testing"},{"line_number":144,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"1ae5cdf2_c7085131","line":141,"in_reply_to":"1ae5cdf2_e742b739","updated":"2014-06-23 10:38:22.000000000","message":"Have you seen my comments on https://review.openstack.org/#/c/72538/ in which I suggest that somewhere (not clear if in Ironic or Ceilometer) some adjustments may need to be made the the keys used in the data structure to make them useful to various consumers of the notifications?","commit_id":"824a94560ff6304ce62caff6df2302e5a90d59bf"},{"author":{"_account_id":8106,"name":"Haomeng,Wang","email":"wanghaomeng@gmail.com","username":"whaom"},"change_message_id":"bca4974089f0531cfb5e81fa189dc053d1df9f63","unresolved":false,"context_lines":[{"line_number":138,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":139,"context_line":""},{"line_number":140,"context_line":"The primary dependency is work described in an `Ironic spec`_. Once"},{"line_number":141,"context_line":"that is implemented, notifications will be present on the bus."},{"line_number":142,"context_line":""},{"line_number":143,"context_line":"Testing"},{"line_number":144,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d"}],"source_content_type":"text/x-rst","patch_set":5,"id":"1ae5cdf2_32a02d91","line":141,"in_reply_to":"1ae5cdf2_f32df4fe","updated":"2014-06-24 06:15:37.000000000","message":"@Eoghan:\n\nYes, we need identify sensor data main value, the for \u0027Fan\u0027 and \u0027Temperature\u0027 sensor data, the *actual* numerical data should be \u0027Sensor Reading\u0027 field value. \n\nHowever for other sensor types, such as \u0027Voltage\u0027 type, there is no such  \u0027Sensor Reading\u0027 as main value:\n\n\n  \u0027Voltage\u0027: {\n\t\t\t\u00275V Riser PG (0x1c)\u0027: {\n\t\t\t  \u0027States Asserted\u0027: \u0027Digital State\u0027,\n\t\t\t  \u0027Sensor ID\u0027: \u00275V Riser PG (0x1c)\u0027,\n\t\t\t  \u0027Entity ID\u0027: \u002716.1 (System Internal Expansion Board)\u0027,\n\t\t\t  \u0027Assertions Enabled\u0027: \u0027Digital State\u0027,\n\t\t\t  \u0027Sensor Type (Discrete)\u0027: \u0027Voltage\u0027,\n\t\t\t  \u0027Assertion Events\u0027: \u0027Digital State\u0027\n\t\t\t},\n\t\t\t\u00271.5V PXH PG (0x1b)\u0027: {\n\t\t\t  \u0027States Asserted\u0027: \u0027Digital State\u0027,\n\t\t\t  \u0027Sensor ID\u0027: \u00271.5V PXH PG (0x1b)\u0027,\n\t\t\t  \u0027Entity ID\u0027: \u002716.1 (System Internal Expansion Board)\u0027,\n\t\t\t  \u0027Assertions Enabled\u0027: \u0027Digital State\u0027,\n\t\t\t  \u0027Sensor Type (Discrete)\u0027: \u0027Voltage\u0027,\n\t\t\t  \u0027Assertion Events\u0027: \u0027Digital State\u0027\n\t\t\t},\n\n\n http://paste.openstack.org/show/68808/\n\nSo the question is, from the Ceilometer view, if the  *actual* numerical data is required for each sample, or just these metered datapoint are enough?","commit_id":"824a94560ff6304ce62caff6df2302e5a90d59bf"},{"author":{"_account_id":10202,"name":"Wan-yen Hsu","email":"wanyenhsu@gmail.com"},"change_message_id":"eaef13d766a6db9945713ff1fba25e61f78a982f","unresolved":false,"context_lines":[{"line_number":20,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"The Ironic project is doing work to emit notifications on the message bus"},{"line_number":23,"context_line":"containing IPMI sensor data. If Ceilometer processes these notifications other"},{"line_number":24,"context_line":"services will be able use queries and alarms to monitor and scale as required."},{"line_number":25,"context_line":""},{"line_number":26,"context_line":"It is not possible to simply dump a message on the bus and for Ceilometer to"}],"source_content_type":"text/x-rst","patch_set":6,"id":"1ae5cdf2_7d5846f9","line":23,"updated":"2014-06-26 23:38:06.000000000","message":"Wan-yen Hsu: Is this spec specifically for IPMI sensors or is it a generic spec for Ironic hardware sensors that can be collected via IPMI or other non-IPMI BMC (e.g., iLO or other vendor\u0027s BMC)? \r\n\r\n  Haomeng\u0027s sample sensor output is IPMITool specific.  Different BMC interface may return different sensor data for power/voltage or other sensor types.  For instance, HP iLO is able to return \"current power reading\" sensor as a numerical value.  It can also return other power sensors (e.g., capacity watts) as numical values.  These  will be userful information for some use cases.  These sensor data were not shown on IPMI Voltage sample output.\r\n\r\n   I felt that there is a need for a generic spec to descirbe the overall framework which is flexible to accomodate and process either IPMI or non-IPMI (e.g., iLO or other BMC) sensors.   For instance, a framework to allow vendor to process vendor specific hardware sensors in Ceilometer notification plugin.  My knowledge of Ceilometer is limited so if there is already a framework to do so, I will be happy to learn more about it.\r\n\r\n\r\n   If we further define more and more key or entitity names based on ipmitool, then it might become very difficult (sometimes it may not be possible) for other non-IPMI sensors to map.  Please take non-ipmi sensors into design consideration.  Perhaps Ironic extra_sensors and a frame work for vendor to process the extra_sensor  data in the notification plugin might help.   Thanks!\r\n\r\n\r\nWan-yen Hsu: Haomeng\u0027s sample sensor output is IPMITool specific.  Different BMC interface may return different sensor data for power/voltage.  For instance, HP iLO is able to return \"current power reading\" sensor as a numerical value.  It can also return other power sensors (e.g., capacity watts) as numical values.  These  will be userful information for some use cases.  Is this design spec specifically for IPMI sensors?  \r\n\r\n   I felt that there is a need for a generic spec to descirbe the overall framework which is flexible to accomodate and process either IPMI or non-IPMI (e.g., iLO or other BMC) sensors.   For instance, a framework to allow vendor to process vendor specific hardware sensors in Ceilometer notification plugin.  My knowledge of Ceilometer is limited so if there is already a framework to do so, I will be happy to learn more about it.\r\n\r\n   If we further define more and more key or entitity names based on ipmitool, then it might become very difficult (sometimes it may not be possible) for other non-IPMI sensors to map.  Please take non-ipmi sensors into design consideration.  Perhaps Ironic extra_sensors and a frame work for vendor to process the extra_sensor  data in the notification plugin might help.   Thanks!\r\nEditSaveCancelDiscard","commit_id":"b868a5ed66d5a4bc21bb905bc50de13d267f10e1"},{"author":{"_account_id":11564,"name":"Chris Dent","email":"cdent@anticdent.org","username":"chdent"},"change_message_id":"031c835a6bf6c0df9b1df17662e73c7c2fbf1e23","unresolved":false,"context_lines":[{"line_number":20,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"The Ironic project is doing work to emit notifications on the message bus"},{"line_number":23,"context_line":"containing IPMI sensor data. If Ceilometer processes these notifications other"},{"line_number":24,"context_line":"services will be able use queries and alarms to monitor and scale as required."},{"line_number":25,"context_line":""},{"line_number":26,"context_line":"It is not possible to simply dump a message on the bus and for Ceilometer to"}],"source_content_type":"text/x-rst","patch_set":6,"id":"dab895d6_0ec1772e","line":23,"in_reply_to":"1ae5cdf2_7d5846f9","updated":"2014-06-29 15:14:32.000000000","message":"\u003e  I felt that there is a need for a generic spec to descirbe the overall\n\u003e framework which is flexible to accomodate and process...\n\nI agree, there should be an abstraction which describes the kind of data that sensors like IPMITool gathers and it is those data which are then sent to ceilometer.","commit_id":"b868a5ed66d5a4bc21bb905bc50de13d267f10e1"},{"author":{"_account_id":8106,"name":"Haomeng,Wang","email":"wanghaomeng@gmail.com","username":"whaom"},"change_message_id":"15d4d98c150f4ea3cb43a92c815c1a80063e4f57","unresolved":false,"context_lines":[{"line_number":20,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"The Ironic project is doing work to emit notifications on the message bus"},{"line_number":23,"context_line":"containing IPMI sensor data. If Ceilometer processes these notifications other"},{"line_number":24,"context_line":"services will be able use queries and alarms to monitor and scale as required."},{"line_number":25,"context_line":""},{"line_number":26,"context_line":"It is not possible to simply dump a message on the bus and for Ceilometer to"}],"source_content_type":"text/x-rst","patch_set":6,"id":"dab895d6_7b7c7273","line":23,"in_reply_to":"1ae5cdf2_7d5846f9","updated":"2014-06-30 00:00:55.000000000","message":"@Wan-yen,\n\nWe defined the common data structure \u0027Sensor Type\u0027 -\u003e \u0027Sensor ID\u0027 -\u003e \u0027Field\u0027:\u0027Value\u0027 which will be fine with all hardware sensors for both IPMI or non-IPMI. The sensor current value should be find in \u0027Sensor Reading\u0027 field.\n\nAnd from Ironic part, we just implement the *get_sensors_data* interface method for our IPMITool driver, so that means, our Ironic support other driver such as HP iLO driver to get the sensor data, there is the plugin framework based on different driver in Ironic already.\n\nSo we can create more driver to cover non-ipmi sensor data collection to return to Ceilometer same common data structure \u0027Sensor Type\u0027 -\u003e \u0027Sensor ID\u0027 -\u003e \u0027Field\u0027:\u0027Value\u0027, and make the necessary formatting to make sure the same \u0027Sensor Reading\u0027 field to hold the sensor current reading data for all drivers.","commit_id":"b868a5ed66d5a4bc21bb905bc50de13d267f10e1"},{"author":{"_account_id":2284,"name":"Eoghan Glynn","email":"eglynn@redhat.com","username":"eglynn"},"change_message_id":"4a8994e5cedc3920f7e6a6dd8f77ddd37e5d5b5e","unresolved":false,"context_lines":[{"line_number":20,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"The Ironic project is doing work to emit notifications on the message bus"},{"line_number":23,"context_line":"containing IPMI sensor data. If Ceilometer processes these notifications other"},{"line_number":24,"context_line":"services will be able use queries and alarms to monitor and scale as required."},{"line_number":25,"context_line":""},{"line_number":26,"context_line":"It is not possible to simply dump a message on the bus and for Ceilometer to"}],"source_content_type":"text/x-rst","patch_set":6,"id":"1ae5cdf2_34ddebdd","line":23,"in_reply_to":"1ae5cdf2_7d5846f9","updated":"2014-06-27 10:44:29.000000000","message":"See Haomeng\u0027s response to a similar question posed on the corresponding Ironic review:\n\n  https://review.openstack.org/#/c/102435/16/specs/juno/send-data-to-ceilometer.rst\n\nThe issue of IPMI specificity is really a concern for the Ironic side of the equation, i.e. ceilometer is just consuming whatever data Ironic provides us with.","commit_id":"b868a5ed66d5a4bc21bb905bc50de13d267f10e1"},{"author":{"_account_id":8106,"name":"Haomeng,Wang","email":"wanghaomeng@gmail.com","username":"whaom"},"change_message_id":"acf4418e4d0f6ffd97691a73bf00616e29716751","unresolved":false,"context_lines":[{"line_number":169,"context_line":""},{"line_number":170,"context_line":".. _Ironic spec: https://blueprints.launchpad.net/ironic/+spec/send-data-to-ceilometer"},{"line_number":171,"context_line":".. _Review in progress: https://review.openstack.org/#/c/72538/"},{"line_number":172,"context_line":".. _Sample data: http://paste.openstack.org/show/68808/"}],"source_content_type":"text/x-rst","patch_set":6,"id":"1ae5cdf2_8f686af5","line":172,"updated":"2014-06-27 11:31:25.000000000","message":"Ironic change the spec, the notification to be sent to Ceilometer will not cover the sensors which has no \u0027Sensor Reading\u0027 field, because it is non-varying data, will not be recorded by Ceilometer.\n\n\n[1] http://paste.openstack.org/show/85053/","commit_id":"b868a5ed66d5a4bc21bb905bc50de13d267f10e1"},{"author":{"_account_id":9562,"name":"Ildiko Vancsa","email":"ildiko.vancsa@gmail.com","username":"ildikov"},"change_message_id":"5c69b4337b976a8cfb33f7787c23033569a3a1ad","unresolved":false,"context_lines":[{"line_number":169,"context_line":""},{"line_number":170,"context_line":".. _Ironic spec: https://blueprints.launchpad.net/ironic/+spec/send-data-to-ceilometer"},{"line_number":171,"context_line":".. _Review in progress: https://review.openstack.org/#/c/72538/"},{"line_number":172,"context_line":".. _Sample data: http://paste.openstack.org/show/68808/"}],"source_content_type":"text/x-rst","patch_set":6,"id":"dab895d6_46e505a7","line":172,"in_reply_to":"1ae5cdf2_8f686af5","updated":"2014-06-30 12:27:10.000000000","message":"Would be nice to include this link in the spec, to have every info in one place once if it is approved.","commit_id":"b868a5ed66d5a4bc21bb905bc50de13d267f10e1"}]}