)]}'
{"ceilometer/service.py":[{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"fbb634187dfbb11df513b69f07b9b398aaba3cbf","unresolved":false,"context_lines":[{"line_number":21,"context_line":"from oslo_config import cfg"},{"line_number":22,"context_line":"import oslo_i18n"},{"line_number":23,"context_line":"from oslo_log import log"},{"line_number":24,"context_line":"from oslo_reports import guru_meditation_report as gmr"},{"line_number":25,"context_line":""},{"line_number":26,"context_line":"from ceilometer.i18n import _"},{"line_number":27,"context_line":"from ceilometer import messaging"}],"source_content_type":"text/x-python","patch_set":5,"id":"1a4dcd0f_9c6861bc","line":24,"updated":"2015-08-10 13:58:14.000000000","message":"what\u0027s the affect on memory when this is loaded, maybe this should be conditionally loaded. ie load with:\n\n  gmr.TextGuruMeditation.setup_autorun(version)","commit_id":"0381f6c8bef70f4de862a9c882443f406a21b962"},{"author":{"_account_id":8290,"name":"liusheng","email":"liusheng2048@gmail.com","username":"LiuSheng"},"change_message_id":"d009ab678d1666f39d260c899cfb66ef1e1548d1","unresolved":false,"context_lines":[{"line_number":21,"context_line":"from oslo_config import cfg"},{"line_number":22,"context_line":"import oslo_i18n"},{"line_number":23,"context_line":"from oslo_log import log"},{"line_number":24,"context_line":"from oslo_reports import guru_meditation_report as gmr"},{"line_number":25,"context_line":""},{"line_number":26,"context_line":"from ceilometer.i18n import _"},{"line_number":27,"context_line":"from ceilometer import messaging"}],"source_content_type":"text/x-python","patch_set":5,"id":"1a4dcd0f_fa9389f8","line":24,"in_reply_to":"1a4dcd0f_9c6861bc","updated":"2015-08-11 03:38:00.000000000","message":"even the this will include a long import chain in oslo.reports, but it seems on series effects when this is loaded. the other projects like Nova used gmr like this. \n\nAre you concerning about the memory cost of this import, or the affect if this \u0027service\u0027 module would be imported in other place ?\n\nmaybe it is better moving this to following, but not a good code style ;) what do you think ?","commit_id":"0381f6c8bef70f4de862a9c882443f406a21b962"},{"author":{"_account_id":13637,"name":"Divya K Konoor","email":"dikonoor@in.ibm.com","username":"dikonoor"},"change_message_id":"cd7014b419188309ba0128b931671f8f8e47c03a","unresolved":false,"context_lines":[{"line_number":124,"context_line":"    # ceilometer-api running with mod_wsgi, the argv is [], we don\u0027t start"},{"line_number":125,"context_line":"    # guru."},{"line_number":126,"context_line":"    if argv:"},{"line_number":127,"context_line":"        gmr.TextGuruMeditation.setup_autorun(version)"},{"line_number":128,"context_line":"    messaging.setup()"}],"source_content_type":"text/x-python","patch_set":5,"id":"3a50d1a3_ba3813ef","line":127,"updated":"2015-07-29 17:26:07.000000000","message":"The setup_autorun method has an option to take in the log_dir into which the GMR file should be generated. IF this is not passed in, then GMR is generated into stderr. Generating into standard error means GMR gets lost along with the other stderr, Each service should give an option to specify the log directory as part of their respective conf file..for eg. support should be added for a new section something named [oslo_reports]\nlog_dir\u003d\n\nThe other services where GMR is currently implemented don\u0027t have this option and I think it was just missed out","commit_id":"0381f6c8bef70f4de862a9c882443f406a21b962"},{"author":{"_account_id":8290,"name":"liusheng","email":"liusheng2048@gmail.com","username":"LiuSheng"},"change_message_id":"d009ab678d1666f39d260c899cfb66ef1e1548d1","unresolved":false,"context_lines":[{"line_number":124,"context_line":"    # ceilometer-api running with mod_wsgi, the argv is [], we don\u0027t start"},{"line_number":125,"context_line":"    # guru."},{"line_number":126,"context_line":"    if argv:"},{"line_number":127,"context_line":"        gmr.TextGuruMeditation.setup_autorun(version)"},{"line_number":128,"context_line":"    messaging.setup()"}],"source_content_type":"text/x-python","patch_set":5,"id":"1a4dcd0f_7a6e99ca","line":127,"in_reply_to":"3a50d1a3_ba3813ef","updated":"2015-08-11 03:38:00.000000000","message":"Hi, Divya K Konoor, sorry for reply lately. maybe the stderr can    also be redirected to the service log file, it seems the jenkins does like this, see[1]. I agree with adding an dedicated log option for GMR, but maybe it is better in oslo.reports. like other oslo.*, we generate the config options by tox, see [2].\n\n[1] http://logs.openstack.org/79/211379/1/check/gate-tempest-dsvm-full/e511f38/logs/screen-n-cpu.txt.gz#_2015-08-11_01_57_41_737\n\n[2] https://github.com/openstack/ceilometer/blob/master/tox.ini#L81-L91","commit_id":"0381f6c8bef70f4de862a9c882443f406a21b962"},{"author":{"_account_id":6537,"name":"gordon chung","email":"gord@live.ca","username":"chungg"},"change_message_id":"6e9eb86561e9c42739d082f646ab3afb25d7a4cf","unresolved":false,"context_lines":[{"line_number":21,"context_line":"from oslo_config import cfg"},{"line_number":22,"context_line":"import oslo_i18n"},{"line_number":23,"context_line":"from oslo_log import log"},{"line_number":24,"context_line":"from oslo_reports import guru_meditation_report as gmr"},{"line_number":25,"context_line":""},{"line_number":26,"context_line":"from ceilometer.i18n import _"},{"line_number":27,"context_line":"from ceilometer import messaging"}],"source_content_type":"text/x-python","patch_set":6,"id":"1a4dcd0f_e8df1ad4","line":24,"updated":"2015-08-11 16:22:56.000000000","message":"the main idea was to not load something that would never be used. if it\u0027s not on here, it\u0027ll never be on.\n\nthe same was done for migrations. we only load it in upgrade method rather than in global connection.","commit_id":"ee6a2ad5661f9a3ff6a5bbb98272b64addc912f1"}]}