)]}'
{"cinder/service.py":[{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"9c5cf4e7af6465118ac127001b42f002ae16828b","unresolved":false,"context_lines":[{"line_number":127,"context_line":"        if not rpc.initialized():"},{"line_number":128,"context_line":"            rpc.init(CONF)"},{"line_number":129,"context_line":""},{"line_number":130,"context_line":"        self.cluster \u003d cluster"},{"line_number":131,"context_line":"        self.host \u003d host"},{"line_number":132,"context_line":"        self.binary \u003d binary"},{"line_number":133,"context_line":"        self.topic \u003d topic"}],"source_content_type":"text/x-python","patch_set":8,"id":"3aaa91ec_4848195c","line":130,"updated":"2016-06-28 07:17:16.000000000","message":"-1,\n\nI wondering whether it is easier to just set:\n\nself.cluster \u003d cluster if not self.is_upgrading else None\n\nSo that it doesn\u0027t need to do check in service and manager. \n\nEspecially in manager, you transfer cluster_name to it. And then when creating volume, it needs to decide whether setting cluster_name or not: if not upgrading set cluster_name of a created volume, or else not set cluster_name.\n\nI noticed your later patch: \nhttps://review.openstack.org/#/c/318573/9/cinder/volume/rpcapi.py","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"},{"author":{"_account_id":9535,"name":"Gorka Eguileor","email":"geguileo@redhat.com","username":"Gorka"},"change_message_id":"132cb996276ed858350539f9db650a468175b5cf","unresolved":false,"context_lines":[{"line_number":127,"context_line":"        if not rpc.initialized():"},{"line_number":128,"context_line":"            rpc.init(CONF)"},{"line_number":129,"context_line":""},{"line_number":130,"context_line":"        self.cluster \u003d cluster"},{"line_number":131,"context_line":"        self.host \u003d host"},{"line_number":132,"context_line":"        self.binary \u003d binary"},{"line_number":133,"context_line":"        self.topic \u003d topic"}],"source_content_type":"text/x-python","patch_set":8,"id":"3aaa91ec_a28711d6","line":130,"in_reply_to":"3aaa91ec_4848195c","updated":"2016-06-28 10:37:46.000000000","message":"If we do that then we\u0027ll have problems when we upgrade hosts that are in clusters from version N to O.  See my reply in 2 comments below.\n\nCreation is done in the scheduler, and we\u0027ll deal with when we can add the cluster_name field for new volumes in there.","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"b528ad45e0b9b7936a8c25d75e929a365af68e19","unresolved":false,"context_lines":[{"line_number":127,"context_line":"        if not rpc.initialized():"},{"line_number":128,"context_line":"            rpc.init(CONF)"},{"line_number":129,"context_line":""},{"line_number":130,"context_line":"        self.cluster \u003d cluster"},{"line_number":131,"context_line":"        self.host \u003d host"},{"line_number":132,"context_line":"        self.binary \u003d binary"},{"line_number":133,"context_line":"        self.topic \u003d topic"}],"source_content_type":"text/x-python","patch_set":8,"id":"3aaa91ec_ce745837","line":130,"in_reply_to":"3aaa91ec_a28711d6","updated":"2016-07-01 08:03:43.000000000","message":"In fact, we can check whether whether cinder is upgrading from M release to N. How about we just do the check?","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"},{"author":{"_account_id":9535,"name":"Gorka Eguileor","email":"geguileo@redhat.com","username":"Gorka"},"change_message_id":"99935820a1800a86e6eeac760afac310a9debf07","unresolved":false,"context_lines":[{"line_number":127,"context_line":"        if not rpc.initialized():"},{"line_number":128,"context_line":"            rpc.init(CONF)"},{"line_number":129,"context_line":""},{"line_number":130,"context_line":"        self.cluster \u003d cluster"},{"line_number":131,"context_line":"        self.host \u003d host"},{"line_number":132,"context_line":"        self.binary \u003d binary"},{"line_number":133,"context_line":"        self.topic \u003d topic"}],"source_content_type":"text/x-python","patch_set":8,"id":"1aa78d24_72c4b75e","line":130,"in_reply_to":"3aaa91ec_ce745837","updated":"2016-07-05 12:20:50.000000000","message":"That\u0027s probably the best thing to do.","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"9c5cf4e7af6465118ac127001b42f002ae16828b","unresolved":false,"context_lines":[{"line_number":163,"context_line":"            # configuration has a cluster value.  We don\u0027t want to do anything"},{"line_number":164,"context_line":"            # automatic if the cluster is changed, in those cases we\u0027ll want"},{"line_number":165,"context_line":"            # to use cinder manage command and to it manually."},{"line_number":166,"context_line":"            self.added_to_cluster \u003d (not service_ref.cluster_name and cluster"},{"line_number":167,"context_line":"                                     and not self.is_upgrading)"},{"line_number":168,"context_line":"            if self.added_to_cluster:"},{"line_number":169,"context_line":"                # We pass copy service\u0027s disable status in the cluster if we"}],"source_content_type":"text/x-python","patch_set":8,"id":"3aaa91ec_6549bed0","line":166,"updated":"2016-06-28 07:17:16.000000000","message":"-1,\nI think we also need to check whether cluster_name changed.","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"b528ad45e0b9b7936a8c25d75e929a365af68e19","unresolved":false,"context_lines":[{"line_number":163,"context_line":"            # configuration has a cluster value.  We don\u0027t want to do anything"},{"line_number":164,"context_line":"            # automatic if the cluster is changed, in those cases we\u0027ll want"},{"line_number":165,"context_line":"            # to use cinder manage command and to it manually."},{"line_number":166,"context_line":"            self.added_to_cluster \u003d (not service_ref.cluster_name and cluster"},{"line_number":167,"context_line":"                                     and not self.is_upgrading)"},{"line_number":168,"context_line":"            if self.added_to_cluster:"},{"line_number":169,"context_line":"                # We pass copy service\u0027s disable status in the cluster if we"}],"source_content_type":"text/x-python","patch_set":8,"id":"3aaa91ec_53b48f80","line":166,"in_reply_to":"3aaa91ec_422f4da1","updated":"2016-07-01 08:03:43.000000000","message":"Sorry, sometimes I was lazy and skipped the comments.\n\nShould we at least give a warning here if cluster value is changed?","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"},{"author":{"_account_id":9535,"name":"Gorka Eguileor","email":"geguileo@redhat.com","username":"Gorka"},"change_message_id":"99935820a1800a86e6eeac760afac310a9debf07","unresolved":false,"context_lines":[{"line_number":163,"context_line":"            # configuration has a cluster value.  We don\u0027t want to do anything"},{"line_number":164,"context_line":"            # automatic if the cluster is changed, in those cases we\u0027ll want"},{"line_number":165,"context_line":"            # to use cinder manage command and to it manually."},{"line_number":166,"context_line":"            self.added_to_cluster \u003d (not service_ref.cluster_name and cluster"},{"line_number":167,"context_line":"                                     and not self.is_upgrading)"},{"line_number":168,"context_line":"            if self.added_to_cluster:"},{"line_number":169,"context_line":"                # We pass copy service\u0027s disable status in the cluster if we"}],"source_content_type":"text/x-python","patch_set":8,"id":"1aa78d24_b750d10b","line":166,"in_reply_to":"3aaa91ec_53b48f80","updated":"2016-07-05 12:20:50.000000000","message":"Yes, that\u0027s a good idea","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"},{"author":{"_account_id":9535,"name":"Gorka Eguileor","email":"geguileo@redhat.com","username":"Gorka"},"change_message_id":"132cb996276ed858350539f9db650a468175b5cf","unresolved":false,"context_lines":[{"line_number":163,"context_line":"            # configuration has a cluster value.  We don\u0027t want to do anything"},{"line_number":164,"context_line":"            # automatic if the cluster is changed, in those cases we\u0027ll want"},{"line_number":165,"context_line":"            # to use cinder manage command and to it manually."},{"line_number":166,"context_line":"            self.added_to_cluster \u003d (not service_ref.cluster_name and cluster"},{"line_number":167,"context_line":"                                     and not self.is_upgrading)"},{"line_number":168,"context_line":"            if self.added_to_cluster:"},{"line_number":169,"context_line":"                # We pass copy service\u0027s disable status in the cluster if we"}],"source_content_type":"text/x-python","patch_set":8,"id":"3aaa91ec_422f4da1","line":166,"in_reply_to":"3aaa91ec_6549bed0","updated":"2016-06-28 10:37:46.000000000","message":"I gave quite some thought to that, but as I mention in L163-L165 we don\u0027t want to do that, in those cases we want the user to use cinder manage command and to it manually, as we don\u0027t know if they have also changed other things like going from multibackend to single backend, etc., so we can\u0027t be sure how we want to do the cluster renaming.","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"9c5cf4e7af6465118ac127001b42f002ae16828b","unresolved":false,"context_lines":[{"line_number":229,"context_line":"        self.rpcserver \u003d rpc.get_server(target, endpoints, serializer)"},{"line_number":230,"context_line":"        self.rpcserver.start()"},{"line_number":231,"context_line":""},{"line_number":232,"context_line":"        if self.cluster:"},{"line_number":233,"context_line":"            LOG.info(_LI(\u0027Starting %(topic)s cluster %(cluster)s (version \u0027"},{"line_number":234,"context_line":"                         \u0027%(version)s)\u0027),"},{"line_number":235,"context_line":"                     {\u0027topic\u0027: self.topic, \u0027version\u0027: version_string,"}],"source_content_type":"text/x-python","patch_set":8,"id":"3aaa91ec_459002ed","line":232,"updated":"2016-06-28 07:17:16.000000000","message":"-1, I think it should be:\nif not self.is_upgrading and self.cluster:\n\nas no need to start cluster_rpcserver.","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"},{"author":{"_account_id":9535,"name":"Gorka Eguileor","email":"geguileo@redhat.com","username":"Gorka"},"change_message_id":"132cb996276ed858350539f9db650a468175b5cf","unresolved":false,"context_lines":[{"line_number":229,"context_line":"        self.rpcserver \u003d rpc.get_server(target, endpoints, serializer)"},{"line_number":230,"context_line":"        self.rpcserver.start()"},{"line_number":231,"context_line":""},{"line_number":232,"context_line":"        if self.cluster:"},{"line_number":233,"context_line":"            LOG.info(_LI(\u0027Starting %(topic)s cluster %(cluster)s (version \u0027"},{"line_number":234,"context_line":"                         \u0027%(version)s)\u0027),"},{"line_number":235,"context_line":"                     {\u0027topic\u0027: self.topic, \u0027version\u0027: version_string,"}],"source_content_type":"text/x-python","patch_set":8,"id":"3aaa91ec_6299a91f","line":232,"in_reply_to":"3aaa91ec_459002ed","updated":"2016-06-28 10:37:46.000000000","message":"We need to start it because we could be running N with nodes in clusters and then upgrade to O and if we don\u0027t add the cluster then we\u0027ll break during the upgrade from N to O.","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"},{"author":{"_account_id":15961,"name":"lisali","email":"xiaoyan.li@intel.com","username":"lisali"},"change_message_id":"b528ad45e0b9b7936a8c25d75e929a365af68e19","unresolved":false,"context_lines":[{"line_number":229,"context_line":"        self.rpcserver \u003d rpc.get_server(target, endpoints, serializer)"},{"line_number":230,"context_line":"        self.rpcserver.start()"},{"line_number":231,"context_line":""},{"line_number":232,"context_line":"        if self.cluster:"},{"line_number":233,"context_line":"            LOG.info(_LI(\u0027Starting %(topic)s cluster %(cluster)s (version \u0027"},{"line_number":234,"context_line":"                         \u0027%(version)s)\u0027),"},{"line_number":235,"context_line":"                     {\u0027topic\u0027: self.topic, \u0027version\u0027: version_string,"}],"source_content_type":"text/x-python","patch_set":8,"id":"3aaa91ec_ce9d7867","line":232,"in_reply_to":"3aaa91ec_6299a91f","updated":"2016-07-01 08:03:43.000000000","message":"You are right. When N with nodes upgrade to O, no need to check is_upgrading.\n\nBut it is ok to check when M with nodes upgrade to N.","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"},{"author":{"_account_id":9535,"name":"Gorka Eguileor","email":"geguileo@redhat.com","username":"Gorka"},"change_message_id":"99935820a1800a86e6eeac760afac310a9debf07","unresolved":false,"context_lines":[{"line_number":229,"context_line":"        self.rpcserver \u003d rpc.get_server(target, endpoints, serializer)"},{"line_number":230,"context_line":"        self.rpcserver.start()"},{"line_number":231,"context_line":""},{"line_number":232,"context_line":"        if self.cluster:"},{"line_number":233,"context_line":"            LOG.info(_LI(\u0027Starting %(topic)s cluster %(cluster)s (version \u0027"},{"line_number":234,"context_line":"                         \u0027%(version)s)\u0027),"},{"line_number":235,"context_line":"                     {\u0027topic\u0027: self.topic, \u0027version\u0027: version_string,"}],"source_content_type":"text/x-python","patch_set":8,"id":"1aa78d24_1710c500","line":232,"in_reply_to":"3aaa91ec_ce9d7867","updated":"2016-07-05 12:20:50.000000000","message":"I don\u0027t think it\u0027s worth it, as the only gain is that we won\u0027t be running the additional RPC server during the upgrade.  But I\u0027ll add it since it doesn\u0027t cost anything.","commit_id":"ce7bb7269cb7bbb90327bd82c5b57d8ab25459fe"}],"releasenotes/notes/cluster_job_distribution-f916dd2e4cce6c1b.yaml":[{"author":{"_account_id":7173,"name":"Scott DAngelo","email":"scott.dangelo@gmail.com","username":"scottda"},"change_message_id":"9f719af1a0d9c7fe5b7600313c9f39d0167a2cc2","unresolved":false,"context_lines":[{"line_number":7,"context_line":"    production."},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"features:"},{"line_number":10,"context_line":"  - HA A-A: Add cluster configuration option to allow grouping hosts that"},{"line_number":11,"context_line":"    share the same backend configurations and should work in Active-Active"},{"line_number":12,"context_line":"    fashion."}],"source_content_type":"text/x-yaml","patch_set":2,"id":"7aa08908_61b4f5f1","line":10,"updated":"2016-06-10 12:50:29.000000000","message":"I think the colon \":\" is what causes the release notes problem. Not positive on this.","commit_id":"708960ff72d18a66c5881d0cc6d983792113962f"},{"author":{"_account_id":9535,"name":"Gorka Eguileor","email":"geguileo@redhat.com","username":"Gorka"},"change_message_id":"e364dbaee1f080e5a539f610065f5fa46ac00a05","unresolved":false,"context_lines":[{"line_number":7,"context_line":"    production."},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"features:"},{"line_number":10,"context_line":"  - HA A-A: Add cluster configuration option to allow grouping hosts that"},{"line_number":11,"context_line":"    share the same backend configurations and should work in Active-Active"},{"line_number":12,"context_line":"    fashion."}],"source_content_type":"text/x-yaml","patch_set":2,"id":"7aa08908_84773d3a","line":10,"in_reply_to":"7aa08908_61b4f5f1","updated":"2016-06-12 11:39:41.000000000","message":"Yep, it\u0027s the colon, it\u0027s being interpreted as a key separator.  :-(","commit_id":"708960ff72d18a66c5881d0cc6d983792113962f"}]}