)]}'
{"ironic/api/controllers/v1/port.py":[{"author":{"_account_id":14525,"name":"Vasyl Saienko","email":"vsaienko@mirantis.com","username":"vsaienko"},"change_message_id":"776d0ac3355916e3a4f86f65f6cefb77629ba737","unresolved":false,"context_lines":[{"line_number":547,"context_line":"        vif \u003d extra.get(\u0027vif_port_id\u0027) if extra else None"},{"line_number":548,"context_line":"        if vif:"},{"line_number":549,"context_line":"            common_utils.warn_about_deprecated_extra_vif_port_id()"},{"line_number":550,"context_line":"            # NOTE(rloo): this value should really be in .internal_info[..]"},{"line_number":551,"context_line":"            #             which is what would happen if they had used the"},{"line_number":552,"context_line":"            #             POST /v1/nodes/\u003cnode\u003e/vifs API."},{"line_number":553,"context_line":"            internal_info \u003d {\u0027tenant_vif_port_id\u0027: vif}"},{"line_number":554,"context_line":"            pdict[\u0027internal_info\u0027] \u003d internal_info"},{"line_number":555,"context_line":""},{"line_number":556,"context_line":"        if (pdict.get(\u0027portgroup_uuid\u0027) and"},{"line_number":557,"context_line":"                (pdict.get(\u0027pxe_enabled\u0027) or vif)):"}],"source_content_type":"text/x-python","patch_set":3,"id":"3f4b6375_0aa8b48a","line":554,"range":{"start_line":550,"start_character":12,"end_line":554,"end_character":50},"updated":"2017-10-24 07:05:53.000000000","message":"We show deprecation here, and I doubt that we need to copy value to internal_info. We should force users to update theirs scripts to use vif attach/detach. Copying tenant_vif_port_id to internal_info when run dbsync is seems enough","commit_id":"43048be11f76f0b7c3e1a5eadb23c78a160d4012"},{"author":{"_account_id":6618,"name":"Ruby Loo","email":"opensrloo@gmail.com","username":"rloo"},"change_message_id":"5f9c411efa3a0fb86b0edd5a53740418cc45623c","unresolved":false,"context_lines":[{"line_number":547,"context_line":"        vif \u003d extra.get(\u0027vif_port_id\u0027) if extra else None"},{"line_number":548,"context_line":"        if vif:"},{"line_number":549,"context_line":"            common_utils.warn_about_deprecated_extra_vif_port_id()"},{"line_number":550,"context_line":"            # NOTE(rloo): this value should really be in .internal_info[..]"},{"line_number":551,"context_line":"            #             which is what would happen if they had used the"},{"line_number":552,"context_line":"            #             POST /v1/nodes/\u003cnode\u003e/vifs API."},{"line_number":553,"context_line":"            internal_info \u003d {\u0027tenant_vif_port_id\u0027: vif}"},{"line_number":554,"context_line":"            pdict[\u0027internal_info\u0027] \u003d internal_info"},{"line_number":555,"context_line":""},{"line_number":556,"context_line":"        if (pdict.get(\u0027portgroup_uuid\u0027) and"},{"line_number":557,"context_line":"                (pdict.get(\u0027pxe_enabled\u0027) or vif)):"}],"source_content_type":"text/x-python","patch_set":3,"id":"3f4b6375_6043c0b0","line":554,"range":{"start_line":550,"start_character":12,"end_line":554,"end_character":50},"in_reply_to":"3f4b6375_0aa8b48a","updated":"2017-10-24 12:40:35.000000000","message":"the problem is that we don\u0027t know when they will run \u0027ironic-dbsync online_data_migrations\u0027.\n\nThey upgrade to Queens release, which has this code. Before they upgrade to Rocky, they will run \u0027ironic-dbsync online...\u0027. After they run the dbsync, they might call this API. If we don\u0027t copy to internal_info, after upgrading to Rocky, the Rocky code will ignore .extra[] and only look at .internal and the info won\u0027t be there.\n\nThe chance of this occurring (calling this API to attach vif via .extra at end of Queens but actually using vif info in Rocky) may be low but even if there is one instance of this, we would have broken it.\n\nOr am I missing something?","commit_id":"43048be11f76f0b7c3e1a5eadb23c78a160d4012"},{"author":{"_account_id":14525,"name":"Vasyl Saienko","email":"vsaienko@mirantis.com","username":"vsaienko"},"change_message_id":"59164ecba86a506f859192d30f846e9f569680ba","unresolved":false,"context_lines":[{"line_number":547,"context_line":"        vif \u003d extra.get(\u0027vif_port_id\u0027) if extra else None"},{"line_number":548,"context_line":"        if vif:"},{"line_number":549,"context_line":"            common_utils.warn_about_deprecated_extra_vif_port_id()"},{"line_number":550,"context_line":"            # NOTE(rloo): this value should really be in .internal_info[..]"},{"line_number":551,"context_line":"            #             which is what would happen if they had used the"},{"line_number":552,"context_line":"            #             POST /v1/nodes/\u003cnode\u003e/vifs API."},{"line_number":553,"context_line":"            internal_info \u003d {\u0027tenant_vif_port_id\u0027: vif}"},{"line_number":554,"context_line":"            pdict[\u0027internal_info\u0027] \u003d internal_info"},{"line_number":555,"context_line":""},{"line_number":556,"context_line":"        if (pdict.get(\u0027portgroup_uuid\u0027) and"},{"line_number":557,"context_line":"                (pdict.get(\u0027pxe_enabled\u0027) or vif)):"}],"source_content_type":"text/x-python","patch_set":3,"id":"3f4b6375_f09642f0","line":554,"range":{"start_line":550,"start_character":12,"end_line":554,"end_character":50},"in_reply_to":"3f4b6375_3232b360","updated":"2017-10-25 08:36:24.000000000","message":"sounds like a plan, I missed that user may still set tenant_vif_port_id in extra after run of \u0027ironic-dbsync online_data_migrations\u0027","commit_id":"43048be11f76f0b7c3e1a5eadb23c78a160d4012"},{"author":{"_account_id":6618,"name":"Ruby Loo","email":"opensrloo@gmail.com","username":"rloo"},"change_message_id":"a8036da5521ea7cb8e077529958bde445a9a01e6","unresolved":false,"context_lines":[{"line_number":547,"context_line":"        vif \u003d extra.get(\u0027vif_port_id\u0027) if extra else None"},{"line_number":548,"context_line":"        if vif:"},{"line_number":549,"context_line":"            common_utils.warn_about_deprecated_extra_vif_port_id()"},{"line_number":550,"context_line":"            # NOTE(rloo): this value should really be in .internal_info[..]"},{"line_number":551,"context_line":"            #             which is what would happen if they had used the"},{"line_number":552,"context_line":"            #             POST /v1/nodes/\u003cnode\u003e/vifs API."},{"line_number":553,"context_line":"            internal_info \u003d {\u0027tenant_vif_port_id\u0027: vif}"},{"line_number":554,"context_line":"            pdict[\u0027internal_info\u0027] \u003d internal_info"},{"line_number":555,"context_line":""},{"line_number":556,"context_line":"        if (pdict.get(\u0027portgroup_uuid\u0027) and"},{"line_number":557,"context_line":"                (pdict.get(\u0027pxe_enabled\u0027) or vif)):"}],"source_content_type":"text/x-python","patch_set":3,"id":"3f4b6375_3232b360","line":554,"range":{"start_line":550,"start_character":12,"end_line":554,"end_character":50},"in_reply_to":"3f4b6375_412fa38d","updated":"2017-10-24 17:05:15.000000000","message":"I\u0027m not sure I understand. This is my plan (I hope it is clear):\n\nIn Queens release: 1. copy extra[vif_port_id] to internal_info if the user tries to set it in extra; 2. ironic-dbsync online_data_migrations will also copy extra.vif_port_id.\n\nOperator upgrades to Queens release.\n\nIn Queens:\n- If they try to set extra[vif_port_id], it will get copied to internal_info via 1.\n- Before they upgrade to Rocky, they must run ironic-dbsync online_data_migrations (2.)\n\nOperator upgrades to Rocky release. At this point, we know that all extra[vif_port_id] got copied to internal_info because of 1 \u0026 2. (If I didn\u0027t do 1, the operator could have set extra[] after they ran the ironic-dbsync command, which would mean that there was an .extra[] that did NOT get copied to internal_info.)\n\nRocky release ignores extra[vif_port_id]. User can set/modify it via API, but we just leave the value there for user purposes. Code will only look at internal_info. (In Rocky+, we cannot drop extra[vif_port_id] because extra was meant for operators to use, so we don\u0027t know if they are setting it due to wanting to do the vif attach, or setting it cuz they want to set it for their own use...)\n\n[Which reminds me that I have to update the bug description with this plan...]","commit_id":"43048be11f76f0b7c3e1a5eadb23c78a160d4012"},{"author":{"_account_id":14525,"name":"Vasyl Saienko","email":"vsaienko@mirantis.com","username":"vsaienko"},"change_message_id":"4929b6cd1e76bcdecb8e36a98f41a6bcf327022d","unresolved":false,"context_lines":[{"line_number":547,"context_line":"        vif \u003d extra.get(\u0027vif_port_id\u0027) if extra else None"},{"line_number":548,"context_line":"        if vif:"},{"line_number":549,"context_line":"            common_utils.warn_about_deprecated_extra_vif_port_id()"},{"line_number":550,"context_line":"            # NOTE(rloo): this value should really be in .internal_info[..]"},{"line_number":551,"context_line":"            #             which is what would happen if they had used the"},{"line_number":552,"context_line":"            #             POST /v1/nodes/\u003cnode\u003e/vifs API."},{"line_number":553,"context_line":"            internal_info \u003d {\u0027tenant_vif_port_id\u0027: vif}"},{"line_number":554,"context_line":"            pdict[\u0027internal_info\u0027] \u003d internal_info"},{"line_number":555,"context_line":""},{"line_number":556,"context_line":"        if (pdict.get(\u0027portgroup_uuid\u0027) and"},{"line_number":557,"context_line":"                (pdict.get(\u0027pxe_enabled\u0027) or vif)):"}],"source_content_type":"text/x-python","patch_set":3,"id":"3f4b6375_412fa38d","line":554,"range":{"start_line":550,"start_character":12,"end_line":554,"end_character":50},"in_reply_to":"3f4b6375_6043c0b0","updated":"2017-10-24 15:10:26.000000000","message":"This will not update nodes that were deployed in previous release, we still need to be sure that online_data_migration was executed before dropping it.\nSo my suggestion is to:\n\n1. Copy vif_port_id to internal_info with online_data_migration\n2. Drop extra[\u0027tenant_vif_port_id\u0027] after one more release (when we sure that users launch online_data_migration in Rocky","commit_id":"43048be11f76f0b7c3e1a5eadb23c78a160d4012"},{"author":{"_account_id":14525,"name":"Vasyl Saienko","email":"vsaienko@mirantis.com","username":"vsaienko"},"change_message_id":"e1cc77bcbe09472e20de0c7ca6cbf2d0c17cf992","unresolved":false,"context_lines":[{"line_number":657,"context_line":"            orig_info \u003d rpc_port.internal_info.get(\u0027tenant_vif_port_id\u0027)"},{"line_number":658,"context_line":"            if (not orig_info or"},{"line_number":659,"context_line":"                orig_info \u003d\u003d rpc_port.extra.get(\u0027vif_port_id\u0027)):"},{"line_number":660,"context_line":"                port.internal_info \u003d {\u0027tenant_vif_port_id\u0027: vif}"},{"line_number":661,"context_line":""},{"line_number":662,"context_line":"        if api_utils.is_path_removed(patch, \u0027/portgroup_uuid\u0027):"},{"line_number":663,"context_line":"            rpc_port.portgroup_id \u003d None"}],"source_content_type":"text/x-python","patch_set":4,"id":"3f4b6375_aedde6e0","line":660,"updated":"2017-10-26 12:08:05.000000000","message":"we lost here all info that were in internal_info before this update, we need to update only tenant_vif_port_id key.","commit_id":"3510aae0007855e501a5b738c8d7398e3d4f0041"},{"author":{"_account_id":6618,"name":"Ruby Loo","email":"opensrloo@gmail.com","username":"rloo"},"change_message_id":"1f81e24abfd29844d1ecf4aa62d74fbea1c749f2","unresolved":false,"context_lines":[{"line_number":657,"context_line":"            orig_info \u003d rpc_port.internal_info.get(\u0027tenant_vif_port_id\u0027)"},{"line_number":658,"context_line":"            if (not orig_info or"},{"line_number":659,"context_line":"                orig_info \u003d\u003d rpc_port.extra.get(\u0027vif_port_id\u0027)):"},{"line_number":660,"context_line":"                port.internal_info \u003d {\u0027tenant_vif_port_id\u0027: vif}"},{"line_number":661,"context_line":""},{"line_number":662,"context_line":"        if api_utils.is_path_removed(patch, \u0027/portgroup_uuid\u0027):"},{"line_number":663,"context_line":"            rpc_port.portgroup_id \u003d None"}],"source_content_type":"text/x-python","patch_set":4,"id":"3f4b6375_99a8e0db","line":660,"in_reply_to":"3f4b6375_aedde6e0","updated":"2017-10-30 18:18:32.000000000","message":"Ack! Good catch!","commit_id":"3510aae0007855e501a5b738c8d7398e3d4f0041"},{"author":{"_account_id":6618,"name":"Ruby Loo","email":"opensrloo@gmail.com","username":"rloo"},"change_message_id":"2f0a9c703dfb42ce33e27e95ff199434ec5b1bc8","unresolved":false,"context_lines":[{"line_number":649,"context_line":"            # if specified more than once, use the last value"},{"line_number":650,"context_line":"            vif \u003d vif_list[-1]"},{"line_number":651,"context_line":"            # TODO(rloo): in Rocky cycle: if API version \u003e\u003d 1.28, remove this"},{"line_number":652,"context_line":"            #             warning and don\u0027t copy to internal_info (bug 1722850)"},{"line_number":653,"context_line":"            if api_utils.allow_vifs_subcontroller():"},{"line_number":654,"context_line":"                common_utils.warn_about_deprecated_extra_vif_port_id()"},{"line_number":655,"context_line":"            # NOTE(rloo): if the user isn\u0027t also using the REST API"}],"source_content_type":"text/x-python","patch_set":6,"id":"1f9dbf25_84e3b133","line":652,"updated":"2018-03-02 11:31:01.000000000","message":"and if \u003c 1.28, remove warning but keep code from l655.","commit_id":"97d76fb6fd49f4634f3ff2b65156bb75256cbbba"},{"author":{"_account_id":6637,"name":"Sam Betts","email":"sam@code-smash.net","username":"sambetts"},"change_message_id":"c5f549117069f3ffb9b3c98e62ceb5e3d663097f","unresolved":false,"context_lines":[{"line_number":658,"context_line":"            orig_info \u003d rpc_port.internal_info.get(\u0027tenant_vif_port_id\u0027)"},{"line_number":659,"context_line":"            if (not orig_info or"},{"line_number":660,"context_line":"                orig_info \u003d\u003d rpc_port.extra.get(\u0027vif_port_id\u0027)):"},{"line_number":661,"context_line":"                port.internal_info[\u0027tenant_vif_port_id\u0027] \u003d vif"},{"line_number":662,"context_line":""},{"line_number":663,"context_line":"        if api_utils.is_path_removed(patch, \u0027/portgroup_uuid\u0027):"},{"line_number":664,"context_line":"            rpc_port.portgroup_id \u003d None"}],"source_content_type":"text/x-python","patch_set":6,"id":"ff82abbf_5fa844da","line":661,"updated":"2017-11-30 16:03:16.000000000","message":"Don\u0027t we also need to support the reverse?? If the vif is removed from port.extra we need to remove it from the internal info field.\n\nThis code also makes the assumption that the network interface being loaded on the nodes is one of the current in-tree implementations that stores its vif value in internal_info[\u0027tenant_vif_port_id\u0027] that might not be the case, so I wondered if instead of setting the value manually in internal_info like this we should call out to the network_interface to attach the VIF like it would be through the vifs API except with a specific port/portgroup specified so it doesn\u0027t attach to any other port???","commit_id":"97d76fb6fd49f4634f3ff2b65156bb75256cbbba"},{"author":{"_account_id":12356,"name":"Vladyslav Drok","email":"vdrok@mirantis.com","username":"vdrok"},"change_message_id":"39801173154504ba5976bd17e18a3da70574ff25","unresolved":false,"context_lines":[{"line_number":658,"context_line":"            orig_info \u003d rpc_port.internal_info.get(\u0027tenant_vif_port_id\u0027)"},{"line_number":659,"context_line":"            if (not orig_info or"},{"line_number":660,"context_line":"                orig_info \u003d\u003d rpc_port.extra.get(\u0027vif_port_id\u0027)):"},{"line_number":661,"context_line":"                port.internal_info[\u0027tenant_vif_port_id\u0027] \u003d vif"},{"line_number":662,"context_line":""},{"line_number":663,"context_line":"        if api_utils.is_path_removed(patch, \u0027/portgroup_uuid\u0027):"},{"line_number":664,"context_line":"            rpc_port.portgroup_id \u003d None"}],"source_content_type":"text/x-python","patch_set":6,"id":"1f9dbf25_dcd520bb","line":661,"in_reply_to":"1f9dbf25_5cf03091","updated":"2018-03-01 12:21:23.000000000","message":"release note should be ok, we had warnings for a while already. otoh we are basically forcing people to use api version where attach/detach interfaces was introduced, as there will be no way to detach vif from internal_info apart from calling vif detach endpoint","commit_id":"97d76fb6fd49f4634f3ff2b65156bb75256cbbba"},{"author":{"_account_id":6618,"name":"Ruby Loo","email":"opensrloo@gmail.com","username":"rloo"},"change_message_id":"2f0a9c703dfb42ce33e27e95ff199434ec5b1bc8","unresolved":false,"context_lines":[{"line_number":658,"context_line":"            orig_info \u003d rpc_port.internal_info.get(\u0027tenant_vif_port_id\u0027)"},{"line_number":659,"context_line":"            if (not orig_info or"},{"line_number":660,"context_line":"                orig_info \u003d\u003d rpc_port.extra.get(\u0027vif_port_id\u0027)):"},{"line_number":661,"context_line":"                port.internal_info[\u0027tenant_vif_port_id\u0027] \u003d vif"},{"line_number":662,"context_line":""},{"line_number":663,"context_line":"        if api_utils.is_path_removed(patch, \u0027/portgroup_uuid\u0027):"},{"line_number":664,"context_line":"            rpc_port.portgroup_id \u003d None"}],"source_content_type":"text/x-python","patch_set":6,"id":"1f9dbf25_041481f4","line":661,"in_reply_to":"1f9dbf25_dcd520bb","updated":"2018-03-02 11:31:01.000000000","message":"talked again with sam, vdrok, dmitry. we need to handle case where user removes the vif from .extra. Then in S*, we handle this also if api version \u003c 1.28","commit_id":"97d76fb6fd49f4634f3ff2b65156bb75256cbbba"},{"author":{"_account_id":6618,"name":"Ruby Loo","email":"opensrloo@gmail.com","username":"rloo"},"change_message_id":"bda14f6f2507906ee6b4bfbed4b4769eaa20fa8a","unresolved":false,"context_lines":[{"line_number":658,"context_line":"            orig_info \u003d rpc_port.internal_info.get(\u0027tenant_vif_port_id\u0027)"},{"line_number":659,"context_line":"            if (not orig_info or"},{"line_number":660,"context_line":"                orig_info \u003d\u003d rpc_port.extra.get(\u0027vif_port_id\u0027)):"},{"line_number":661,"context_line":"                port.internal_info[\u0027tenant_vif_port_id\u0027] \u003d vif"},{"line_number":662,"context_line":""},{"line_number":663,"context_line":"        if api_utils.is_path_removed(patch, \u0027/portgroup_uuid\u0027):"},{"line_number":664,"context_line":"            rpc_port.portgroup_id \u003d None"}],"source_content_type":"text/x-python","patch_set":6,"id":"1f9dbf25_5cf03091","line":661,"in_reply_to":"1f9dbf25_f904b37d","updated":"2018-03-01 12:12:17.000000000","message":"Sam and I looked at the code and Sam thinks that the code he thought was there, has been removed, so it is OK, we don\u0027t need to handle the reverse.\n\nand at this point, Sam is ok, we don\u0027t know what we can do about out-of-tree drivers.","commit_id":"97d76fb6fd49f4634f3ff2b65156bb75256cbbba"},{"author":{"_account_id":6618,"name":"Ruby Loo","email":"opensrloo@gmail.com","username":"rloo"},"change_message_id":"951e168bc3d53da7f3502559b4598dae682abbb9","unresolved":false,"context_lines":[{"line_number":658,"context_line":"            orig_info \u003d rpc_port.internal_info.get(\u0027tenant_vif_port_id\u0027)"},{"line_number":659,"context_line":"            if (not orig_info or"},{"line_number":660,"context_line":"                orig_info \u003d\u003d rpc_port.extra.get(\u0027vif_port_id\u0027)):"},{"line_number":661,"context_line":"                port.internal_info[\u0027tenant_vif_port_id\u0027] \u003d vif"},{"line_number":662,"context_line":""},{"line_number":663,"context_line":"        if api_utils.is_path_removed(patch, \u0027/portgroup_uuid\u0027):"},{"line_number":664,"context_line":"            rpc_port.portgroup_id \u003d None"}],"source_content_type":"text/x-python","patch_set":6,"id":"1f9dbf25_f904b37d","line":661,"in_reply_to":"ff82abbf_5fa844da","updated":"2018-03-01 08:27:47.000000000","message":"wrt supporting the reverse. I looked through the code, and I don\u0027t see where that is currently being done. Ie, if the vif is removed from port.extra, it doesn\u0027t look like that triggers any code to remove it from the internal info field. If we didn\u0027t do that already, I\u0027d prefer not to do it now; I\u0027m just trying to mimic as much as possible, the existing behaviour.\n\nwrt network interface assumption. If I don\u0027t do this, I have no idea how we\u0027d be able to guarantee wrt the online_data_migrations, that we\u0027ve migrated the info from .extra to .internal_info. I\u0027m not comfortable with the online_data_migrations calling the network_interface\u0027s vif_attache methods. I could put something in the release notes about this potentially breaking out-of-tree network interfaces.","commit_id":"97d76fb6fd49f4634f3ff2b65156bb75256cbbba"}],"ironic/api/controllers/v1/portgroup.py":[{"author":{"_account_id":14525,"name":"Vasyl Saienko","email":"vsaienko@mirantis.com","username":"vsaienko"},"change_message_id":"e1cc77bcbe09472e20de0c7ca6cbf2d0c17cf992","unresolved":false,"context_lines":[{"line_number":553,"context_line":"            orig_info \u003d rpc_portgroup.internal_info.get(\u0027tenant_vif_port_id\u0027)"},{"line_number":554,"context_line":"            if (not orig_info or"},{"line_number":555,"context_line":"                orig_info \u003d\u003d rpc_portgroup.extra.get(\u0027vif_port_id\u0027)):"},{"line_number":556,"context_line":"                portgroup.internal_info \u003d {\u0027tenant_vif_port_id\u0027: vif}"},{"line_number":557,"context_line":""},{"line_number":558,"context_line":"        # Update only the fields that have changed"},{"line_number":559,"context_line":"        for field in objects.Portgroup.fields:"}],"source_content_type":"text/x-python","patch_set":4,"id":"3f4b6375_0e699af7","line":556,"updated":"2017-10-26 12:08:05.000000000","message":"ditto about original values.","commit_id":"3510aae0007855e501a5b738c8d7398e3d4f0041"}],"ironic/common/utils.py":[{"author":{"_account_id":14525,"name":"Vasyl Saienko","email":"vsaienko@mirantis.com","username":"vsaienko"},"change_message_id":"59164ecba86a506f859192d30f846e9f569680ba","unresolved":false,"context_lines":[{"line_number":509,"context_line":"    global warn_deprecated_extra_vif_port_id"},{"line_number":510,"context_line":"    if not warn_deprecated_extra_vif_port_id:"},{"line_number":511,"context_line":"        warn_deprecated_extra_vif_port_id \u003d True"},{"line_number":512,"context_line":"        LOG.warning(\"Attaching VIF to a port/portgroup via \""},{"line_number":513,"context_line":"                    \"extra[\u0027vif_port_id\u0027] is deprecated and will not \""},{"line_number":514,"context_line":"                    \"be supported in the Rocky release. API endpoint \""},{"line_number":515,"context_line":"                    \"v1/nodes/\u003cnode\u003e/vifs should be used instead.\")"}],"source_content_type":"text/x-python","patch_set":3,"id":"3f4b6375_736b7cb6","line":515,"range":{"start_line":512,"start_character":21,"end_line":515,"end_character":65},"updated":"2017-10-25 08:36:24.000000000","message":"Now when I think about exact deprecation here one thing is not clear. We introduce vif endpoint in api microversion 1.28. Does it mean that we should drop support of \u0027extra\u0027 only starting from api microversion 1.28 and still support it until we drop support of all microversions prior 1.28?","commit_id":"43048be11f76f0b7c3e1a5eadb23c78a160d4012"},{"author":{"_account_id":14525,"name":"Vasyl Saienko","email":"vsaienko@mirantis.com","username":"vsaienko"},"change_message_id":"7096acc1d9269ca643421192202efe1ef5e6ab7a","unresolved":false,"context_lines":[{"line_number":509,"context_line":"    global warn_deprecated_extra_vif_port_id"},{"line_number":510,"context_line":"    if not warn_deprecated_extra_vif_port_id:"},{"line_number":511,"context_line":"        warn_deprecated_extra_vif_port_id \u003d True"},{"line_number":512,"context_line":"        LOG.warning(\"Attaching VIF to a port/portgroup via \""},{"line_number":513,"context_line":"                    \"extra[\u0027vif_port_id\u0027] is deprecated and will not \""},{"line_number":514,"context_line":"                    \"be supported in the Rocky release. API endpoint \""},{"line_number":515,"context_line":"                    \"v1/nodes/\u003cnode\u003e/vifs should be used instead.\")"}],"source_content_type":"text/x-python","patch_set":3,"id":"3f4b6375_1ba1f694","line":515,"range":{"start_line":512,"start_character":21,"end_line":515,"end_character":65},"in_reply_to":"3f4b6375_736b7cb6","updated":"2017-10-25 14:59:54.000000000","message":"vote -1 here until we get a decision here.","commit_id":"43048be11f76f0b7c3e1a5eadb23c78a160d4012"}],"ironic/objects/port.py":[{"author":{"_account_id":14525,"name":"Vasyl Saienko","email":"vsaienko@mirantis.com","username":"vsaienko"},"change_message_id":"c1c1d585b26ec7ded0b3dbe32afeacb23e90a0a8","unresolved":false,"context_lines":[{"line_number":36,"context_line":"              migrated (at the beginning of this call) and the number"},{"line_number":37,"context_line":"              of migrated objects."},{"line_number":38,"context_line":"    \"\"\""},{"line_number":39,"context_line":"    db_ports \u003d Port.dbapi.get_not_versions(\u0027Port\u0027, [\u00271.8\u0027])"},{"line_number":40,"context_line":"    total \u003d len(db_ports)"},{"line_number":41,"context_line":"    max_count \u003d max_count or total"},{"line_number":42,"context_line":"    done \u003d 0"}],"source_content_type":"text/x-python","patch_set":6,"id":"1f485f77_3d0d7f90","line":39,"updated":"2017-11-09 13:49:06.000000000","message":"just wondered if we introduce a new port version 1.9 this will return all previous version ie: 1.6, 1.7 and 1.9 right?\nbut we shouldn\u0027t run migration for ports with version higher than 1.8.","commit_id":"97d76fb6fd49f4634f3ff2b65156bb75256cbbba"},{"author":{"_account_id":6618,"name":"Ruby Loo","email":"opensrloo@gmail.com","username":"rloo"},"change_message_id":"3cf007b3083ee1a70718af734818abcda3a58c5a","unresolved":false,"context_lines":[{"line_number":36,"context_line":"              migrated (at the beginning of this call) and the number"},{"line_number":37,"context_line":"              of migrated objects."},{"line_number":38,"context_line":"    \"\"\""},{"line_number":39,"context_line":"    db_ports \u003d Port.dbapi.get_not_versions(\u0027Port\u0027, [\u00271.8\u0027])"},{"line_number":40,"context_line":"    total \u003d len(db_ports)"},{"line_number":41,"context_line":"    max_count \u003d max_count or total"},{"line_number":42,"context_line":"    done \u003d 0"}],"source_content_type":"text/x-python","patch_set":6,"id":"1f485f77_ec84dab3","line":39,"in_reply_to":"1f485f77_3d0d7f90","updated":"2017-11-09 14:58:55.000000000","message":"Yes. That list should be updated if we add more port versions in Queens, otherwise, as you mentioned, ports with v1.9 would also be \u0027migrated\u0027. (Should I add a comment about that?)\n\nI guess the \u0027good\u0027 thing is that if we forgot to add in new versions, it wouldn\u0027t hurt; the code would just do more work and save the same port info that was already in the DB.\n\nHmm. I put a comment in dbsync.py, to remove the migrations in Rocky, I was too lazy (or forgot) to do so here -- let me know if you think it might be useful.","commit_id":"97d76fb6fd49f4634f3ff2b65156bb75256cbbba"},{"author":{"_account_id":14525,"name":"Vasyl Saienko","email":"vsaienko@mirantis.com","username":"vsaienko"},"change_message_id":"3fb45041743c5db61a55862b416ca39c8eb4f03d","unresolved":false,"context_lines":[{"line_number":36,"context_line":"              migrated (at the beginning of this call) and the number"},{"line_number":37,"context_line":"              of migrated objects."},{"line_number":38,"context_line":"    \"\"\""},{"line_number":39,"context_line":"    db_ports \u003d Port.dbapi.get_not_versions(\u0027Port\u0027, [\u00271.8\u0027])"},{"line_number":40,"context_line":"    total \u003d len(db_ports)"},{"line_number":41,"context_line":"    max_count \u003d max_count or total"},{"line_number":42,"context_line":"    done \u003d 0"}],"source_content_type":"text/x-python","patch_set":6,"id":"1f485f77_f62f8e3c","line":39,"in_reply_to":"1f485f77_ec84dab3","updated":"2017-11-15 13:09:47.000000000","message":"I\u0027m ok with as is.","commit_id":"97d76fb6fd49f4634f3ff2b65156bb75256cbbba"}],"ironic/objects/portgroup.py":[{"author":{"_account_id":12356,"name":"Vladyslav Drok","email":"vdrok@mirantis.com","username":"vdrok"},"change_message_id":"890fc0e6dad6b534d57e0af35b34f2742a5b63e6","unresolved":false,"context_lines":[{"line_number":27,"context_line":"from ironic.objects import notification"},{"line_number":28,"context_line":""},{"line_number":29,"context_line":""},{"line_number":30,"context_line":"def migrate_vif_port_id(context, max_count):"},{"line_number":31,"context_line":"    \"\"\"Copy portgroup\u0027s VIF info from extra to internal_info."},{"line_number":32,"context_line":""},{"line_number":33,"context_line":"    :param context: The context."}],"source_content_type":"text/x-python","patch_set":4,"id":"3f4b6375_344e91c1","line":30,"updated":"2017-10-30 17:21:36.000000000","message":"this can be refactored in a single function for both port and port group, but not a big deal","commit_id":"3510aae0007855e501a5b738c8d7398e3d4f0041"},{"author":{"_account_id":6618,"name":"Ruby Loo","email":"opensrloo@gmail.com","username":"rloo"},"change_message_id":"1f81e24abfd29844d1ecf4aa62d74fbea1c749f2","unresolved":false,"context_lines":[{"line_number":27,"context_line":"from ironic.objects import notification"},{"line_number":28,"context_line":""},{"line_number":29,"context_line":""},{"line_number":30,"context_line":"def migrate_vif_port_id(context, max_count):"},{"line_number":31,"context_line":"    \"\"\"Copy portgroup\u0027s VIF info from extra to internal_info."},{"line_number":32,"context_line":""},{"line_number":33,"context_line":"    :param context: The context."}],"source_content_type":"text/x-python","patch_set":4,"id":"3f4b6375_9dc8c974","line":30,"in_reply_to":"3f4b6375_344e91c1","updated":"2017-10-30 18:18:32.000000000","message":"Yup. I did this cuz we should be able to delete it in 1-2 cycles, and we don\u0027t have a \u0027common\u0027 file for common code and I was too lazy to create one, esp if it just has this one function that we may end up deleting after the migration has been done.","commit_id":"3510aae0007855e501a5b738c8d7398e3d4f0041"}]}
