)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"7252d79286e82eea9eec53d34d00e97cc08efec4","unresolved":true,"context_lines":[{"line_number":8,"context_line":""},{"line_number":9,"context_line":"This reverts commit 85d3fff97e55ba85f72cda4365ad0441c10bd9f6."},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"Reason for revert: Using \"joined\" relation can cause very long query times when dealing with many(hundreds) of tags on a large deployments. The initial reason for this change was to reduce amount of SELECT logs during the testing, but given that this change can have negative side effects on a large deployments(in some corner cases), we will favor the original `subquery` option. See closes-bug link for further details."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Change-Id: I83ec349e49e1f343da8996cab149d76443120873"},{"line_number":14,"context_line":"Closes-bug: #2068761"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"8556482e_654b6ca2","line":11,"range":{"start_line":11,"start_character":152,"end_line":11,"end_character":158},"updated":"2024-06-07 20:50:30.000000000","message":"this commit message should explain why we now believe it was a bad reason; or what the next step is to tackle this presumed degradation.","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"7252d79286e82eea9eec53d34d00e97cc08efec4","unresolved":true,"context_lines":[{"line_number":8,"context_line":""},{"line_number":9,"context_line":"This reverts commit 85d3fff97e55ba85f72cda4365ad0441c10bd9f6."},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"Reason for revert: Using \"joined\" relation can cause very long query times when dealing with many(hundreds) of tags on a large deployments. The initial reason for this change was to reduce amount of SELECT logs during the testing, but given that this change can have negative side effects on a large deployments(in some corner cases), we will favor the original `subquery` option. See closes-bug link for further details."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Change-Id: I83ec349e49e1f343da8996cab149d76443120873"},{"line_number":14,"context_line":"Closes-bug: #2068761"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"441be52d_4e91bf61","line":11,"updated":"2024-06-07 20:50:30.000000000","message":"this should be wrapped.","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":34271,"name":"Miro Tomaska","display_name":"Miro Tomaska","email":"mtomaska@redhat.com","username":"mtomaska"},"change_message_id":"e96b78adc40942b71c8d4b067b814a275ecddea1","unresolved":true,"context_lines":[{"line_number":8,"context_line":""},{"line_number":9,"context_line":"This reverts commit 85d3fff97e55ba85f72cda4365ad0441c10bd9f6."},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"Reason for revert: Using \"joined\" relation can cause very long query times when dealing with many(hundreds) of tags on a large deployments. The initial reason for this change was to reduce amount of SELECT logs during the testing, but given that this change can have negative side effects on a large deployments(in some corner cases), we will favor the original `subquery` option. See closes-bug link for further details."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Change-Id: I83ec349e49e1f343da8996cab149d76443120873"},{"line_number":14,"context_line":"Closes-bug: #2068761"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"036b9297_9ab712d8","line":11,"in_reply_to":"441be52d_4e91bf61","updated":"2024-06-11 15:24:21.000000000","message":"Thanks, I used the gerrit revert option and the provided text box gave me impression that the text will be wrapped correctly. I will use my editor next time around.","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"a838d99bbdd7484064dbac40a8ae8600382bb12d","unresolved":true,"context_lines":[{"line_number":8,"context_line":""},{"line_number":9,"context_line":"This reverts commit 85d3fff97e55ba85f72cda4365ad0441c10bd9f6."},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"Reason for revert: Using \"joined\" relation can cause very long query times when dealing with many(hundreds) of tags on a large deployments. The initial reason for this change was to reduce amount of SELECT logs during the testing, but given that this change can have negative side effects on a large deployments(in some corner cases), we will favor the original `subquery` option. See closes-bug link for further details."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Change-Id: I83ec349e49e1f343da8996cab149d76443120873"},{"line_number":14,"context_line":"Closes-bug: #2068761"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"e65a7236_2816de2c","line":11,"range":{"start_line":11,"start_character":152,"end_line":11,"end_character":158},"in_reply_to":"8556482e_654b6ca2","updated":"2024-06-10 14:45:36.000000000","message":"Copying from elsewhere to clarify what I mean:\n\n```\nI’d like to see a narrative captured in commit message explaining why the original change was misguided - perhaps because the number of SELECTs is a bad indicator to track - btw I actually don’t know if that’s the case!\n\nalternatively - why it was a needed but incomplete change - and if so, what we may expect to regress by reverting it - and how this topic will be revisited (a bug report or whatever).\n\nbasically, there are two cases 1) the original patch was a complete mistake; 2) it had some legs but was executed incorrectly.\n\n(1) needs an explanation why we think so (2) needs a follow-up to avoid regressing on whatever the original patch was trying to resolve.\n```","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":34271,"name":"Miro Tomaska","display_name":"Miro Tomaska","email":"mtomaska@redhat.com","username":"mtomaska"},"change_message_id":"e96b78adc40942b71c8d4b067b814a275ecddea1","unresolved":false,"context_lines":[{"line_number":8,"context_line":""},{"line_number":9,"context_line":"This reverts commit 85d3fff97e55ba85f72cda4365ad0441c10bd9f6."},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"Reason for revert: Using \"joined\" relation can cause very long query times when dealing with many(hundreds) of tags on a large deployments. The initial reason for this change was to reduce amount of SELECT logs during the testing, but given that this change can have negative side effects on a large deployments(in some corner cases), we will favor the original `subquery` option. See closes-bug link for further details."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Change-Id: I83ec349e49e1f343da8996cab149d76443120873"},{"line_number":14,"context_line":"Closes-bug: #2068761"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"7de6a135_7f90353e","line":11,"range":{"start_line":11,"start_character":152,"end_line":11,"end_character":158},"in_reply_to":"8db0c1c8_0152fe8d","updated":"2024-06-11 15:24:21.000000000","message":"Done","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":6773,"name":"Lucas Alvares Gomes","email":"lucasagomes@gmail.com","username":"lucasagomes"},"change_message_id":"11052129b0c345bd15534e452d4eff839575cf73","unresolved":true,"context_lines":[{"line_number":8,"context_line":""},{"line_number":9,"context_line":"This reverts commit 85d3fff97e55ba85f72cda4365ad0441c10bd9f6."},{"line_number":10,"context_line":""},{"line_number":11,"context_line":"Reason for revert: Using \"joined\" relation can cause very long query times when dealing with many(hundreds) of tags on a large deployments. The initial reason for this change was to reduce amount of SELECT logs during the testing, but given that this change can have negative side effects on a large deployments(in some corner cases), we will favor the original `subquery` option. See closes-bug link for further details."},{"line_number":12,"context_line":""},{"line_number":13,"context_line":"Change-Id: I83ec349e49e1f343da8996cab149d76443120873"},{"line_number":14,"context_line":"Closes-bug: #2068761"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":1,"id":"8db0c1c8_0152fe8d","line":11,"range":{"start_line":11,"start_character":152,"end_line":11,"end_character":158},"in_reply_to":"e65a7236_2816de2c","updated":"2024-06-11 13:37:34.000000000","message":"\u003e Copying from elsewhere to clarify what I mean:\n\u003e \n\u003e ```\n\u003e I’d like to see a narrative captured in commit message explaining why the original change was misguided - perhaps because the number of SELECTs is a bad indicator to track - btw I actually don’t know if that’s the case!\n\u003e \n\u003e alternatively - why it was a needed but incomplete change - and if so, what we may expect to regress by reverting it - and how this topic will be revisited (a bug report or whatever).\n\u003e \n\u003e basically, there are two cases 1) the original patch was a complete mistake; 2) it had some legs but was executed incorrectly.\n\u003e \n\u003e (1) needs an explanation why we think so (2) needs a follow-up to avoid regressing on whatever the original patch was trying to resolve.\n\u003e ```","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"20088ec40881ce7df85cbf3ef1ca296d0e9c5227","unresolved":true,"context_lines":[{"line_number":16,"context_line":"would define hundreds of tags per port across a large deployment."},{"line_number":17,"context_line":"I am proposing to revert this change in favor of the old “subquery”"},{"line_number":18,"context_line":"relation in order to fix this regression. In addition, I filed the"},{"line_number":19,"context_line":"Related-Bug #2069061 to investigate using `selectin` as the more"},{"line_number":20,"context_line":"appropriate long term solution."},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"Change-Id: I83ec349e49e1f343da8996cab149d76443120873"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":2,"id":"8c8e8578_32b434c4","line":19,"updated":"2024-06-11 15:36:28.000000000","message":"there\u0027s already https://bugs.launchpad.net/neutron/+bug/2067770 for this\n\nI was thinking more about investigating the effect of this revert on the number of SELECTs. But I guess I will let it go. 😊","commit_id":"bf123dfb385f85077e7e6c277da3789afa02f422"}],"/PATCHSET_LEVEL":[{"author":{"_account_id":6773,"name":"Lucas Alvares Gomes","email":"lucasagomes@gmail.com","username":"lucasagomes"},"change_message_id":"c82ca9ba61991c397fd0127aa36ac997f1d82392","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"530a0c94_ba0fc98c","updated":"2024-06-11 14:40:21.000000000","message":"Downgrading vote until we get a consensus about reporting the original issue and updating the commit message","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"541f8ef343fa86acf417af59e214562b769d63ea","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"404d87ae_6cac6f8e","updated":"2024-06-11 14:10:36.000000000","message":"Guys, you are +2ing a patch with a line in commit message that spans hundreds of characters.","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"b1abb2675f28f28a1e08cd5f949d020f85c19b69","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"f4d97a23_267a68f4","updated":"2024-06-07 20:48:18.000000000","message":"I know the original patch was due to a high number of SELECTs observed in gate and vague concerns about this maybe being an indicator of performance issues. Do you plan to track the effect of this revert on this indicator? Or do we now believe that this indicator should be ignored completely?\n\nAfraid we may fix one issue while regressing back on some other scenario.","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":6773,"name":"Lucas Alvares Gomes","email":"lucasagomes@gmail.com","username":"lucasagomes"},"change_message_id":"11052129b0c345bd15534e452d4eff839575cf73","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"f60daf74_be63a6e4","updated":"2024-06-11 13:37:34.000000000","message":"This has been identified as a regression so I am in favor of reverting it until further investigation on how we can solve the original problem without the side-effect. I see the LP mentioning the \"selectin\" mode for this option in sqlalchemy 2.0. This is something that we can experiment with I guess. But for now, I think it\u0027s reasonable to make sure we do not introduce a regression.","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":6773,"name":"Lucas Alvares Gomes","email":"lucasagomes@gmail.com","username":"lucasagomes"},"change_message_id":"199a3949957ae2f236e9678770d1df18763727cc","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"ea45b5aa_c0fa5cb4","in_reply_to":"08846033_9ad6fab2","updated":"2024-06-11 14:20:26.000000000","message":"I would expect a LP for it. The original patch does not have one which is was bad form, otherwise we could just re-open it. \n\n@Miro or @Slaweq, what you think ? Can you create a LP for the original issue ?","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"40d7500644e53e3584a1d42fee93178459c5c6fc","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"bde640c5_caf80a02","in_reply_to":"4af4ac4a_fc25f20d","updated":"2024-06-11 14:12:25.000000000","message":"I don\u0027t question the patch - I don\u0027t have knowledge to do so. I question the fact the commit message doesn\u0027t provide enough explanation why the indicator is misguided / why we shouldn\u0027t follow up on it.\n\nThe original patch should have given more details about the reasoning. It didn\u0027t. I ask to at least add details now that we revert it.","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":11975,"name":"Slawek Kaplonski","email":"skaplons@redhat.com","username":"slaweq"},"change_message_id":"254898351e0b9d738b77bf3975630514fa659fbf","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"4af4ac4a_fc25f20d","in_reply_to":"f4d97a23_267a68f4","updated":"2024-06-11 07:12:30.000000000","message":"I personally think that we should not worry that much about that indicator now, at least in this example as we saw that this wasn\u0027t good optimization. Originally we were trying to optimize number of sql queries in the ci jobs to reduce load on the mysql server and ci resources but it seems that this wasn\u0027t good optimization simply.","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":9656,"name":"Ihar Hrachyshka","email":"ihrachys@redhat.com","username":"ihrachys","status":"Red Hat Networking Systems Engineer"},"change_message_id":"dcf82ff65061ec37c885abd1bf2add9474d6be9e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"08846033_9ad6fab2","in_reply_to":"f60daf74_be63a6e4","updated":"2024-06-11 14:14:22.000000000","message":"\u003e until further investigation\n\nwhere will this investigation happen? was a LP reported? Then it should be linked here, as Related-Bug.","commit_id":"52e4521ba8806f721059259a09cfa988a50e0839"},{"author":{"_account_id":6773,"name":"Lucas Alvares Gomes","email":"lucasagomes@gmail.com","username":"lucasagomes"},"change_message_id":"b7b6adda0ad0c1368ea4946bfb20a66643f14050","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"4588af86_40507134","updated":"2024-06-11 15:30:53.000000000","message":"Thanks Miro!","commit_id":"bf123dfb385f85077e7e6c277da3789afa02f422"},{"author":{"_account_id":11975,"name":"Slawek Kaplonski","email":"skaplons@redhat.com","username":"slaweq"},"change_message_id":"db997f755ce21dad401174be9b9beced8943140f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"3f8a8a15_fecdcb25","updated":"2024-06-11 15:34:07.000000000","message":"looks good for me but I would like Ihar to review before approving it.","commit_id":"bf123dfb385f85077e7e6c277da3789afa02f422"},{"author":{"_account_id":16688,"name":"Rodolfo Alonso","email":"ralonsoh@redhat.com","username":"rodolfo-alonso-hernandez"},"change_message_id":"51e659b6a0a71d30ea2a8a793a74f6b38bfa3986","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"be5de1c9_e7e5763a","updated":"2024-06-12 05:29:48.000000000","message":"recheck unrelated error in neutron-functional-with-uwsgi","commit_id":"bf123dfb385f85077e7e6c277da3789afa02f422"}]}