)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":17669,"name":"Doug Szumski","email":"doug@stackhpc.com","username":"DougSzumski"},"change_message_id":"07157e358dc79609df3ae6277fa69241e190aa17","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"c75ff3a2_52d5242e","updated":"2025-10-24 16:06:57.000000000","message":"Good job Seunghun. I\u0027m curious if we see a real word speed up in a typical deployment.","commit_id":"84078bdf85bb5c7d266155503d9aeea5baae6c35"},{"author":{"_account_id":35264,"name":"Alex Welsh","email":"alex@stackhpc.com","username":"alex-welsh"},"change_message_id":"fc7d1c56af8437d0c3890769d611a6252e6bdf15","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":1,"id":"1fdc5614_7ab1c7b3","updated":"2025-10-24 15:47:08.000000000","message":"Release note with some details on impact, previous default etc might be useful. If there\u0027s a bug report could throw that in there too","commit_id":"84078bdf85bb5c7d266155503d9aeea5baae6c35"},{"author":{"_account_id":17669,"name":"Doug Szumski","email":"doug@stackhpc.com","username":"DougSzumski"},"change_message_id":"07157e358dc79609df3ae6277fa69241e190aa17","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":1,"id":"fb70c3e2_854585e0","in_reply_to":"1fdc5614_7ab1c7b3","updated":"2025-10-24 16:06:57.000000000","message":"It\u0027s worth highlighting that it may take longer to recover, but performance should improve, and that sites may wish to tune the size based on their deployment.","commit_id":"84078bdf85bb5c7d266155503d9aeea5baae6c35"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"cc10cc1be1884931f85ccf5d279a724344cf81ca","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":1,"id":"b474c1ac_3405b85d","in_reply_to":"1fdc5614_7ab1c7b3","updated":"2025-10-24 16:00:00.000000000","message":"reno and bug report sounds good. Thanks.","commit_id":"84078bdf85bb5c7d266155503d9aeea5baae6c35"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"c47180cf5cb7b412be0ea5963b9c63277b0c6124","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"ea61f444_932b350b","in_reply_to":"fb70c3e2_854585e0","updated":"2025-10-27 12:02:43.000000000","message":"Done","commit_id":"84078bdf85bb5c7d266155503d9aeea5baae6c35"},{"author":{"_account_id":17669,"name":"Doug Szumski","email":"doug@stackhpc.com","username":"DougSzumski"},"change_message_id":"4c9993863b3290bfefdcdf41abc92649fbe878ca","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"c9bd0bbb_87f8970b","updated":"2025-10-27 12:08:34.000000000","message":"Looks good - it\u0027s perhaps worth adding an upgrade note as well.","commit_id":"fa84ab26ba9bafb31cc8e0a22e590abcb013e666"},{"author":{"_account_id":17669,"name":"Doug Szumski","email":"doug@stackhpc.com","username":"DougSzumski"},"change_message_id":"ed6cc0b6c473451f906f7ebc213a2ab143cf880c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"a53378ef_7a441088","in_reply_to":"0c41b4e0_265e1d18","updated":"2025-10-27 13:45:53.000000000","message":"I was thinking that if you were planning an upgrade, it\u0027s something important to be aware of. You might want to consider adjusting the value. You might want to do a test recovery to work out how long it would take with the new, larger value. Hopefully you aren\u0027t extremely tight on disk space etc. You could split the note up so that the feature is the new variable which can be used to set the value, and the upgrade note tells you to consider the recovery time and that the default has changed etc?\n\nIt all feels pretty trivial, but it might help someone.","commit_id":"fa84ab26ba9bafb31cc8e0a22e590abcb013e666"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"74bed7e91bf265b190bb2bfd2ac3cd7238fdd699","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"2d1416eb_4692b978","in_reply_to":"a53378ef_7a441088","updated":"2025-10-27 14:44:35.000000000","message":"Aha I see. Thank you. Yeah that sounds good. I\u0027ll amend the release note.","commit_id":"fa84ab26ba9bafb31cc8e0a22e590abcb013e666"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"d393c4a8a096531cc9f2188409cca2c0ff3a3dfd","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"0c41b4e0_265e1d18","in_reply_to":"c9bd0bbb_87f8970b","updated":"2025-10-27 13:03:38.000000000","message":"Sorry, what would be the topics on upgrade note? Or do we want to move features note to upgrade note?","commit_id":"fa84ab26ba9bafb31cc8e0a22e590abcb013e666"},{"author":{"_account_id":17669,"name":"Doug Szumski","email":"doug@stackhpc.com","username":"DougSzumski"},"change_message_id":"0eba5fb3e143471db3bd253071f70a45c527001c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"9e026449_a7f9fd82","updated":"2025-10-27 16:41:52.000000000","message":"It seems that this doesn\u0027t entirely fix the bug although it is still a useful tuning. Do you think it has helped reduce the failure rate?","commit_id":"4400468f76459511c08baaa38a7379050217320f"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"59df2f9acf435bf1453b3aa709a607c6395c0b30","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"b1bccf3a_be10fa13","in_reply_to":"9e026449_a7f9fd82","updated":"2025-10-29 09:37:00.000000000","message":"Ah that\u0027s a shame. But yeah I didn\u0027t see the same failure on Zuul runs of this patch","commit_id":"4400468f76459511c08baaa38a7379050217320f"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"4df4b4d76960a7133d232b0cdd7b07e7786a7197","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"becf5a7c_c867e835","updated":"2025-10-29 15:54:26.000000000","message":"Let\u0027s try with 2GB","commit_id":"fa8cc448b8cff1520c2819848e35a6321919085a"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"2abb44a18a4d94ec75b0590603dfa19fe5bf0de6","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"594b5d31_a6f90227","updated":"2025-10-30 08:49:38.000000000","message":"https://review.opendev.org/c/openstack/kolla-ansible/+/961675 also had successful CI run with 2GB of innodb_log_file_size","commit_id":"fa8cc448b8cff1520c2819848e35a6321919085a"},{"author":{"_account_id":22629,"name":"Michal Nasiadka","email":"mnasiadka@gmail.com","username":"mnasiadka"},"change_message_id":"23f0b4fb63393505ca134e051e287ee01d722935","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":6,"id":"428e9b7b_b4925139","updated":"2026-01-23 17:10:51.000000000","message":"I don\u0027t feel backporting this is safe - unless somebody proves me wrong","commit_id":"daa4d512ffd73c80777bfc226ea1065c3cabde4c"},{"author":{"_account_id":17669,"name":"Doug Szumski","email":"doug@stackhpc.com","username":"DougSzumski"},"change_message_id":"75f203c7a451c00ac6056e3445b09903182a9608","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"e1956052_111e62ff","updated":"2025-12-10 13:27:04.000000000","message":"I\u0027m not sure whether this closes the bug or just helps reduce the change of hitting it. \n\nIgnoring that, it provides a better default setting for typical deployments, without too much of a resource burden.\n\nNote, OSA is doing something similar by default (1GB):\n\nhttps://docs.openstack.org/openstack-ansible-galera_server/2025.1/","commit_id":"daa4d512ffd73c80777bfc226ea1065c3cabde4c"},{"author":{"_account_id":28048,"name":"Will Szumski","email":"will@stackhpc.com","username":"jovial"},"change_message_id":"b55b111d565586a3b46c48a739f5ba5e5fedadf7","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":6,"id":"04f3b2ee_2bece873","updated":"2026-02-10 11:00:16.000000000","message":"Seemed to help for me","commit_id":"daa4d512ffd73c80777bfc226ea1065c3cabde4c"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"6d0188d2c6d028e022f39e3e3c39bcec04f22807","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"e14e307b_fcf2eec9","updated":"2025-10-31 09:13:28.000000000","message":"recheck\nUnsure if the error \"[ERROR] WSREP: Receiving IST failed, node restart required: IST receiver reported failure: \u0027IST didn\u0027t contain all write sets, expected last: 25 last received: 24\u0027 (71)\" is related to this change.\nLet\u0027s see if this error is consistent.","commit_id":"daa4d512ffd73c80777bfc226ea1065c3cabde4c"},{"author":{"_account_id":17669,"name":"Doug Szumski","email":"doug@stackhpc.com","username":"DougSzumski"},"change_message_id":"1aade547f2a00a613c265498572c10b7b1246dff","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"e704caed_66e84f32","updated":"2026-02-11 11:11:32.000000000","message":"recheck - failed during bootstrap with ssh connection issues","commit_id":"daa4d512ffd73c80777bfc226ea1065c3cabde4c"},{"author":{"_account_id":13252,"name":"Dr. Jens Harbott","display_name":"Jens Harbott (frickler)","email":"frickler@offenerstapel.de","username":"jrosenboom"},"change_message_id":"95749f417248ffc7cf60375a19ccccdfc7c50f84","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"a73e181e_2318c82c","updated":"2026-02-11 08:34:06.000000000","message":"recheck get some fresh logs","commit_id":"daa4d512ffd73c80777bfc226ea1065c3cabde4c"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"3ae3904c47abaa9440a377bbe2f53d57325b898d","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"0303210c_606f790a","in_reply_to":"04f3b2ee_2bece873","updated":"2026-02-10 11:48:39.000000000","message":"Acknowledged","commit_id":"daa4d512ffd73c80777bfc226ea1065c3cabde4c"},{"author":{"_account_id":17669,"name":"Doug Szumski","email":"doug@stackhpc.com","username":"DougSzumski"},"change_message_id":"93baf9e4b368f5a38487f546c76c4a6a5df23ad6","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":6,"id":"d21069ab_8f4be53f","in_reply_to":"1c1b8d84_6e4b0cb0","updated":"2026-01-29 09:35:32.000000000","message":"Apparently, it can be changed dynamically so long as MariaDB \u003e\u003d 10.9. Check https://mariadb.com/docs/server/server-usage/storage-engines/innodb/innodb-system-variables#innodb_log_file_size\n\nNone-the-less, I agree, let\u0027s not risk backporting it.","commit_id":"daa4d512ffd73c80777bfc226ea1065c3cabde4c"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"ae4e43c3c880744010778e9c524150d98418e3a1","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":6,"id":"1c1b8d84_6e4b0cb0","in_reply_to":"428e9b7b_b4925139","updated":"2026-01-26 10:23:57.000000000","message":"I don\u0027t think I could see this problem in stable releases. We probably don\u0027t need to backport this.","commit_id":"daa4d512ffd73c80777bfc226ea1065c3cabde4c"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"5592d444dd52b0e9446ad05e17b8c930b4cf36ec","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"f00ab921_f2f46578","in_reply_to":"d21069ab_8f4be53f","updated":"2026-01-29 09:59:21.000000000","message":"Acknowledged","commit_id":"daa4d512ffd73c80777bfc226ea1065c3cabde4c"},{"author":{"_account_id":17669,"name":"Doug Szumski","email":"doug@stackhpc.com","username":"DougSzumski"},"change_message_id":"3e000f7dad8cc83dafbb39dc50b258dbd9b4f04e","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"2dc52908_358fa2ee","in_reply_to":"e1956052_111e62ff","updated":"2025-12-10 13:27:31.000000000","message":"*chance","commit_id":"daa4d512ffd73c80777bfc226ea1065c3cabde4c"}],"ansible/roles/mariadb/templates/galera.cnf.j2":[{"author":{"_account_id":17669,"name":"Doug Szumski","email":"doug@stackhpc.com","username":"DougSzumski"},"change_message_id":"07157e358dc79609df3ae6277fa69241e190aa17","unresolved":true,"context_lines":[{"line_number":57,"context_line":"innodb_buffer_pool_size \u003d \u00278192M\u0027"},{"line_number":58,"context_line":"{% endif %}"},{"line_number":59,"context_line":""},{"line_number":60,"context_line":"innodb_log_file_size \u003d \u00271G\u0027"},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"[server]"},{"line_number":63,"context_line":"pid-file\u003d/var/lib/mysql/mariadb.pid"}],"source_content_type":"text/x-jinja2","patch_set":1,"id":"d6ef2b83_072c7be3","line":60,"updated":"2025-10-24 16:06:57.000000000","message":"It seems that the main tradeoff from increasing this is crash recovery time. \n\nhttps://www.percona.com/blog/what-is-a-big-innodb_log_file_size/\n\nIt seems like a better default than that standard","commit_id":"84078bdf85bb5c7d266155503d9aeea5baae6c35"},{"author":{"_account_id":36765,"name":"Seunghun Lee","display_name":"Seunghun Lee","email":"seunghun@stackhpc.com","username":"seunghun1ee","status":"Cloud Engineer at StackHPC"},"change_message_id":"c47180cf5cb7b412be0ea5963b9c63277b0c6124","unresolved":false,"context_lines":[{"line_number":57,"context_line":"innodb_buffer_pool_size \u003d \u00278192M\u0027"},{"line_number":58,"context_line":"{% endif %}"},{"line_number":59,"context_line":""},{"line_number":60,"context_line":"innodb_log_file_size \u003d \u00271G\u0027"},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"[server]"},{"line_number":63,"context_line":"pid-file\u003d/var/lib/mysql/mariadb.pid"}],"source_content_type":"text/x-jinja2","patch_set":1,"id":"fe12ae2f_f6717423","line":60,"in_reply_to":"cf7898ce_fb49b2eb","updated":"2025-10-27 12:02:43.000000000","message":"Acknowledged","commit_id":"84078bdf85bb5c7d266155503d9aeea5baae6c35"},{"author":{"_account_id":17669,"name":"Doug Szumski","email":"doug@stackhpc.com","username":"DougSzumski"},"change_message_id":"64899205eae81e37514de2b7522cd4b5f75fb713","unresolved":true,"context_lines":[{"line_number":57,"context_line":"innodb_buffer_pool_size \u003d \u00278192M\u0027"},{"line_number":58,"context_line":"{% endif %}"},{"line_number":59,"context_line":""},{"line_number":60,"context_line":"innodb_log_file_size \u003d \u00271G\u0027"},{"line_number":61,"context_line":""},{"line_number":62,"context_line":"[server]"},{"line_number":63,"context_line":"pid-file\u003d/var/lib/mysql/mariadb.pid"}],"source_content_type":"text/x-jinja2","patch_set":1,"id":"cf7898ce_fb49b2eb","line":60,"in_reply_to":"d6ef2b83_072c7be3","updated":"2025-10-24 16:08:06.000000000","message":"*the standard.","commit_id":"84078bdf85bb5c7d266155503d9aeea5baae6c35"}]}