)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"e5f45c17d2a23afe38bdfa8c0439c9fc7e05d5dd","unresolved":true,"context_lines":[{"line_number":17,"context_line":""},{"line_number":18,"context_line":"NOTE: Setting this to zero is actually undefined and will"},{"line_number":19,"context_line":"cause the kernel to choose random delay in the u32 space"},{"line_number":20,"context_line":"so is not advised. From __get_random_u32_below():"},{"line_number":21,"context_line":""},{"line_number":22,"context_line":"/*"},{"line_number":23,"context_line":" * This function is technically undefined for ceil \u003d\u003d 0, and in fact"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":7,"id":"bcab3736_56bf4c5b","line":20,"updated":"2023-01-13 23:36:59.000000000","message":"I realize I could have worded this better, can update if it doesn\u0027t name sense.","commit_id":"55fc83df05a4f10e1c7d32ac4feca82c7ddf5071"}],"/PATCHSET_LEVEL":[{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"a9122b81b3b9449c0a4979f30e56f034194df90f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":4,"id":"10df216a_63e43096","updated":"2023-01-05 21:34:59.000000000","message":"I would be fine putting proxy_delay to zero (and having it before the other value), so will un-abandon this so we can update. Ed - let me know if you want me to push the update.","commit_id":"792ab31d56828a1bad658699cc3471ec00d154b5"},{"author":{"_account_id":11975,"name":"Slawek Kaplonski","email":"skaplons@redhat.com","username":"slaweq"},"change_message_id":"e5f9ede5b0221f914e18374fabb6a9e4e93ac7ad","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"8c3f7405_0aaa29b3","updated":"2023-01-10 08:59:06.000000000","message":"It seems that functional tests failures may be related to this patch. Please check it before rechecking. Logs: https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_3c0/782570/5/check/neutron-functional-with-uwsgi/3c0055c/testr_results.html","commit_id":"8fe5008f3853e8df9d198ec819d82e5ff984cc2b"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"fddbe44e8ddda6b39db99e73638331ce76b922ee","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"d4208c2b_f241a726","updated":"2023-01-11 15:03:12.000000000","message":"recheck test_direct_route_for_address_scope failure\n\nI didn\u0027t see anything obvious in the logs, will try once more and locally too","commit_id":"8fe5008f3853e8df9d198ec819d82e5ff984cc2b"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"b5988f95679570a8415b0e5b9fac085d19fe1201","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"613aebb1_0390cdf3","updated":"2023-01-12 15:20:57.000000000","message":"recheck test_direct_route_for_address_scope failure\n\nI ran this locally and it passes, so will try once more. The only commonality is it\u0027s the same tests in three classes.","commit_id":"8fe5008f3853e8df9d198ec819d82e5ff984cc2b"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"9b2b41c784866ce978b6d3e97d4f7286e4b6af78","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":5,"id":"19567729_a9f9e4a4","updated":"2023-01-06 23:01:36.000000000","message":"recheck unrelated functional test failure","commit_id":"8fe5008f3853e8df9d198ec819d82e5ff984cc2b"},{"author":{"_account_id":16688,"name":"Rodolfo Alonso","email":"ralonsoh@redhat.com","username":"rodolfo-alonso-hernandez"},"change_message_id":"a5b8fe36bc12017f5618a861f49b4f566ffe7027","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"5d358012_3a3a4380","updated":"2023-01-13 17:27:19.000000000","message":"Please check \"test_direct_route_for_address_scope\", the error looks related to this patch.","commit_id":"bbed658840a1ee4b9663d21fed741fcbabcd8745"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"478845c58af62f5e3352af698a610c97e409be62","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":6,"id":"09f759da_242842d6","updated":"2023-01-13 22:25:40.000000000","message":"Think I figured out the problem after looking at the kernel source code, if so I\u0027ll work on addressing that, but will push an update soon that should work without it.","commit_id":"bbed658840a1ee4b9663d21fed741fcbabcd8745"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"e5f45c17d2a23afe38bdfa8c0439c9fc7e05d5dd","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":7,"id":"c3dc64e7_066731f6","updated":"2023-01-13 23:36:59.000000000","message":"Functional test success 😊","commit_id":"55fc83df05a4f10e1c7d32ac4feca82c7ddf5071"},{"author":{"_account_id":6737,"name":"Edward Hope-Morley","email":"edward.hope-morley@canonical.com","username":"hopem"},"change_message_id":"e6f6e56cee4a2d9b437077e45071072cadfdf5c4","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":9,"id":"d042e4b5_c6e0e537","updated":"2023-01-16 10:18:05.000000000","message":"thanks for the updates brian!","commit_id":"d7f68a0ce76ffb9a93dfba167dfffba53189350d"}],"neutron/agent/l3/dvr_fip_ns.py":[{"author":{"_account_id":9531,"name":"liuyulong","display_name":"LIU Yulong","email":"i@liuyulong.me","username":"LIU-Yulong"},"change_message_id":"4deed79bda385be689d4bbfd7bcbae96e7058803","unresolved":true,"context_lines":[{"line_number":193,"context_line":""},{"line_number":194,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.conf.%s.proxy_arp\u003d1\u0027 % interface_name]"},{"line_number":195,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":196,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.neigh.%s.proxy_delay\u003d5\u0027 %"},{"line_number":197,"context_line":"               interface_name]"},{"line_number":198,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":199,"context_line":""}],"source_content_type":"text/x-python","patch_set":2,"id":"95165797_ced5f4a2","line":196,"range":{"start_line":196,"start_character":62,"end_line":196,"end_character":63},"updated":"2021-03-24 11:22:20.000000000","message":"So the unit here is ms (millisecond)？Why the value is 5? Is there a best practice?\n\nA self-explain constant defination should be good for this number here, how about:\n\nARP_PROXY_DELAY_MS \u003d 5\n\nOn the other hand, chaning the value of kernel gc_thresh, can be a workaround to overcome this? From my personal experiences, the network node which is serving large number of IPs should change the gc_thresh1/gc_thresh2/gc_thresh3 to higher value, especially when they are facing the internet directly.","commit_id":"a09c42af302a72c2fc1a5348a7953e4bcacb4bba"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"9b2b41c784866ce978b6d3e97d4f7286e4b6af78","unresolved":false,"context_lines":[{"line_number":193,"context_line":""},{"line_number":194,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.conf.%s.proxy_arp\u003d1\u0027 % interface_name]"},{"line_number":195,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":196,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.neigh.%s.proxy_delay\u003d5\u0027 %"},{"line_number":197,"context_line":"               interface_name]"},{"line_number":198,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":199,"context_line":""}],"source_content_type":"text/x-python","patch_set":2,"id":"f01ee9ec_7b98a522","line":196,"range":{"start_line":196,"start_character":62,"end_line":196,"end_character":63},"in_reply_to":"95165797_ced5f4a2","updated":"2023-01-06 23:01:36.000000000","message":"Done","commit_id":"a09c42af302a72c2fc1a5348a7953e4bcacb4bba"},{"author":{"_account_id":5948,"name":"Oleg Bondarev","email":"obondarev@mirantis.com","username":"obondarev"},"change_message_id":"7a540c0229b0b5cc30f97a6bf9d77a8ad42879dd","unresolved":true,"context_lines":[{"line_number":193,"context_line":""},{"line_number":194,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.conf.%s.proxy_arp\u003d1\u0027 % interface_name]"},{"line_number":195,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":196,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.neigh.%s.proxy_delay\u003d5\u0027 %"},{"line_number":197,"context_line":"               interface_name]"},{"line_number":198,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":199,"context_line":""}],"source_content_type":"text/x-python","patch_set":3,"id":"d1b7edb3_475f3819","line":196,"range":{"start_line":196,"start_character":62,"end_line":196,"end_character":63},"updated":"2021-03-28 08:40:57.000000000","message":"Please make it a const as per LIU\u0027s suggestion","commit_id":"d24f72696059f3cd4c9c2bf5b3e451b2ca991681"},{"author":{"_account_id":11975,"name":"Slawek Kaplonski","email":"skaplons@redhat.com","username":"slaweq"},"change_message_id":"f81635c2672a468e26daef106af293af96ce9ae7","unresolved":true,"context_lines":[{"line_number":193,"context_line":""},{"line_number":194,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.conf.%s.proxy_arp\u003d1\u0027 % interface_name]"},{"line_number":195,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":196,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.neigh.%s.proxy_delay\u003d5\u0027 %"},{"line_number":197,"context_line":"               interface_name]"},{"line_number":198,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":199,"context_line":""}],"source_content_type":"text/x-python","patch_set":3,"id":"55b27d22_5d83be58","line":196,"range":{"start_line":196,"start_character":62,"end_line":196,"end_character":64},"updated":"2021-03-26 08:47:43.000000000","message":"what is the unit of that? I just found that this setting defaults to 0.8 second, so is that 5 seconds now? Or 5 miliseconds? Or what exactly? Can You explain a bit more?","commit_id":"d24f72696059f3cd4c9c2bf5b3e451b2ca991681"},{"author":{"_account_id":6737,"name":"Edward Hope-Morley","email":"edward.hope-morley@canonical.com","username":"hopem"},"change_message_id":"7262dd91eb5dfda60d89a10afb6ebe026c67ffc6","unresolved":true,"context_lines":[{"line_number":193,"context_line":""},{"line_number":194,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.conf.%s.proxy_arp\u003d1\u0027 % interface_name]"},{"line_number":195,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":196,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.neigh.%s.proxy_delay\u003d5\u0027 %"},{"line_number":197,"context_line":"               interface_name]"},{"line_number":198,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":199,"context_line":""}],"source_content_type":"text/x-python","patch_set":3,"id":"e03796c4_122e51ca","line":196,"range":{"start_line":196,"start_character":62,"end_line":196,"end_character":64},"in_reply_to":"55b27d22_5d83be58","updated":"2021-03-26 15:49:16.000000000","message":"Hi Slaweq, the unit is jiffies where one jiffie is (typically) 100 meaning one jiffie is 0.01s. The defaults for this sysctl are:\n\nubuntu@focal:~$ sysctl net.ipv4.neigh.default.proxy_delay\nnet.ipv4.neigh.default.proxy_delay \u003d 80\n\nwhere\n\nubuntu@focal:~$ getconf CLK_TCK\n100\n\ni.e. default is 800ms and we are now setting to 50ms which feels like a more palatable max value.","commit_id":"d24f72696059f3cd4c9c2bf5b3e451b2ca991681"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"9b2b41c784866ce978b6d3e97d4f7286e4b6af78","unresolved":false,"context_lines":[{"line_number":193,"context_line":""},{"line_number":194,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.conf.%s.proxy_arp\u003d1\u0027 % interface_name]"},{"line_number":195,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":196,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.neigh.%s.proxy_delay\u003d5\u0027 %"},{"line_number":197,"context_line":"               interface_name]"},{"line_number":198,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":199,"context_line":""}],"source_content_type":"text/x-python","patch_set":3,"id":"a895fdf8_5d8ded68","line":196,"range":{"start_line":196,"start_character":62,"end_line":196,"end_character":63},"in_reply_to":"d1b7edb3_475f3819","updated":"2023-01-06 23:01:36.000000000","message":"Done","commit_id":"d24f72696059f3cd4c9c2bf5b3e451b2ca991681"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"9b2b41c784866ce978b6d3e97d4f7286e4b6af78","unresolved":false,"context_lines":[{"line_number":193,"context_line":""},{"line_number":194,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.conf.%s.proxy_arp\u003d1\u0027 % interface_name]"},{"line_number":195,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":196,"context_line":"        cmd \u003d [\u0027sysctl\u0027, \u0027-w\u0027, \u0027net.ipv4.neigh.%s.proxy_delay\u003d5\u0027 %"},{"line_number":197,"context_line":"               interface_name]"},{"line_number":198,"context_line":"        ip_wrapper.netns.execute(cmd, check_exit_code\u003dFalse, privsep_exec\u003dTrue)"},{"line_number":199,"context_line":""}],"source_content_type":"text/x-python","patch_set":3,"id":"dfc9a0ab_90a5f659","line":196,"range":{"start_line":196,"start_character":62,"end_line":196,"end_character":64},"in_reply_to":"e03796c4_122e51ca","updated":"2023-01-06 23:01:36.000000000","message":"Done","commit_id":"d24f72696059f3cd4c9c2bf5b3e451b2ca991681"},{"author":{"_account_id":9531,"name":"liuyulong","display_name":"LIU Yulong","email":"i@liuyulong.me","username":"LIU-Yulong"},"change_message_id":"ba2c4a36d0283c37ca58a0130ecaf6526009f867","unresolved":true,"context_lines":[{"line_number":45,"context_line":"FIP_PR_END \u003d FIP_PR_START + 40000"},{"line_number":46,"context_line":"# Fixed rule priority for Fast Path Exit rules"},{"line_number":47,"context_line":"FAST_PATH_EXIT_PR \u003d 80000"},{"line_number":48,"context_line":"SYSCTL_PROXY_DELAY_JIFFIES \u003d 5"},{"line_number":49,"context_line":""},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"class FipNamespace(namespaces.Namespace):"}],"source_content_type":"text/x-python","patch_set":4,"id":"cb100209_a8525ca6","line":48,"range":{"start_line":48,"start_character":19,"end_line":48,"end_character":30},"updated":"2021-03-31 15:16:42.000000000","message":"Maybe we can hide this long stroy to a defination with suffix \"_MS\" here by inverse operation of kernel related algorithms.\n\ndef get_kernel_config_of_hz():\n    return X\n\nHZ \u003d get_kernel_config_of_hz()\n\nSYSCTL_ARP_PROXY_MAX_DELAY_MS \u003d 50  (50 ms is our wanted value, so jiffies then will be based on the kernel HZ).\n\nnet.ipv4.neigh.%s.proxy_delay\u003d (SYSCTL_ARP_PROXY_MAX_DELAY_MS / 1000) * HZ.","commit_id":"792ab31d56828a1bad658699cc3471ec00d154b5"},{"author":{"_account_id":16688,"name":"Rodolfo Alonso","email":"ralonsoh@redhat.com","username":"rodolfo-alonso-hernandez"},"change_message_id":"9b9e45aadac3b11b69004d244e8270d107aeb494","unresolved":true,"context_lines":[{"line_number":45,"context_line":"FIP_PR_END \u003d FIP_PR_START + 40000"},{"line_number":46,"context_line":"# Fixed rule priority for Fast Path Exit rules"},{"line_number":47,"context_line":"FAST_PATH_EXIT_PR \u003d 80000"},{"line_number":48,"context_line":"SYSCTL_PROXY_DELAY_JIFFIES \u003d 5"},{"line_number":49,"context_line":""},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"class FipNamespace(namespaces.Namespace):"}],"source_content_type":"text/x-python","patch_set":4,"id":"d0f0dfc6_cdaec447","line":48,"range":{"start_line":48,"start_character":19,"end_line":48,"end_character":26},"updated":"2021-03-31 14:02:06.000000000","message":"Quite curious this time unit (dependent on the arch https://www.netadmintools.com/html/7arp.man.html).\n\nMaybe this value, if needed, could be a config knob (but only if needed, not now).","commit_id":"792ab31d56828a1bad658699cc3471ec00d154b5"},{"author":{"_account_id":21107,"name":"Trent Lloyd","email":"trent.lloyd@canonical.com","username":"lathiat"},"change_message_id":"108ed021f3caf7946db1dafd684b775c5cd711a7","unresolved":true,"context_lines":[{"line_number":45,"context_line":"FIP_PR_END \u003d FIP_PR_START + 40000"},{"line_number":46,"context_line":"# Fixed rule priority for Fast Path Exit rules"},{"line_number":47,"context_line":"FAST_PATH_EXIT_PR \u003d 80000"},{"line_number":48,"context_line":"SYSCTL_PROXY_DELAY_JIFFIES \u003d 5"},{"line_number":49,"context_line":""},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"class FipNamespace(namespaces.Namespace):"}],"source_content_type":"text/x-python","patch_set":4,"id":"d74a46aa_bde154a3","line":48,"range":{"start_line":48,"start_character":19,"end_line":48,"end_character":26},"in_reply_to":"31734c7a_8b38725b","updated":"2021-05-11 12:37:13.000000000","message":"I think we could solve this whole problem by simply disabling it. It might be a sensible default but we know what software we\u0027re running, we don\u0027t need it, we should just disable it.\n\nCalico appears to do exactly that (issue introducing the fix, the lines have changed, but it remains):\nhttps://github.com/projectcalico/felix/issues/164\n\nJust disable it and it resolves any issue surrounding unit calculations etc. That issue also suggests even having a smaller value can introduce extra unnecessary delay to the response.\n\nI would note however there is apparently a kernel bug that can be triggered in such a case that we may need to be mindful of - it suggests to set proxy_delay -before- proxy_arp\nhttps://github.com/torvalds/linux/commit/a533b70a657c03137dd49cbcfee70aac086ab2b1\nhttps://github.com/projectcalico/felix/issues/1937\n\nFWIW the reason for having proxy_delay at all seems lost to the sands of time, it seems pre-git (pre-2005) and there are multiple sources stating it\u0027s unclear why other than a comment to \u0027prevent looping in some cases\u0027 that is equally as old.","commit_id":"792ab31d56828a1bad658699cc3471ec00d154b5"},{"author":{"_account_id":9531,"name":"liuyulong","display_name":"LIU Yulong","email":"i@liuyulong.me","username":"LIU-Yulong"},"change_message_id":"98834a46fbb73c45e56c6c45bb57d58fb161b63f","unresolved":true,"context_lines":[{"line_number":45,"context_line":"FIP_PR_END \u003d FIP_PR_START + 40000"},{"line_number":46,"context_line":"# Fixed rule priority for Fast Path Exit rules"},{"line_number":47,"context_line":"FAST_PATH_EXIT_PR \u003d 80000"},{"line_number":48,"context_line":"SYSCTL_PROXY_DELAY_JIFFIES \u003d 5"},{"line_number":49,"context_line":""},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"class FipNamespace(namespaces.Namespace):"}],"source_content_type":"text/x-python","patch_set":4,"id":"31734c7a_8b38725b","line":48,"range":{"start_line":48,"start_character":19,"end_line":48,"end_character":26},"in_reply_to":"584dd416_531e735f","updated":"2021-03-31 15:23:29.000000000","message":"Or a shorter name:\nARP_PROXY_MAX_DELAY_JIFFIES \u003d 5","commit_id":"792ab31d56828a1bad658699cc3471ec00d154b5"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"9b2b41c784866ce978b6d3e97d4f7286e4b6af78","unresolved":false,"context_lines":[{"line_number":45,"context_line":"FIP_PR_END \u003d FIP_PR_START + 40000"},{"line_number":46,"context_line":"# Fixed rule priority for Fast Path Exit rules"},{"line_number":47,"context_line":"FAST_PATH_EXIT_PR \u003d 80000"},{"line_number":48,"context_line":"SYSCTL_PROXY_DELAY_JIFFIES \u003d 5"},{"line_number":49,"context_line":""},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"class FipNamespace(namespaces.Namespace):"}],"source_content_type":"text/x-python","patch_set":4,"id":"9133fec1_05d773bd","line":48,"range":{"start_line":48,"start_character":19,"end_line":48,"end_character":30},"in_reply_to":"cb100209_a8525ca6","updated":"2023-01-06 23:01:36.000000000","message":"Done","commit_id":"792ab31d56828a1bad658699cc3471ec00d154b5"},{"author":{"_account_id":9531,"name":"liuyulong","display_name":"LIU Yulong","email":"i@liuyulong.me","username":"LIU-Yulong"},"change_message_id":"ba2c4a36d0283c37ca58a0130ecaf6526009f867","unresolved":true,"context_lines":[{"line_number":45,"context_line":"FIP_PR_END \u003d FIP_PR_START + 40000"},{"line_number":46,"context_line":"# Fixed rule priority for Fast Path Exit rules"},{"line_number":47,"context_line":"FAST_PATH_EXIT_PR \u003d 80000"},{"line_number":48,"context_line":"SYSCTL_PROXY_DELAY_JIFFIES \u003d 5"},{"line_number":49,"context_line":""},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"class FipNamespace(namespaces.Namespace):"}],"source_content_type":"text/x-python","patch_set":4,"id":"584dd416_531e735f","line":48,"range":{"start_line":48,"start_character":19,"end_line":48,"end_character":26},"in_reply_to":"d0f0dfc6_cdaec447","updated":"2021-03-31 15:16:42.000000000","message":"Looks a bit complicated based on the kernel implementation [1]. So it is a variable value based on different kernel version and hardware.\nOne example, if HZ \u003d 1 tick(s)/sec, 5 jiffies will be 5 seconds. (never mind, just a simple example.)😊\nThere are some example in the kernel header: \"100 Hz for the SunOS kernel, 256 Hz for the Ultrix kernel and 1024 Hz\". So maybe in some deployment, \"5\" will not be the real value they want.\n\nThe delay value is variable not fixed, so the name here (with ARP) should be:\n\nSYSCTL_ARP_PROXY_MAX_DELAY_JIFFIES \u003d 5 (-1 for this)\n\n[1] https://github.com/torvalds/linux/blob/master/include/linux/jiffies.h","commit_id":"792ab31d56828a1bad658699cc3471ec00d154b5"},{"author":{"_account_id":1131,"name":"Brian Haley","email":"haleyb.dev@gmail.com","username":"brian-haley"},"change_message_id":"9b2b41c784866ce978b6d3e97d4f7286e4b6af78","unresolved":false,"context_lines":[{"line_number":45,"context_line":"FIP_PR_END \u003d FIP_PR_START + 40000"},{"line_number":46,"context_line":"# Fixed rule priority for Fast Path Exit rules"},{"line_number":47,"context_line":"FAST_PATH_EXIT_PR \u003d 80000"},{"line_number":48,"context_line":"SYSCTL_PROXY_DELAY_JIFFIES \u003d 5"},{"line_number":49,"context_line":""},{"line_number":50,"context_line":""},{"line_number":51,"context_line":"class FipNamespace(namespaces.Namespace):"}],"source_content_type":"text/x-python","patch_set":4,"id":"3741785a_0f41ba37","line":48,"range":{"start_line":48,"start_character":19,"end_line":48,"end_character":26},"in_reply_to":"d74a46aa_bde154a3","updated":"2023-01-06 23:01:36.000000000","message":"Done","commit_id":"792ab31d56828a1bad658699cc3471ec00d154b5"}]}