)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":28522,"name":"Hervé Beraud","email":"herveberaud.pro@gmail.com","username":"hberaud"},"change_message_id":"6e91a4d4afaa20bc2e24b2fc551356b9abfd999a","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"47c65b40_61e4570e","updated":"2022-01-20 18:13:50.000000000","message":"@Grzegorz @Takashi: FYI, to fully fix the TLS issue on OSP16-17 these changes and the parent patch also needs to be reflected in puppet as you did with socket-keepalive of patches https://review.opendev.org/c/openstack/puppet-ceilometer/+/816633","commit_id":"751ec77f7802dc04e78f5bbbb814e6a2b238e4f0"},{"author":{"_account_id":28522,"name":"Hervé Beraud","email":"herveberaud.pro@gmail.com","username":"hberaud"},"change_message_id":"d1fd2b69c1f6275d247056c007547c1ecb5f51a5","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"37b6e8e5_8a5575e0","updated":"2022-01-17 14:31:55.000000000","message":"Depends-on a dogpile.cache patch that\u0027s not yet merged and released: https://gerrit.sqlalchemy.org/c/sqlalchemy/dogpile.cache/+/3528 \nHowever, we are good to go with the 2 previous parent patches.","commit_id":"751ec77f7802dc04e78f5bbbb814e6a2b238e4f0"},{"author":{"_account_id":28522,"name":"Hervé Beraud","email":"herveberaud.pro@gmail.com","username":"hberaud"},"change_message_id":"cb4febfa7ca3f0457746b02d97eaff6c72ac9174","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"bb30db22_cb3dec64","updated":"2022-01-20 07:09:16.000000000","message":"Dogpile.cache changes are now merged [1] and released (1.1.5). I proposed a patch to bump the dogpile.cache version [2].\n\n[1] https://gerrit.sqlalchemy.org/c/sqlalchemy/dogpile.cache/+/3528\n[2] https://review.opendev.org/c/openstack/requirements/+/825427","commit_id":"751ec77f7802dc04e78f5bbbb814e6a2b238e4f0"},{"author":{"_account_id":28522,"name":"Hervé Beraud","email":"herveberaud.pro@gmail.com","username":"hberaud"},"change_message_id":"b04e8b9cbcb34f4f2f35094946e11f0b7db33fc9","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"477c43bc_0423c296","updated":"2022-01-20 18:10:39.000000000","message":"recheck\n\nThe dogpile.cache requirements is now updated [1].\n\n[1] https://review.opendev.org/c/openstack/requirements/+/825427","commit_id":"751ec77f7802dc04e78f5bbbb814e6a2b238e4f0"},{"author":{"_account_id":15334,"name":"Stephen Finucane","display_name":"stephenfin","email":"stephenfin@redhat.com","username":"sfinucan"},"change_message_id":"f791e6224ff904492c1078cbb63e0b8a0cf8b7a8","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"1ef23b84_2ef67a9a","updated":"2022-01-24 17:28:11.000000000","message":"I think bnemec\u0027s comment has been addressed and I\u0027m otherwise happy with this","commit_id":"cb118d04cea318df23831d7445f6b33c866cd0c6"}],"oslo_cache/_opts.py":[{"author":{"_account_id":6928,"name":"Ben Nemec","email":"openstack@nemebean.com","username":"bnemec"},"change_message_id":"a1a1ae6bb7dd06b3e8900ac49ca68ccf53baa3b8","unresolved":true,"context_lines":[{"line_number":192,"context_line":"            \u0027hashclient_retry_attempts\u0027,"},{"line_number":193,"context_line":"            min\u003d1,"},{"line_number":194,"context_line":"            default\u003d2,"},{"line_number":195,"context_line":"            help\u003d\u0027Amount of times a client should be tried \u0027"},{"line_number":196,"context_line":"            \u0027before it is marked dead and removed from the pool in \u0027"},{"line_number":197,"context_line":"            \u0027the HashClient\\\u0027s internal mechanisms.\u0027),"},{"line_number":198,"context_line":"        cfg.FloatOpt("}],"source_content_type":"text/x-python","patch_set":2,"id":"3be1f195_cfa93655","line":195,"updated":"2022-01-21 16:14:33.000000000","message":"I\u0027m wondering if we want to mention that enable_retry_client must be true for these to have any effect. It\u0027s not necessarily clear that retry client relates to hashclient.","commit_id":"cb118d04cea318df23831d7445f6b33c866cd0c6"},{"author":{"_account_id":28522,"name":"Hervé Beraud","email":"herveberaud.pro@gmail.com","username":"hberaud"},"change_message_id":"e845cde1435e40c5c127c25cc51b38695f22bab5","unresolved":true,"context_lines":[{"line_number":192,"context_line":"            \u0027hashclient_retry_attempts\u0027,"},{"line_number":193,"context_line":"            min\u003d1,"},{"line_number":194,"context_line":"            default\u003d2,"},{"line_number":195,"context_line":"            help\u003d\u0027Amount of times a client should be tried \u0027"},{"line_number":196,"context_line":"            \u0027before it is marked dead and removed from the pool in \u0027"},{"line_number":197,"context_line":"            \u0027the HashClient\\\u0027s internal mechanisms.\u0027),"},{"line_number":198,"context_line":"        cfg.FloatOpt("}],"source_content_type":"text/x-python","patch_set":2,"id":"4e04deb4_bbb4886b","line":195,"in_reply_to":"3be1f195_cfa93655","updated":"2022-01-24 08:59:11.000000000","message":"No, enable_retry_client don\u0027t need to be true to use hashclient_retry_attempts [1]... pymemcache come with 2 retry mechanisms. the first one is enable_retry_client which is an external wrapper that can wrap any memcached client [2], the second one is internal to the pymemcache\u0027s hashclient (a client) [3], so even if the wrapper is not activated the internal retrying feature of this client (used by dogpile.cache to communicate with pymemcache) can be used [4].\n\n[1] https://github.com/sqlalchemy/dogpile.cache/blob/main/dogpile/cache/backends/memcached.py#L605 \n[2] https://pymemcache.readthedocs.io/en/latest/getting_started.html#using-the-built-in-retrying-mechanism\n[3] https://github.com/pinterest/pymemcache/blob/master/pymemcache/client/hash.py#L66\n[4] https://github.com/sqlalchemy/dogpile.cache/blob/main/dogpile/cache/backends/memcached.py#L607","commit_id":"cb118d04cea318df23831d7445f6b33c866cd0c6"},{"author":{"_account_id":6928,"name":"Ben Nemec","email":"openstack@nemebean.com","username":"bnemec"},"change_message_id":"4234e25a2e268fe3b2edabbda8aef2ebbb253734","unresolved":false,"context_lines":[{"line_number":192,"context_line":"            \u0027hashclient_retry_attempts\u0027,"},{"line_number":193,"context_line":"            min\u003d1,"},{"line_number":194,"context_line":"            default\u003d2,"},{"line_number":195,"context_line":"            help\u003d\u0027Amount of times a client should be tried \u0027"},{"line_number":196,"context_line":"            \u0027before it is marked dead and removed from the pool in \u0027"},{"line_number":197,"context_line":"            \u0027the HashClient\\\u0027s internal mechanisms.\u0027),"},{"line_number":198,"context_line":"        cfg.FloatOpt("}],"source_content_type":"text/x-python","patch_set":2,"id":"a5d8f708_512046be","line":195,"in_reply_to":"4e04deb4_bbb4886b","updated":"2022-01-26 17:29:04.000000000","message":"Ah, okay. Thanks for the explanation!","commit_id":"cb118d04cea318df23831d7445f6b33c866cd0c6"}]}