)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":22629,"name":"Michal Nasiadka","email":"mnasiadka@gmail.com","username":"mnasiadka"},"change_message_id":"cd6683c438ed291b6978f9843296cb01212c5e6f","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":3,"id":"d3bc9ecf_a0438594","updated":"2025-08-08 19:54:50.000000000","message":"Changing overcommit_memory requires a reboot, and I don’t see why would we support this.\n\nRedis should be fine with its fork() memory routines on a properly sized controller, it only needs special overcommit settings if your redis db is large and you don’t have two times memory of the current process size for redis to be able to do fork().\nit should not happen since redis in Kolla only holds temporary coordination data","commit_id":"2c0ef2780a99dd52057d6e3eea3b49a197f20ef2"}],"ansible/roles/redis/tasks/config-host.yml":[{"author":{"_account_id":14200,"name":"Maksim Malchuk","email":"maksim.malchuk@gmail.com","username":"mmalchuk"},"change_message_id":"ea33a90a63759ae4dc8edd03541a25c174f7b84b","unresolved":true,"context_lines":[{"line_number":4,"context_line":"    name: sysctl"},{"line_number":5,"context_line":"  vars:"},{"line_number":6,"context_line":"    settings:"},{"line_number":7,"context_line":"      - { name: \"vm.overcommit_memory\", value: \"1\"}"},{"line_number":8,"context_line":"  when:"},{"line_number":9,"context_line":"    - set_sysctl | bool"}],"source_content_type":"text/x-yaml","patch_set":3,"id":"7a8cfe68_ff0fa093","line":7,"range":{"start_line":7,"start_character":10,"end_line":7,"end_character":50},"updated":"2025-08-08 17:17:22.000000000","message":"this is system-wide changes. may be this is good for Redis, but it can be bad for other deployments.","commit_id":"2c0ef2780a99dd52057d6e3eea3b49a197f20ef2"}]}