)]}'
{"specs/train/approved/reduce-quota-usage-when-shelve.rst":[{"author":{"_account_id":4690,"name":"melanie witt","display_name":"melwitt","email":"melwittt@gmail.com","username":"melwitt"},"change_message_id":"ebf630d7b53bfb401020137084b7945d92ed9e5a","unresolved":false,"context_lines":[{"line_number":14,"context_line":"Problem description"},{"line_number":15,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"Currently, when a server gets shelved its CPU and RAM quota isn\u0027t changed."},{"line_number":18,"context_line":"It makes shelve operation useless for end users as it doesn\u0027t motivate them"},{"line_number":19,"context_line":"to use this feature as it doesn\u0027t allow them to start new servers using"},{"line_number":20,"context_line":"existing quota."}],"source_content_type":"text/x-rst","patch_set":4,"id":"dfbec78f_b2b42637","line":17,"updated":"2019-05-02 17:42:28.000000000","message":"Are you talking of SHELVED or SHELVED_OFFLOADED? For SHELVED, I think it still makes sense to consume resources because the server is still present on the hypervisor. However, when the server becomes SHELVED_OFFLOADED, I agree it makes sense to free resources.\n\nThere\u0027s some work going on presently to enable quota usage counting of cores and ram from placement:\n\nhttps://review.opendev.org/638073\n\nand when that merges and with the feature enabled ([quota]count_usage_from_placement \u003d True), cores and ram usage will not be consumed while a server is in SHELVED_OFFLOADED state.","commit_id":"4675a118141d8816a15bdc31f5f549f1433f7e3a"},{"author":{"_account_id":4690,"name":"melanie witt","display_name":"melwitt","email":"melwittt@gmail.com","username":"melwitt"},"change_message_id":"10593085a8df664ce3ea1573178394d8732b3cee","unresolved":false,"context_lines":[{"line_number":14,"context_line":"Problem description"},{"line_number":15,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"Currently, when a server gets shelved its CPU and RAM quota isn\u0027t changed."},{"line_number":18,"context_line":"It makes shelve operation useless for end users as it doesn\u0027t motivate them"},{"line_number":19,"context_line":"to use this feature as it doesn\u0027t allow them to start new servers using"},{"line_number":20,"context_line":"existing quota."}],"source_content_type":"text/x-rst","patch_set":4,"id":"dfbec78f_1d40a307","line":17,"in_reply_to":"dfbec78f_927562fd","updated":"2019-05-02 18:25:30.000000000","message":"\u003e In https://review.opendev.org/#/c/638073/ you state that the\n \u003e changes proposed will not be beneficial to single cell deployments.\n \u003e Am I missing something here?\n\nSorry, I meant only in the context of down cell resiliency for multiple cells (in a single cell deployment you have to assume the one cell is always available and aren\u0027t concerned about handling some down cells). I\u0027ll change the wording to hopefully make it more clear in the patches.\n\nThere are other benefits of the proposed changes but the down cells was the main one I was aware of at the time.\n\n \u003e If in fact SHELVED_OFFLOADED servers don\u0027t have placement\n \u003e allocation then this spec is unnecessary as soon as your changes\n \u003e land.\n\nThat is true, if you use \u0027[quota]count_usage_from_placement \u003d True\u0027 and are also able to tolerate the changes in quota usage behavior for servers that are in the middle of a resize (VERIFY_RESIZE).","commit_id":"4675a118141d8816a15bdc31f5f549f1433f7e3a"},{"author":{"_account_id":12408,"name":"Maxim Nestratov","email":"mnestratov@virtuozzo.com","username":"mnestratov"},"change_message_id":"69cab414777d5fdf9d5eed83868bcccaafb538e6","unresolved":false,"context_lines":[{"line_number":14,"context_line":"Problem description"},{"line_number":15,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"Currently, when a server gets shelved its CPU and RAM quota isn\u0027t changed."},{"line_number":18,"context_line":"It makes shelve operation useless for end users as it doesn\u0027t motivate them"},{"line_number":19,"context_line":"to use this feature as it doesn\u0027t allow them to start new servers using"},{"line_number":20,"context_line":"existing quota."}],"source_content_type":"text/x-rst","patch_set":4,"id":"dfbec78f_927562fd","line":17,"in_reply_to":"dfbec78f_b2b42637","updated":"2019-05-02 18:00:21.000000000","message":"In https://review.opendev.org/#/c/638073/ you state that the changes proposed will not be beneficial to single cell deployments. Am I missing something here?\nIf in fact SHELVED_OFFLOADED servers don\u0027t have placement allocation then this spec is unnecessary as soon as your changes land.","commit_id":"4675a118141d8816a15bdc31f5f549f1433f7e3a"},{"author":{"_account_id":12408,"name":"Maxim Nestratov","email":"mnestratov@virtuozzo.com","username":"mnestratov"},"change_message_id":"c70a5bc1a99dd0bb3233e592b0d87d729bf4e236","unresolved":false,"context_lines":[{"line_number":14,"context_line":"Problem description"},{"line_number":15,"context_line":"\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d"},{"line_number":16,"context_line":""},{"line_number":17,"context_line":"Currently, when a server gets shelved its CPU and RAM quota isn\u0027t changed."},{"line_number":18,"context_line":"It makes shelve operation useless for end users as it doesn\u0027t motivate them"},{"line_number":19,"context_line":"to use this feature as it doesn\u0027t allow them to start new servers using"},{"line_number":20,"context_line":"existing quota."}],"source_content_type":"text/x-rst","patch_set":4,"id":"dfbec78f_b269868c","line":17,"in_reply_to":"dfbec78f_b2b42637","updated":"2019-05-02 17:49:42.000000000","message":"Thanks for pointing this out, sure I\u0027m talking about SHELVED_OFFLOADED","commit_id":"4675a118141d8816a15bdc31f5f549f1433f7e3a"}]}