)]}'
{"/COMMIT_MSG":[{"author":{"_account_id":18002,"name":"John Fulton","email":"fulton@redhat.com","username":"fultonj"},"change_message_id":"dedf242a5fa11742c5682d9180823352ee839905","unresolved":false,"context_lines":[{"line_number":6,"context_line":""},{"line_number":7,"context_line":"Document the Derived Parameters feature"},{"line_number":8,"context_line":""},{"line_number":9,"context_line":"Depends-On: I26c29d2dad02e753688cc9f21c9f375aa48cfda0"},{"line_number":10,"context_line":"Change-Id: Id31218797b9a757eb31eb4574980c27718df0c1d"}],"source_content_type":"text/x-gerrit-commit-message","patch_set":14,"id":"9f560f44_3504de49","line":9,"range":{"start_line":9,"start_character":0,"end_line":9,"end_character":53},"updated":"2020-09-10 16:27:21.000000000","message":"Not everything described in this documentation will be true until this depends on patch merges.","commit_id":"b7936b24eb95486f1fdba784d756921128e991b2"}],"deploy-guide/source/features/derived_parameters.rst":[{"author":{"_account_id":21129,"name":"Alan Bishop","email":"abishopsweng@gmail.com","username":"ASBishop","status":"ex Red Hat"},"change_message_id":"87053fa394e9bdcce3ccbdca2085a25dc5443589","unresolved":false,"context_lines":[{"line_number":127,"context_line":"``ironic_node_id`` and that the average guest will use 50% of its CPU"},{"line_number":128,"context_line":"and will use 8 GB of RAM. The ``heat_environment_input_file`` must"},{"line_number":129,"context_line":"be set to the path of the Heat environment file where the"},{"line_number":130,"context_line":"``CephAnsibleDisksConfig`` parameter is set. This parameter is used"},{"line_number":131,"context_line":"define which disks are used as Ceph OSDs and might look like the"},{"line_number":132,"context_line":"following::"},{"line_number":133,"context_line":""},{"line_number":134,"context_line":"  CephAnsibleDisksConfig:"}],"source_content_type":"text/x-rst","patch_set":11,"id":"9f560f44_dd75f7cd","line":131,"range":{"start_line":130,"start_character":45,"end_line":131,"end_character":6},"updated":"2020-09-08 18:54:31.000000000","message":"Should this be \"This parameter is used _to_ define...\" (\"to\" missing?)?","commit_id":"fd2fc169903343e84af9775959f89d7f86680d00"},{"author":{"_account_id":18002,"name":"John Fulton","email":"fulton@redhat.com","username":"fultonj"},"change_message_id":"0a1a0c74e7530c70d61aa03f81c335beed71d03f","unresolved":false,"context_lines":[{"line_number":127,"context_line":"``ironic_node_id`` and that the average guest will use 50% of its CPU"},{"line_number":128,"context_line":"and will use 8 GB of RAM. The ``heat_environment_input_file`` must"},{"line_number":129,"context_line":"be set to the path of the Heat environment file where the"},{"line_number":130,"context_line":"``CephAnsibleDisksConfig`` parameter is set. This parameter is used"},{"line_number":131,"context_line":"define which disks are used as Ceph OSDs and might look like the"},{"line_number":132,"context_line":"following::"},{"line_number":133,"context_line":""},{"line_number":134,"context_line":"  CephAnsibleDisksConfig:"}],"source_content_type":"text/x-rst","patch_set":11,"id":"9f560f44_be090f2c","line":131,"range":{"start_line":130,"start_character":45,"end_line":131,"end_character":6},"in_reply_to":"9f560f44_dd75f7cd","updated":"2020-09-09 11:02:30.000000000","message":"Done","commit_id":"fd2fc169903343e84af9775959f89d7f86680d00"},{"author":{"_account_id":21129,"name":"Alan Bishop","email":"abishopsweng@gmail.com","username":"ASBishop","status":"ex Red Hat"},"change_message_id":"87053fa394e9bdcce3ccbdca2085a25dc5443589","unresolved":false,"context_lines":[{"line_number":188,"context_line":"  [stack@undercloud ~]$"},{"line_number":189,"context_line":""},{"line_number":190,"context_line":"The above could be used during a deployment by running a command like"},{"line_number":191,"context_line":"``openstack overcloud deploy ... -e /home/stack/hci_result.yaml``."},{"line_number":192,"context_line":""},{"line_number":193,"context_line":"The second file, defined by the ``report_path`` variable, will contain"},{"line_number":194,"context_line":"an explanation of how the parameters were derived and what relevant"}],"source_content_type":"text/x-rst","patch_set":11,"id":"9f560f44_dd2337bd","line":191,"updated":"2020-09-08 18:54:31.000000000","message":"Should we recommend the user add the env file near the end of their deployment command? That will ensure the derived values take precedence.","commit_id":"fd2fc169903343e84af9775959f89d7f86680d00"},{"author":{"_account_id":18002,"name":"John Fulton","email":"fulton@redhat.com","username":"fultonj"},"change_message_id":"0a1a0c74e7530c70d61aa03f81c335beed71d03f","unresolved":false,"context_lines":[{"line_number":188,"context_line":"  [stack@undercloud ~]$"},{"line_number":189,"context_line":""},{"line_number":190,"context_line":"The above could be used during a deployment by running a command like"},{"line_number":191,"context_line":"``openstack overcloud deploy ... -e /home/stack/hci_result.yaml``."},{"line_number":192,"context_line":""},{"line_number":193,"context_line":"The second file, defined by the ``report_path`` variable, will contain"},{"line_number":194,"context_line":"an explanation of how the parameters were derived and what relevant"}],"source_content_type":"text/x-rst","patch_set":11,"id":"9f560f44_3e15ff3d","line":191,"in_reply_to":"9f560f44_dd2337bd","updated":"2020-09-09 11:02:30.000000000","message":"Done","commit_id":"fd2fc169903343e84af9775959f89d7f86680d00"},{"author":{"_account_id":21129,"name":"Alan Bishop","email":"abishopsweng@gmail.com","username":"ASBishop","status":"ex Red Hat"},"change_message_id":"87053fa394e9bdcce3ccbdca2085a25dc5443589","unresolved":false,"context_lines":[{"line_number":198,"context_line":"  [stack@undercloud ~]$ cat /home/stack/hci_report.txt"},{"line_number":199,"context_line":"  Derived Parameters results"},{"line_number":200,"context_line":"   Inputs:"},{"line_number":201,"context_line":"   - Total host RAM in GB: 7"},{"line_number":202,"context_line":"   - Total host vCPUs: 8"},{"line_number":203,"context_line":"   - Ceph OSDs per host: 1"},{"line_number":204,"context_line":"   - Average guest memory size in GB: 8"}],"source_content_type":"text/x-rst","patch_set":11,"id":"9f560f44_584c5575","line":201,"updated":"2020-09-08 18:54:31.000000000","message":"I\u0027m guessing these numbers come from a very lightweight dev system (probably a VM). Is there any way to get data that\u0027s more reflective of a real world deployment?\n\nI mention this because the output on L208 might confuse readers (\"number of guests allowed based on memory \u003d 0\" doesn\u0027t make for a very interesting compute node :-/)","commit_id":"fd2fc169903343e84af9775959f89d7f86680d00"},{"author":{"_account_id":18002,"name":"John Fulton","email":"fulton@redhat.com","username":"fultonj"},"change_message_id":"0a1a0c74e7530c70d61aa03f81c335beed71d03f","unresolved":false,"context_lines":[{"line_number":198,"context_line":"  [stack@undercloud ~]$ cat /home/stack/hci_report.txt"},{"line_number":199,"context_line":"  Derived Parameters results"},{"line_number":200,"context_line":"   Inputs:"},{"line_number":201,"context_line":"   - Total host RAM in GB: 7"},{"line_number":202,"context_line":"   - Total host vCPUs: 8"},{"line_number":203,"context_line":"   - Ceph OSDs per host: 1"},{"line_number":204,"context_line":"   - Average guest memory size in GB: 8"}],"source_content_type":"text/x-rst","patch_set":11,"id":"9f560f44_5ee27331","line":201,"in_reply_to":"9f560f44_584c5575","updated":"2020-09-09 11:02:30.000000000","message":"OK, I\u0027ll modify it to use values from a Dell RX730 I used in the past.","commit_id":"fd2fc169903343e84af9775959f89d7f86680d00"},{"author":{"_account_id":18002,"name":"John Fulton","email":"fulton@redhat.com","username":"fultonj"},"change_message_id":"70abda404dbfdd04e8abb8895451e821a5970408","unresolved":false,"context_lines":[{"line_number":85,"context_line":"will use on average 20% of it\u0027s available CPU. The ``hci_profile``"},{"line_number":86,"context_line":"parameter should be set to a profile describing the expected workload."},{"line_number":87,"context_line":""},{"line_number":88,"context_line":"In victoria and newer the default profile sets the average memory and"},{"line_number":89,"context_line":"CPU utilization to 0 because by default the workload is unknown. When"},{"line_number":90,"context_line":"the ``tripleo_derive_hci_parameters`` Ansible module is passed these"},{"line_number":91,"context_line":"values it sets the NovaReservedHostMemory to the number of OSDs"},{"line_number":92,"context_line":"requested multiplied by 5 (for 5 GBs of RAM per OSD) but is not able"},{"line_number":93,"context_line":"to take into account the memory overhead per guest for the hypervisor."},{"line_number":94,"context_line":"It also does not set the NovaCPUAllocationRatio. Thus, passing an"},{"line_number":95,"context_line":"expected average workload will produce a more accurate set of derived"},{"line_number":96,"context_line":"HCI parameters. However, this default does allow for a simpler"},{"line_number":97,"context_line":"deployment where derived paramters may be used without having to"},{"line_number":98,"context_line":"specify a workload but the OSDs are protected from having their memory"},{"line_number":99,"context_line":"allocated to Nova guests."},{"line_number":100,"context_line":""},{"line_number":101,"context_line":"Deriving HCI paramters before a deployment"},{"line_number":102,"context_line":"------------------------------------------"}],"source_content_type":"text/x-rst","patch_set":12,"id":"9f560f44_9eefeb85","line":99,"range":{"start_line":88,"start_character":0,"end_line":99,"end_character":25},"updated":"2020-09-09 11:09:29.000000000","message":"Assuming we merge https://review.opendev.org/#/c/750474 and we use the latest change to https://review.opendev.org/#/c/714217/3..4/plan-samples/plan-environment-derived-params.yaml\n\nI added this for deployers don\u0027t know their workload. They can at least always pass the following easily:\n\n-p /usr/share/openstack-tripleo-heat-templates/plan-samples/plan-environment-derived-params.yaml \\\n\nWithout having to edit the plan env file and get the benefit of having their OSD memory protected. Ceph can slow down without as much CPU but losing memory can lead to OOM so I think this is a good trade off.","commit_id":"596a066c48abca8f8b56fe4ccd1bbfc872997377"},{"author":{"_account_id":21129,"name":"Alan Bishop","email":"abishopsweng@gmail.com","username":"ASBishop","status":"ex Red Hat"},"change_message_id":"74052b4b188201ca4158ac5be200d54ed4bbc569","unresolved":false,"context_lines":[{"line_number":68,"context_line":"    hci_profile: default"},{"line_number":69,"context_line":"    hci_profile_config:"},{"line_number":70,"context_line":"      default:"},{"line_number":71,"context_line":"        average_guest_memory_size_in_mb: 0"},{"line_number":72,"context_line":"        average_guest_cpu_utilization_percentage: 0"},{"line_number":73,"context_line":"      many_small_vms:"},{"line_number":74,"context_line":"        average_guest_memory_size_in_mb: 1024"},{"line_number":75,"context_line":"        average_guest_cpu_utilization_percentage: 20"}],"source_content_type":"text/x-rst","patch_set":13,"id":"9f560f44_6a5b1521","line":72,"range":{"start_line":71,"start_character":0,"end_line":72,"end_character":51},"updated":"2020-09-10 15:21:14.000000000","message":"This doesn\u0027t look right. I see these values for the default profile:\n\n      default:\n        average_guest_memory_size_in_mb: 2048\n        average_guest_cpu_utilization_percentage: 50\n\nAnd also this 4th profile:\n\n      nfv_default:\n        average_guest_memory_size_in_mb: 8192\n        average_guest_cpu_utilization_percentage: 90\n\n** UPDATE **\n\nOh, now I see at L88 what seems to be a new feature, and that is values of zero have special meaning. But FWIW, I don\u0027t see that in master just yet (must be other patches to THT and tripleo-ansible?).","commit_id":"ee6b4543e8d03e0b58a6d484ca751d0b7d6f6053"},{"author":{"_account_id":18002,"name":"John Fulton","email":"fulton@redhat.com","username":"fultonj"},"change_message_id":"6ead7bdc52f1cc3ff1d4ece9f9f7e50d6744546c","unresolved":false,"context_lines":[{"line_number":68,"context_line":"    hci_profile: default"},{"line_number":69,"context_line":"    hci_profile_config:"},{"line_number":70,"context_line":"      default:"},{"line_number":71,"context_line":"        average_guest_memory_size_in_mb: 0"},{"line_number":72,"context_line":"        average_guest_cpu_utilization_percentage: 0"},{"line_number":73,"context_line":"      many_small_vms:"},{"line_number":74,"context_line":"        average_guest_memory_size_in_mb: 1024"},{"line_number":75,"context_line":"        average_guest_cpu_utilization_percentage: 20"}],"source_content_type":"text/x-rst","patch_set":13,"id":"9f560f44_75d776ef","line":72,"range":{"start_line":71,"start_character":0,"end_line":72,"end_character":51},"in_reply_to":"9f560f44_6a5b1521","updated":"2020-09-10 16:17:58.000000000","message":"\u003e This doesn\u0027t look right. I see these values for the default\n \u003e profile:\n \u003e \n \u003e default:\n \u003e average_guest_memory_size_in_mb: 2048\n \u003e average_guest_cpu_utilization_percentage: 50\n \u003e \n \u003e And also this 4th profile:\n \u003e \n \u003e nfv_default:\n \u003e average_guest_memory_size_in_mb: 8192\n \u003e average_guest_cpu_utilization_percentage: 90\n \u003e \n \u003e ** UPDATE **\n \u003e \n \u003e Oh, now I see at L88 what seems to be a new feature, and that is\n \u003e values of zero have special meaning. But FWIW, I don\u0027t see that in\n \u003e master just yet (must be other patches to THT and tripleo-ansible?).\n\nYes, both. \n\nhttps://review.opendev.org/#/c/750474 for tripleo-ansible has merged \n\nWhen THT patch https://review.opendev.org/#/c/714217/3..4/plan-samples/plan-environment-derived-params.yaml merges they will be zeros by default and then we can merge this patch.\n\nI\u0027ll add the nfv profile in the next PS.","commit_id":"ee6b4543e8d03e0b58a6d484ca751d0b7d6f6053"}]}