)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":5890,"name":"Doug Goldstein","email":"cardoe@cardoe.com","username":"cardoe"},"change_message_id":"15190af84230df9ad4d8c73906f4404c315477d3","unresolved":true,"context_lines":[],"source_content_type":"","patch_set":1,"id":"b7d39fed_f827d278","updated":"2025-09-15 21:52:50.000000000","message":"Creating this as a conversation item.","commit_id":"947b89544d7c6f4b42b27e22cdc097c61a350fc5"},{"author":{"_account_id":3009,"name":"Vladimir Kozhukalov","email":"kozhukalov@gmail.com","username":"kozhukalov"},"change_message_id":"70939b7fb2461965354dbda88854521fd3eb1315","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"7c9818db_560fb86d","updated":"2025-09-29 17:19:14.000000000","message":"People also have been intersted about deploying multiple compute daemonsets. OSH now supports having multiple daemonsets with different configs by means of using daemonset overrides [1]. The approach in some sense is similar to what is suggested but does not involve subcharts. Subcharts are probably better in a sense that Helm will put values and templates in a separate release. \n\nI remember someone from Wind River (AFAIU the company behind the StarlingX) was trying to have a separate daemonset for every compute node and encountered the limitation of the Helm release object size (the values was copied N times for every node).\n\nMy personal impression about daemonset overrides is very close to kind of panic when I look at this file [1]. I barely understand what is going on here and every time it takes couple hours to delve into it when I need to fix/debug something.\n\nSo, maybe it is better to have two charts in a repo for compatibilty instead of supporting this compat template which as I understand will be much larger if it provides compatibility for more values.   \n\n[1] https://opendev.org/openstack/openstack-helm/src/branch/master/helm-toolkit/templates/utils/_daemonset_overrides.tpl","commit_id":"947b89544d7c6f4b42b27e22cdc097c61a350fc5"},{"author":{"_account_id":5890,"name":"Doug Goldstein","email":"cardoe@cardoe.com","username":"cardoe"},"change_message_id":"0bd072ed7418ba91771624780b9014e75b235ab9","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"d73f831d_88efbb88","in_reply_to":"7c9818db_560fb86d","updated":"2025-09-29 18:00:54.000000000","message":"I\u0027m happy to have a separate chart. I\u0027d also want to do the same for ironic-conductor. Today we pair up a nova-compute with an ironic-conductor group.","commit_id":"947b89544d7c6f4b42b27e22cdc097c61a350fc5"}],"helm-toolkit/templates/_openstack_compat.tpl":[{"author":{"_account_id":3009,"name":"Vladimir Kozhukalov","email":"kozhukalov@gmail.com","username":"kozhukalov"},"change_message_id":"30ce4de0065215825708f4f54119cdb633f0e795","unresolved":true,"context_lines":[{"line_number":47,"context_line":"{{- $component :\u003d .component -}}"},{"line_number":48,"context_line":"{{- $global :\u003d .global -}}"},{"line_number":49,"context_line":"{{- $legacy_key :\u003d printf \"ironic_%s\" $component -}}"},{"line_number":50,"context_line":"{{- if hasKey $global.Values $component -}}"},{"line_number":51,"context_line":"  {{- if hasKey (index $global.Values $component) \"image\" -}}"},{"line_number":52,"context_line":"    {{- if hasKey (index $global.Values $component \"image\") \"repository\" -}}"},{"line_number":53,"context_line":"      {{- index $global.Values $component \"image\" \"repository\" -}}"}],"source_content_type":"text/x-smarty","patch_set":1,"id":"4c0ccca0_4ff1aa29","line":50,"updated":"2025-09-29 16:53:36.000000000","message":"wouldn\u0027t it be better to use dig function here?","commit_id":"947b89544d7c6f4b42b27e22cdc097c61a350fc5"}]}