)]}'
{"calico/values.yaml":[{"author":{"_account_id":26645,"name":"Luke Philips","email":"lukephilips@gmail.com","username":"lukepatrick"},"change_message_id":"356f0b694438d4c1bafabc70c9c72e1eb89b3dc4","unresolved":false,"context_lines":[{"line_number":23,"context_line":"images:"},{"line_number":24,"context_line":"  tags:"},{"line_number":25,"context_line":"    calico_etcd: quay.io/coreos/etcd:v3.1.10"},{"line_number":26,"context_line":"    calico_node: quay.io/calico/node:v2.4.1"},{"line_number":27,"context_line":"    calico_cni: quay.io/calico/cni:v1.10.0"},{"line_number":28,"context_line":"    calico_kube_policy_controller: quay.io/calico/kube-policy-controller:v0.7.0"},{"line_number":29,"context_line":"    dep_check: quay.io/stackanetes/kubernetes-entrypoint:v0.2.1"}],"source_content_type":"text/x-yaml","patch_set":10,"id":"3f4b6375_09c7b814","line":26,"range":{"start_line":26,"start_character":0,"end_line":26,"end_character":43},"updated":"2017-11-01 20:35:18.000000000","message":"not going for 2.6.x?","commit_id":"308f8a2121882aa5d8edddca19aa4d5008f85cf3"},{"author":{"_account_id":21307,"name":"Brandon B. Jozsa","email":"bjozsa@jinkit.com","username":"v1k0d3n"},"change_message_id":"e55b7b3c834681832eef3878eb7bf6fe4e5dced7","unresolved":false,"context_lines":[{"line_number":23,"context_line":"images:"},{"line_number":24,"context_line":"  tags:"},{"line_number":25,"context_line":"    calico_etcd: quay.io/coreos/etcd:v3.1.10"},{"line_number":26,"context_line":"    calico_node: quay.io/calico/node:v2.4.1"},{"line_number":27,"context_line":"    calico_cni: quay.io/calico/cni:v1.10.0"},{"line_number":28,"context_line":"    calico_kube_policy_controller: quay.io/calico/kube-policy-controller:v0.7.0"},{"line_number":29,"context_line":"    dep_check: quay.io/stackanetes/kubernetes-entrypoint:v0.2.1"}],"source_content_type":"text/x-yaml","patch_set":10,"id":"3f4b6375_6c451a15","line":26,"range":{"start_line":26,"start_character":0,"end_line":26,"end_character":43},"in_reply_to":"3f4b6375_09c7b814","updated":"2017-11-01 20:54:05.000000000","message":"it\u0027s a flexible value, which means it\u0027s easy to override via --set. there shouldn\u0027t be much religion on default values unless there\u0027s a need (bug, feature, etc). change it during instantiation if need be :)","commit_id":"308f8a2121882aa5d8edddca19aa4d5008f85cf3"}],"helm-toolkit/templates/snippets/_kubernetes_entrypoint_rbac.tpl":[{"author":{"_account_id":15518,"name":"Artur Korzeniewski","email":"artur.korzen@wp.pl","username":"akorzeni"},"change_message_id":"90fd0f7f6dcf047ff00c4bcfccfeb6b46b4c2ca4","unresolved":false,"context_lines":[{"line_number":27,"context_line":"    # Secret, ServiceAccount, Role, RoleBinding. The problem is that"},{"line_number":28,"context_line":"    # this Secret will not be created if ServiceAccount doesn\u0027t exist."},{"line_number":29,"context_line":"    # The solution is to add pre-install hook so that these are created first."},{"line_number":30,"context_line":"    helm.sh/hook: pre-install"},{"line_number":31,"context_line":"roleRef:"},{"line_number":32,"context_line":"  apiGroup: rbac.authorization.k8s.io"},{"line_number":33,"context_line":"  kind: ClusterRole"}],"source_content_type":"text/x-smarty","patch_set":10,"id":"3f4b6375_2a389bec","line":30,"updated":"2017-11-08 11:36:27.000000000","message":"Helm is not managing the pre-install resources, so when some service is deleted using helm, the cluster-role-binding will not be removed.\n\nIt is the issue when user would like to reinstall chart - it would fail due to already existing resources like cluster-role-binding.\n\nThe solution to above would be to add code that would remove the resources created during pre-install phase. Or use the post-install phase of creating the secret with service account token :)","commit_id":"308f8a2121882aa5d8edddca19aa4d5008f85cf3"},{"author":{"_account_id":15518,"name":"Artur Korzeniewski","email":"artur.korzen@wp.pl","username":"akorzeni"},"change_message_id":"b73fa58255111f085b7d08bbcdf6520c12edafa8","unresolved":false,"context_lines":[{"line_number":27,"context_line":"    # Secret, ServiceAccount, Role, RoleBinding. The problem is that"},{"line_number":28,"context_line":"    # this Secret will not be created if ServiceAccount doesn\u0027t exist."},{"line_number":29,"context_line":"    # The solution is to add pre-install hook so that these are created first."},{"line_number":30,"context_line":"    helm.sh/hook: pre-install"},{"line_number":31,"context_line":"roleRef:"},{"line_number":32,"context_line":"  apiGroup: rbac.authorization.k8s.io"},{"line_number":33,"context_line":"  kind: ClusterRole"}],"source_content_type":"text/x-smarty","patch_set":10,"id":"1f485f77_fc964df0","line":30,"in_reply_to":"1f485f77_1e6edd00","updated":"2017-11-15 14:45:33.000000000","message":"Ok, I did not notice the \u0027--wait\u0027 would not run post-install resources. \nSo the most reasonable solution would be to add clean-up job with tiller service account, and manually delete the resources created in pre-install phase.\n\nI would try the cleanup job approach.","commit_id":"308f8a2121882aa5d8edddca19aa4d5008f85cf3"},{"author":{"_account_id":23928,"name":"Pete Birley","email":"petebirley@gmail.com","username":"portdirect"},"change_message_id":"374fb7dc0f67c063b8c86d1f787eca6b20835266","unresolved":false,"context_lines":[{"line_number":27,"context_line":"    # Secret, ServiceAccount, Role, RoleBinding. The problem is that"},{"line_number":28,"context_line":"    # this Secret will not be created if ServiceAccount doesn\u0027t exist."},{"line_number":29,"context_line":"    # The solution is to add pre-install hook so that these are created first."},{"line_number":30,"context_line":"    helm.sh/hook: pre-install"},{"line_number":31,"context_line":"roleRef:"},{"line_number":32,"context_line":"  apiGroup: rbac.authorization.k8s.io"},{"line_number":33,"context_line":"  kind: ClusterRole"}],"source_content_type":"text/x-smarty","patch_set":10,"id":"1f485f77_1e6edd00","line":30,"in_reply_to":"3f4b6375_2a389bec","updated":"2017-11-14 18:17:16.000000000","message":"I\u0027ve chatted with a few folks upstream - and the recommendation is to add code that performs the cleanup until pre-install items are tracked. Otherwise, we break some of helms logic as we cannot --wait for a service to come up.","commit_id":"308f8a2121882aa5d8edddca19aa4d5008f85cf3"}]}
