deployer: update daemonset_requests.yaml

deployer/commands/generate/resource_allocation/daemonset_requests.yaml

@@ -14,17 +14,30 @@
 # - We may deploy or change a DaemonSet's requests over time (support-cryptnono,
 #   support-prometheus-node-exporter)
 #
-# Besides requests from DaemonSet resources, GKE apparently has a controller
-# creating a standalone kube-proxy pod per node, currently requesting 100m CPU
-# (no memory requests). These needs to be accounted for as well in the end, but
-# isn't tracked by in this file.
-#
 # This file isn't updated by automation, but can easily be updated by manually
 # running a command once for each cluster:
 #
 #     deployer config get-clusters | xargs -I {} deployer generate resource-allocation daemonset-requests {}
 #
 gke:
+  # Current overhead is 454m and 656Mi (May 15 2024 with EKS GKE 1.29.1).
+  #
+  # About kube-proxy:
+  #
+  #   GKE's node controller creates a standalone kube-proxy pod per node not
+  #   observed by inspecting the daemonsets' requests, currently requesting 100m
+  #   CPU (no memory requests). These needs to be accounted for as well in the
+  #   end, but isn't tracked by in this file.
+  #
+  # About daemonsets varying between GKE clusters:
+  #
+  #   - gke-metrics-agent requests 3+3m CPU and 60+40Mi memory, its not clear
+  #     why some clusters but not all have it.
+  #   - calico-node requests 100m CPU and 0Mi memory, its around if network
+  #     policy enforcement is enabled.
+  #   - ip-masq-agent requests 10m CPU and 16Mi memory, its around if network
+  #     policy enforcement is enabled.
+  #
   2i2c:
     requesting_daemon_sets: calico-node,fluentbit-gke,gke-metadata-server,gke-metrics-agent,ip-masq-agent,netd,pdcsi-node,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: binder-staging-dind,binder-staging-image-cleaner,imagebuilding-demo-binderhub-service-docker-api
@@ -43,13 +56,8 @@ gke:
     cpu_requests: 354m
     memory_requests: 656Mi
     k8s_version: v1.29.1-gke.1589018
-  callysto:
-    requesting_daemon_sets: calico-node,fluentbit-gke,gke-metadata-server,gke-metrics-agent,ip-masq-agent,netd,pdcsi-node,support-cryptnono,support-prometheus-node-exporter
-    other_daemon_sets: ""
-    cpu_requests: 344m
-    memory_requests: 596Mi
-    k8s_version: v1.27.7-gke.1056000
   catalystproject-latam:
+    # missing: gke-metrics-agent
     requesting_daemon_sets: calico-node,fluentbit-gke,gke-metadata-server,ip-masq-agent,netd,pdcsi-node,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 348m
@@ -62,11 +70,12 @@ gke:
     memory_requests: 656Mi
     k8s_version: v1.29.1-gke.1589018
   hhmi:
+    # missing: calico-node, ip-masq-agent, gke-metrics-agent
     requesting_daemon_sets: fluentbit-gke,gke-metadata-server,netd,pdcsi-node,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
-    cpu_requests: 228m
-    memory_requests: 480Mi
-    k8s_version: v1.27.10-gke.1055000
+    cpu_requests: 238m
+    memory_requests: 540Mi
+    k8s_version: v1.29.1-gke.1589020
   leap:
     requesting_daemon_sets: calico-node,fluentbit-gke,gke-metadata-server,gke-metrics-agent,ip-masq-agent,netd,pdcsi-node,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: maintenance-handler
@@ -80,6 +89,7 @@ gke:
     memory_requests: 656Mi
     k8s_version: v1.29.1-gke.1589018
   meom-ige:
+    # missing: calico-node, ip-masq-agent
     requesting_daemon_sets: fluentbit-gke,gke-metadata-server,gke-metrics-agent,netd,pdcsi-node,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 244m
@@ -92,120 +102,136 @@ gke:
     memory_requests: 656Mi
     k8s_version: v1.29.1-gke.1589018
   qcl:
+    # missing: gke-metrics-agent
     requesting_daemon_sets: calico-node,fluentbit-gke,gke-metadata-server,ip-masq-agent,netd,pdcsi-node,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 348m
     memory_requests: 556Mi
     k8s_version: v1.29.1-gke.1589018
 eks:
+  # Current overhead is 195m and 250Mi (May 15 2024 with EKS 1.29.4).
+  #
+  # About daemonsets varying between GKE clusters:
+  #
+  # - aws-node requests 25m or 50m depending on if it has the extra container
+  #   running aws-network-policy-agent or not.
+  #
+  #   ref: https://github.com/2i2c-org/infrastructure/issues/1794
+  #
   2i2c-aws-us:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
   bican:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 195m
     memory_requests: 250Mi
-    k8s_version: v1.29.1-eks-b9c9ed7
+    k8s_version: v1.29.3-eks-adc7111
   catalystproject-africa:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.3-eks-adc7111
   dandi:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 195m
     memory_requests: 250Mi
-    k8s_version: v1.29.1-eks-b9c9ed7
+    k8s_version: v1.29.3-eks-adc7111
   earthscope:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
-    cpu_requests: 195m
+    cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.28.6-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
   gridsst:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
   jupyter-health:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 195m
     memory_requests: 250Mi
-    k8s_version: v1.29.1-eks-b9c9ed7
+    k8s_version: v1.29.3-eks-adc7111
   jupyter-meets-the-earth:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
+  kitware:
+    requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
+    other_daemon_sets: ""
+    cpu_requests: 195m
+    memory_requests: 250Mi
+    k8s_version: v1.29.3-eks-adc7111
   linc:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 195m
     memory_requests: 250Mi
-    k8s_version: v1.29.1-eks-b9c9ed7
+    k8s_version: v1.29.3-eks-adc7111
   nasa-cryo:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
   nasa-esdis:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
   nasa-ghg:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
   nasa-veda:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
   openscapes:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
   opensci:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
-    other_daemon_sets: ""
-    cpu_requests: 195m
+    other_daemon_sets: sciencecore-binderhub-service-docker-api
+    cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.28.6-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
   smithsonian:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
   ubc-eoas:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
   victor:
     requesting_daemon_sets: aws-node,ebs-csi-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter
     other_daemon_sets: ""
     cpu_requests: 170m
     memory_requests: 250Mi
-    k8s_version: v1.27.10-eks-508b6b3
+    k8s_version: v1.29.4-eks-036c24b
 aks:
   utoronto:
     requesting_daemon_sets: cloud-node-manager,csi-azuredisk-node,csi-azurefile-node,kube-proxy,support-cryptnono,support-prometheus-node-exporter