Merge pull request #4240 from yuvipanda/daskhub-coalesce

Coalesce daskhub into basehub
2i2c-org · Jun 28, 2024 · ceeb4bd · ceeb4bd
2 parents 4db4e32 + 736fcad
commit ceeb4bd
Show file tree

Hide file tree

Showing 39 changed files with 673 additions and 512 deletions.
diff --git a/config/clusters/2i2c-aws-us/daskhub-common.values.yaml b/config/clusters/2i2c-aws-us/daskhub-common.values.yaml
@@ -17,7 +17,15 @@ basehub:
         - noresvport
       serverIP: fs-0b70db2b65209a77d.efs.us-west-2.amazonaws.com
       baseShareName: /
+  dask-gateway:
+    enabled: true
   jupyterhub:
     scheduling:
       userScheduler:
         enabled: true
+    custom:
+      daskhubSetup:
+        enabled: true
+    singleuser:
+      cloudMetadata:
+        blockWithIptables: false
diff --git a/config/clusters/2i2c/daskhub-common.values.yaml b/config/clusters/2i2c/daskhub-common.values.yaml
@@ -9,3 +9,12 @@ basehub:
       serverIP: 10.234.45.250
       # MUST HAVE TRAILING SLASH
       baseShareName: /homes/homes/
+  dask-gateway:
+    enabled: true
+  jupyterhub:
+    custom:
+      daskhubSetup:
+        enabled: true
+    singleuser:
+      cloudMetadata:
+        blockWithIptables: false
diff --git a/config/clusters/awi-ciroh/common.values.yaml b/config/clusters/awi-ciroh/common.values.yaml
@@ -10,8 +10,24 @@ basehub:
       serverIP: 10.11.233.234
       # Name of Google Filestore share
       baseShareName: /homes/
+  dask-gateway:
+    enabled: true
+    gateway:
+      backend:
+        scheduler:
+          cores:
+            request: 0.8
+            limit: 1
+          memory:
+            request: 1G
+            limit: 2G
   jupyterhub:
+    singleuser:
+      cloudMetadata:
+        blockWithIptables: false
     custom:
+      daskhubSetup:
+        enabled: true
       2i2c:
         add_staff_user_ids_to_admin_users: true
         add_staff_user_ids_of_type: "github"
@@ -48,13 +64,3 @@ basehub:
             - arpita0911patel
             - sepehrkrz
             - benlee0423
-dask-gateway:
-  gateway:
-    backend:
-      scheduler:
-        cores:
-          request: 0.8
-          limit: 1
-        memory:
-          request: 1G
-          limit: 2G
diff --git a/config/clusters/earthscope/common.values.yaml b/config/clusters/earthscope/common.values.yaml
@@ -13,8 +13,12 @@ basehub:
         - noresvport
       serverIP: fs-08e7747330d833d82.efs.us-east-2.amazonaws.com
       baseShareName: /
+  dask-gateway:
+    enabled: true
   jupyterhub:
     custom:
+      daskhubSetup:
+        enabled: true
       2i2c:
         add_staff_user_ids_to_admin_users: true
         add_staff_user_ids_of_type: "google"
@@ -136,6 +140,8 @@ basehub:
             # Chad Trebant, https://2i2c.freshdesk.com/a/tickets/1279
             - google-oauth2|117718799995701713253
     singleuser:
+      cloudMetadata:
+        blockWithIptables: false
       profileList:
         - display_name: "Shared Small: 1-4 CPU, 8-32 GB"
           description: "A shared machine, the recommended option until you experience a limitation."

diff --git a/config/clusters/gridsst/common.values.yaml b/config/clusters/gridsst/common.values.yaml
@@ -13,8 +13,12 @@ basehub:
         - noresvport
       serverIP: fs-05f68d7e096d7cf16.efs.us-west-2.amazonaws.com
       baseShareName: /
+  dask-gateway:
+    enabled: true
   jupyterhub:
     custom:
+      daskhubSetup:
+        enabled: true
       2i2c:
         add_staff_user_ids_to_admin_users: true
         add_staff_user_ids_of_type: "github"
@@ -73,6 +77,8 @@ basehub:
             - nikki-t
             - dgumustel
     singleuser:
+      cloudMetadata:
+        blockWithIptables: false
       profileList:
         # The mem-guarantees are here so k8s doesn't schedule other pods
         # on these nodes.

diff --git a/config/clusters/hhmi/cluster.yaml b/config/clusters/hhmi/cluster.yaml
@@ -21,15 +21,15 @@ hubs:
     domain: staging.hhmi.2i2c.cloud
     helm_chart: daskhub
     helm_chart_values_files:
-      - common.values.yaml
+      - daskhub-common.values.yaml
       - staging.values.yaml
       - enc-staging.secret.values.yaml
   - name: prod
     display_name: "HHMI"
     domain: hhmi.2i2c.cloud
     helm_chart: daskhub
     helm_chart_values_files:
-      - common.values.yaml
+      - daskhub-common.values.yaml
       - prod.values.yaml
       - enc-prod.secret.values.yaml
   - name: spyglass

diff --git a/config/clusters/hhmi/common.values.yaml → .../clusters/hhmi/daskhub-common.values.yaml b/config/clusters/hhmi/common.values.yaml → .../clusters/hhmi/daskhub-common.values.yaml
@@ -8,8 +8,12 @@ basehub:
         - noatime
       serverIP: 10.55.112.74
       baseShareName: /homes/
+  dask-gateway:
+    enabled: true
   jupyterhub:
     custom:
+      daskhubSetup:
+        enabled: true
       2i2c:
         add_staff_user_ids_to_admin_users: true
         add_staff_user_ids_of_type: "github"
@@ -31,6 +35,8 @@ basehub:
             url: ""
             custom_html: <a href="https://www.hhmi.org/">HHMI</a> and <a href="https://strategiesos.org/about/">Stratos</a>
     singleuser:
+      cloudMetadata:
+        blockWithIptables: false
       extraContainers:
         - name: mysql
           image: datajoint/mysql:8.0 # following the spyglass tutorial at https://lorenfranklab.github.io/spyglass/latest/notebooks/00_Setup/#existing-database

diff --git a/config/clusters/jupyter-meets-the-earth/cluster.yaml b/config/clusters/jupyter-meets-the-earth/cluster.yaml
@@ -16,14 +16,14 @@ hubs:
     domain: staging.jmte.2i2c.cloud
     helm_chart: daskhub
     helm_chart_values_files:
-      - common.values.yaml
+      - daskhub-common.values.yaml
       - staging.values.yaml
       - enc-staging.secret.values.yaml
   - name: prod
     display_name: "Jupyter Meets the Earth"
     domain: jmte.2i2c.cloud
     helm_chart: daskhub
     helm_chart_values_files:
-      - common.values.yaml
+      - daskhub-common.values.yaml
       - prod.values.yaml
       - enc-prod.secret.values.yaml
diff --git a/...upyter-meets-the-earth/common.values.yaml → ...eets-the-earth/daskhub-common.values.yaml b/...upyter-meets-the-earth/common.values.yaml → ...eets-the-earth/daskhub-common.values.yaml
@@ -14,9 +14,28 @@ basehub:
       serverIP: fs-01707b06.efs.us-west-2.amazonaws.com
       # This is different from rest of our hubs!
       baseShareName: /
-
+  dask-gateway:
+    enabled: true
+    gateway:
+      backend:
+        scheduler:
+          # IMPORTANT: We have experienced that the scheduler can fail with
+          #            1GB memory limit. This was observed "stream closed"
+          #            from the python client working against the
+          #            Dask-Gateway created DaskCluster.
+          #
+          #            CommClosedError: in <TLS (closed) ConnectionPool.gather local=tls://192.168.40.210:54296 remote=gateway://traefik-prod-dask-gateway.prod:80/prod.b9600f678bb747c1a5f038b5bef3eb90>: Stream is closed
+          #
+          cores:
+            request: 1
+            limit: 64
+          memory:
+            request: 2G
+            limit: 500G
   jupyterhub:
     custom:
+      daskhubSetup:
+        enabled: true
       2i2c:
         add_staff_user_ids_to_admin_users: true
         add_staff_user_ids_of_type: "github"
@@ -39,6 +58,8 @@ basehub:
             url: https://jupytearth.org
 
     singleuser:
+      cloudMetadata:
+        blockWithIptables: false
       initContainers:
         # Need to explicitly set this up and copy what's in basehub/values.yaml
         # as we have an extra 'shared-public' directory here.
@@ -276,21 +297,3 @@ basehub:
     scheduling:
       userScheduler:
         enabled: true
-
-dask-gateway:
-  gateway:
-    backend:
-      scheduler:
-        # IMPORTANT: We have experienced that the scheduler can fail with
-        #            1GB memory limit. This was observed "stream closed"
-        #            from the python client working against the
-        #            Dask-Gateway created DaskCluster.
-        #
-        #            CommClosedError: in <TLS (closed) ConnectionPool.gather local=tls://192.168.40.210:54296 remote=gateway://traefik-prod-dask-gateway.prod:80/prod.b9600f678bb747c1a5f038b5bef3eb90>: Stream is closed
-        #
-        cores:
-          request: 1
-          limit: 64
-        memory:
-          request: 2G
-          limit: 500G
diff --git a/config/clusters/leap/cluster.yaml b/config/clusters/leap/cluster.yaml
@@ -21,15 +21,15 @@ hubs:
     domain: staging.leap.2i2c.cloud
     helm_chart: daskhub
     helm_chart_values_files:
-      - common.values.yaml
+      - daskhub-common.values.yaml
       - staging.values.yaml
       - enc-staging.secret.values.yaml
   - name: prod
     display_name: "LEAP Prod"
     domain: leap.2i2c.cloud
     helm_chart: daskhub
     helm_chart_values_files:
-      - common.values.yaml
+      - daskhub-common.values.yaml
       - prod.values.yaml
       - enc-prod.secret.values.yaml
   - name: public

diff --git a/config/clusters/leap/common.values.yaml → .../clusters/leap/daskhub-common.values.yaml b/config/clusters/leap/common.values.yaml → .../clusters/leap/daskhub-common.values.yaml
@@ -10,8 +10,21 @@ basehub:
       serverIP: 10.104.103.242
       # Name of Google Filestore share
       baseShareName: /homes/
+  dask-gateway:
+    enabled: true
+    gateway:
+      backend:
+        scheduler:
+          cores:
+            request: 1
+            limit: 2
+          memory:
+            request: 4G
+            limit: 4G
   jupyterhub:
     custom:
+      daskhubSetup:
+        enabled: true
       # Extra mount point for admins to access to all users' home dirs
       # Ref https://github.com/2i2c-org/infrastructure/issues/2105
       singleuserAdmin:
@@ -86,6 +99,8 @@ basehub:
             - rabernat
             - jbusecke
     singleuser:
+      cloudMetadata:
+        blockWithIptables: false
       extraEnv:
         GH_SCOPED_CREDS_CLIENT_ID: "Iv1.0c7df3d4b3191b2f"
         GH_SCOPED_CREDS_APP_URL: https://github.com/apps/leap-hub-push-access
@@ -348,14 +363,3 @@ basehub:
             mem_guarantee: 24G
             extra_resource_limits:
               nvidia.com/gpu: "1"
-
-dask-gateway:
-  gateway:
-    backend:
-      scheduler:
-        cores:
-          request: 1
-          limit: 2
-        memory:
-          request: 4G
-          limit: 4G
diff --git a/config/clusters/linked-earth/common.values.yaml b/config/clusters/linked-earth/common.values.yaml
@@ -10,8 +10,21 @@ basehub:
       serverIP: 10.155.150.2
       # Name of Google Filestore share
       baseShareName: /homes/
+  dask-gateway:
+    enabled: true
+    gateway:
+      backend:
+        scheduler:
+          cores:
+            request: 0.8
+            limit: 1
+          memory:
+            request: 1G
+            limit: 2G
   jupyterhub:
     custom:
+      daskhubSetup:
+        enabled: true
       2i2c:
         add_staff_user_ids_to_admin_users: true
         add_staff_user_ids_of_type: "github"
@@ -43,6 +56,8 @@ basehub:
           admin_users:
             - khider
     singleuser:
+      cloudMetadata:
+        blockWithIptables: false
       image:
         # User image repo: https://quay.io/repository/linkedearth/pyleoclim
         name: quay.io/linkedearth/pyleoclim
@@ -163,13 +178,3 @@ basehub:
             mem_limit: null
             node_selector:
               node.kubernetes.io/instance-type: n2-highmem-16
-dask-gateway:
-  gateway:
-    backend:
-      scheduler:
-        cores:
-          request: 0.8
-          limit: 1
-        memory:
-          request: 1G
-          limit: 2G
diff --git a/config/clusters/meom-ige/common.values.yaml b/config/clusters/meom-ige/common.values.yaml
@@ -13,8 +13,12 @@ basehub:
         - noresvport
       serverIP: nfs-server-01.us-central1-b.c.meom-ige-cnrs.internal
       baseShareName: /export/home-01/homes/
+  dask-gateway:
+    enabled: true
   jupyterhub:
     custom:
+      daskhubSetup:
+        enabled: true
       2i2c:
         add_staff_user_ids_to_admin_users: true
         add_staff_user_ids_of_type: "github"
@@ -34,6 +38,8 @@ basehub:
             name: SWOT Ocean Pangeo Team
             url: https://meom-group.github.io/
     singleuser:
+      cloudMetadata:
+        blockWithIptables: false
       extraEnv:
         DATA_BUCKET: gs://meom-ige-data
         SCRATCH_BUCKET: "gs://meom-ige-scratch/$(JUPYTERHUB_USER)"

diff --git a/config/clusters/nasa-cryo/common.values.yaml b/config/clusters/nasa-cryo/common.values.yaml
@@ -13,8 +13,12 @@ basehub:
         - noresvport
       serverIP: fs-0872256335d483d5f.efs.us-west-2.amazonaws.com
       baseShareName: /
+  dask-gateway:
+    enabled: true
   jupyterhub:
     custom:
+      daskhubSetup:
+        enabled: true
       2i2c:
         add_staff_user_ids_to_admin_users: true
         add_staff_user_ids_of_type: "github"
@@ -66,6 +70,8 @@ basehub:
             - jomey
 
     singleuser:
+      cloudMetadata:
+        blockWithIptables: false
       defaultUrl: /lab
       storage:
         extraVolumes: