Add deployed/error status output to the state file. (#719)

* More status info
* Up default resource limits.
* Need ps

.gitea/workflows/test-k8s-deploy.yml

@@ -6,6 +6,8 @@ on:
     paths:
       - '!**'
       - '.gitea/workflows/triggers/test-k8s-deploy'
+      - '.gitea/workflows/test-k8s-deploy.yml'
+      - 'tests/k8s-deploy/run-deploy-test.sh'
 
 jobs:
   test:
@@ -34,9 +36,15 @@ jobs:
         run: ./scripts/create_build_tag_file.sh
       - name: "Build local shiv package"
         run: ./scripts/build_shiv_package.sh
+      - name: "Check cgroups version"
+        run: mount | grep cgroup
       - name: "Install kind"
         run: ./tests/scripts/install-kind.sh
       - name: "Install Kubectl"
         run: ./tests/scripts/install-kubectl.sh
       - name: "Run k8s deployment test"
-        run: ./tests/k8s-deploy/run-deploy-test.sh
+        run: |
+          source /opt/bash-utils/cgroup-helper.sh
+          join_cgroup
+          ./tests/k8s-deploy/run-deploy-test.sh
+

stack_orchestrator/data/container-build/cerc-nextjs-base/Dockerfile

@@ -30,7 +30,7 @@ RUN \
 
 # [Optional] Uncomment this section to install additional OS packages.
 RUN apt-get update && export DEBIAN_FRONTEND=noninteractive \
-    && apt-get -y install --no-install-recommends jq gettext-base
+    && apt-get -y install --no-install-recommends jq gettext-base procps
 
 # [Optional] Uncomment if you want to install more global node modules
 # RUN su node -c "npm install -g <your-package-list-here>"

stack_orchestrator/deploy/k8s/cluster_info.py

@@ -195,7 +195,7 @@ class ClusterInfo:
                     volume_mounts=volume_mounts,
                     resources=client.V1ResourceRequirements(
                         requests={"cpu": "100m", "memory": "200Mi"},
-                        limits={"cpu": "500m", "memory": "500Mi"},
+                        limits={"cpu": "1000m", "memory": "2000Mi"},
                     ),
                 )
                 containers.append(container)

stack_orchestrator/deploy/webapp/deploy_webapp_from_registry.py

@@ -59,8 +59,8 @@ def process_app_deployment_request(
     dns_record = laconic.get_record(dns_crn)
     if dns_record:
         matched_owner = match_owner(app_deployment_request, dns_record)
-        if not matched_owner and dns_record.request:
-            matched_owner = match_owner(app_deployment_request, laconic.get_record(dns_record.request, require=True))
+        if not matched_owner and dns_record.attributes.request:
+            matched_owner = match_owner(app_deployment_request, laconic.get_record(dns_record.attributes.request, require=True))
 
         if matched_owner:
             print("Matched DnsRecord ownership:", matched_owner)
@@ -136,13 +136,17 @@ def load_known_requests(filename):
     return {}
 
 
-def dump_known_requests(filename, requests):
+def dump_known_requests(filename, requests, status="SEEN"):
     if not filename:
         return
     known_requests = load_known_requests(filename)
     for r in requests:
-        known_requests[r.id] = r.createTime
-    json.dump(known_requests, open(filename, "w"))
+        known_requests[r.id] = {
+            "createTime": r.createTime,
+            "status": status
+        }
+    with open(filename, "w") as f:
+        json.dump(known_requests, f)
 
 
 @click.command()
@@ -201,6 +205,7 @@ def command(ctx, kube_config, laconic_config, image_registry, deployment_parent_
     requests.reverse()
     requests_by_name = {}
     for r in requests:
+        # TODO: Do this _after_ filtering deployments and cancellations to minimize round trips.
         app = laconic.get_record(r.attributes.application)
         if not app:
             print("Skipping request %s, cannot locate app." % r.id)
@@ -256,6 +261,8 @@ def command(ctx, kube_config, laconic_config, image_registry, deployment_parent_
 
     if not dry_run:
         for r in requests_to_execute:
+            dump_known_requests(state_file, [r], "DEPLOYING")
+            status = "ERROR"
             try:
                 process_app_deployment_request(
                     ctx,
@@ -268,5 +275,6 @@ def command(ctx, kube_config, laconic_config, image_registry, deployment_parent_
                     kube_config,
                     image_registry
                 )
+                status = "DEPLOYED"
             finally:
-                dump_known_requests(state_file, [r])
+                dump_known_requests(state_file, [r], status)

stack_orchestrator/deploy/webapp/undeploy_webapp_from_registry.py

@@ -40,8 +40,8 @@ def process_app_removal_request(ctx,
     matched_owner = match_owner(app_removal_request, deployment_record, dns_record)
 
     # Or of the original deployment request.
-    if not matched_owner and deployment_record.request:
-        matched_owner = match_owner(app_removal_request, laconic.get_record(deployment_record.request, require=True))
+    if not matched_owner and deployment_record.attributes.request:
+        matched_owner = match_owner(app_removal_request, laconic.get_record(deployment_record.attributes.request, require=True))
 
     if matched_owner:
         print("Matched deployment ownership:", matched_owner)

tests/k8s-deploy/run-deploy-test.sh

@@ -9,7 +9,7 @@ fi
 
 # Helper functions: TODO move into a separate file
 wait_for_pods_started () {
-    for i in {1..5}
+    for i in {1..50}
     do
         local ps_output=$( $TEST_TARGET_SO deployment --dir $test_deployment_dir ps )
 
@@ -27,7 +27,7 @@ wait_for_pods_started () {
 }
 
 wait_for_log_output () {
-    for i in {1..5}
+    for i in {1..50}
     do
 
         local log_output=$( $TEST_TARGET_SO deployment --dir $test_deployment_dir logs )
@@ -120,7 +120,8 @@ fi
 # Stop then start again and check the volume was preserved
 $TEST_TARGET_SO deployment --dir $test_deployment_dir stop
 # Sleep a bit just in case
-sleep 2
+# sleep for longer to check if that's why the subsequent create cluster fails
+sleep 20
 $TEST_TARGET_SO deployment --dir $test_deployment_dir start
 wait_for_pods_started
 wait_for_log_output