Add schedule for k8s deploy test

* Minor fixes for deploying with k8s and podman.

* ConfigMap support

.gitea/workflows/fixturenet-eth-plugeth-test.yml

@@ -6,6 +6,8 @@ on:
     paths:
       - '!**'
       - '.gitea/workflows/triggers/fixturenet-eth-plugeth-test'
+  schedule: # Note: coordinate with other tests to not overload runners at the same time of day
+    - cron: '2 14 * * *'
 
 # Needed until we can incorporate docker startup into the executor container
 env:

.gitea/workflows/fixturenet-laconicd-test.yml

@@ -6,11 +6,8 @@ on:
     paths:
       - '!**'
       - '.gitea/workflows/triggers/fixturenet-laconicd-test'
-
-# Needed until we can incorporate docker startup into the executor container
-env:
-  DOCKER_HOST: unix:///var/run/dind.sock
-
+  schedule:
+    - cron: '1 13 * * *'
 
 jobs:
   test:
@@ -47,9 +44,5 @@ jobs:
         run: ./scripts/create_build_tag_file.sh
       - name: "Build local shiv package"
         run: ./scripts/build_shiv_package.sh
-      - name: Start dockerd # Also needed until we can incorporate into the executor
-        run: |
-          dockerd -H $DOCKER_HOST --userland-proxy=false &
-          sleep 5
       - name: "Run fixturenet-laconicd tests"
         run: ./tests/fixturenet-laconicd/run-test.sh

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

@@ -1,23 +1,20 @@
 name: K8s Deploy Test
 
 on:
-  pull_request:
-    branches: '*'
   push:
-    branches:
-      - main
-      - ci-test
-    paths-ignore:
-      - '.gitea/workflows/triggers/*'
-
-# Needed until we can incorporate docker startup into the executor container
-env:
-  DOCKER_HOST: unix:///var/run/dind.sock
+    branches: '*'
+    paths:
+      - '!**'
+      - '.gitea/workflows/triggers/test-k8s-deploy'
+      - '.gitea/workflows/test-k8s-deploy.yml'
+      - 'tests/k8s-deploy/run-deploy-test.sh'
+  schedule: # Note: coordinate with other tests to not overload runners at the same time of day
+    - cron: '3 15 * * *'
 
 jobs:
   test:
-    name: "Run deploy test suite"
-    runs-on: ubuntu-latest
+    name: "Run deploy test suite on kind/k8s"
+    runs-on: ubuntu-22.04
     steps:
       - name: "Clone project repository"
         uses: actions/checkout@v3
@@ -41,15 +38,15 @@ jobs:
         run: ./scripts/create_build_tag_file.sh
       - name: "Build local shiv package"
         run: ./scripts/build_shiv_package.sh
-      - name: Start dockerd # Also needed until we can incorporate into the executor
+      - 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: |
-          dockerd -H $DOCKER_HOST --userland-proxy=false &
-          sleep 5
-      - name: "Install Go"
-        uses: actions/setup-go@v4
-        with:
-          go-version: '1.21'
-      - name: "Install Kind"
-        run: go install sigs.k8s.io/kind@v0.20.0
-      - name: "Debug Kind"
-        run: kind create cluster --retain && docker logs kind-control-plane
+          source /opt/bash-utils/cgroup-helper.sh
+          join_cgroup
+          ./tests/k8s-deploy/run-deploy-test.sh
+

.gitea/workflows/triggers/fixturenet-eth-plugeth-test

@@ -1,2 +1,3 @@
 Change this file to trigger running the fixturenet-eth-plugeth-test CI job
 trigger
+trigger

.gitea/workflows/triggers/fixturenet-laconicd-test

@@ -1,2 +1,3 @@
 Change this file to trigger running the fixturenet-laconicd-test CI job
-
+Trigger
+Trigger

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

@@ -0,0 +1 @@
+Change this file to trigger running the test-k8s-deploy CI job

stack_orchestrator/build/build_containers.py

@@ -33,6 +33,7 @@ from stack_orchestrator.base import get_npm_registry_url
 # TODO: find a place for this
 #    epilog="Config provided either in .env or settings.ini or env vars: CERC_REPO_BASE_DIR (defaults to ~/cerc)"
 
+
 def make_container_build_env(dev_root_path: str,
                              container_build_dir: str,
                              debug: bool,
@@ -104,6 +105,9 @@ def process_container(stack: str,
         build_command = os.path.join(container_build_dir,
                                      "default-build.sh") + f" {default_container_tag} {repo_dir_or_build_dir}"
     if not dry_run:
+        # No PATH at all causes failures with podman.
+        if "PATH" not in container_build_env:
+            container_build_env["PATH"] = os.environ["PATH"]
         if verbose:
             print(f"Executing: {build_command} with environment: {container_build_env}")
         build_result = subprocess.run(build_command, shell=True, env=container_build_env)
@@ -119,6 +123,7 @@ def process_container(stack: str,
     else:
         print("Skipped")
 
+
 @click.command()
 @click.option('--include', help="only build these containers")
 @click.option('--exclude', help="don\'t build these containers")

stack_orchestrator/data/compose/docker-compose-grafana.yml

@@ -2,8 +2,10 @@ version: "3.7"
 
 services:
   grafana:
-    image: grafana/grafana
+    image: grafana/grafana:10.2.2
     restart: always
+    environment:
+      GF_SERVER_ROOT_URL: ${GF_SERVER_ROOT_URL}
     volumes:
       - ../config/monitoring/grafana/provisioning:/etc/grafana/provisioning
       - ../config/monitoring/grafana/dashboards:/etc/grafana/dashboards

stack_orchestrator/data/compose/docker-compose-prom-server.yml

@@ -2,7 +2,7 @@ version: "3.7"
 
 services:
   prometheus:
-    image: prom/prometheus
+    image: prom/prometheus:v2.49.1
     restart: always
     volumes:
       - ../config/monitoring/prometheus:/etc/prometheus

stack_orchestrator/data/config/monitoring/watcher-alert-rules.yml

@@ -95,14 +95,10 @@ groups:
               maxDataPoints: 43200
               refId: condition
               type: math
-        dashboardUid: cb9b746a-9abc-482e-9214-5231e0dd75ca
-        panelId: 24
         noDataState: Alerting
         execErrState: Alerting
         for: 15m
         annotations:
-          __dashboardUid__: cb9b746a-9abc-482e-9214-5231e0dd75ca
-          __panelId__: "24"
           summary: Watcher {{ index $labels "instance" }} of group {{ index $labels "job" }} is falling behind external head by {{ index $values "diff" }}
         isPaused: false
       - uid: censures_diff_external
@@ -191,14 +187,10 @@ groups:
               maxDataPoints: 43200
               refId: condition
               type: math
-        dashboardUid: cb9b746a-9abc-482e-9214-5231e0dd75ca
-        panelId: 24
         noDataState: Alerting
         execErrState: Alerting
         for: 15m
         annotations:
-          __dashboardUid__: cb9b746a-9abc-482e-9214-5231e0dd75ca
-          __panelId__: "24"
           summary: Watcher {{ index $labels "instance" }} of group {{ index $labels "job" }} is falling behind external head by {{ index $values "diff" }}
         isPaused: false
       - uid: claims_diff_external
@@ -287,14 +279,10 @@ groups:
               maxDataPoints: 43200
               refId: condition
               type: math
-        dashboardUid: cb9b746a-9abc-482e-9214-5231e0dd75ca
-        panelId: 24
         noDataState: Alerting
         execErrState: Alerting
         for: 15m
         annotations:
-          __dashboardUid__: cb9b746a-9abc-482e-9214-5231e0dd75ca
-          __panelId__: "24"
           summary: Watcher {{ index $labels "instance" }} of group {{ index $labels "job" }} is falling behind external head by {{ index $values "diff" }}
         isPaused: false
       - uid: conditional_star_release_diff_external
@@ -383,14 +371,10 @@ groups:
               maxDataPoints: 43200
               refId: condition
               type: math
-        dashboardUid: cb9b746a-9abc-482e-9214-5231e0dd75ca
-        panelId: 24
         noDataState: Alerting
         execErrState: Alerting
         for: 15m
         annotations:
-          __dashboardUid__: cb9b746a-9abc-482e-9214-5231e0dd75ca
-          __panelId__: "24"
           summary: Watcher {{ index $labels "instance" }} of group {{ index $labels "job" }} is falling behind external head by {{ index $values "diff" }}
         isPaused: false
       - uid: delegated_sending_diff_external
@@ -479,14 +463,10 @@ groups:
               maxDataPoints: 43200
               refId: condition
               type: math
-        dashboardUid: cb9b746a-9abc-482e-9214-5231e0dd75ca
-        panelId: 24
         noDataState: Alerting
         execErrState: Alerting
         for: 15m
         annotations:
-          __dashboardUid__: cb9b746a-9abc-482e-9214-5231e0dd75ca
-          __panelId__: "24"
           summary: Watcher {{ index $labels "instance" }} of group {{ index $labels "job" }} is falling behind external head by {{ index $values "diff" }}
         isPaused: false
       - uid: ecliptic_diff_external
@@ -575,14 +555,10 @@ groups:
               maxDataPoints: 43200
               refId: condition
               type: math
-        dashboardUid: cb9b746a-9abc-482e-9214-5231e0dd75ca
-        panelId: 24
         noDataState: Alerting
         execErrState: Alerting
         for: 15m
         annotations:
-          __dashboardUid__: cb9b746a-9abc-482e-9214-5231e0dd75ca
-          __panelId__: "24"
           summary: Watcher {{ index $labels "instance" }} of group {{ index $labels "job" }} is falling behind external head by {{ index $values "diff" }}
         isPaused: false
       - uid: linear_star_release_diff_external
@@ -671,14 +647,10 @@ groups:
               maxDataPoints: 43200
               refId: condition
               type: math
-        dashboardUid: cb9b746a-9abc-482e-9214-5231e0dd75ca
-        panelId: 24
         noDataState: Alerting
         execErrState: Alerting
         for: 15m
         annotations:
-          __dashboardUid__: cb9b746a-9abc-482e-9214-5231e0dd75ca
-          __panelId__: "24"
           summary: Watcher {{ index $labels "instance" }} of group {{ index $labels "job" }} is falling behind external head by {{ index $values "diff" }}
         isPaused: false
       - uid: polls_diff_external
@@ -767,14 +739,10 @@ groups:
               maxDataPoints: 43200
               refId: condition
               type: math
-        dashboardUid: cb9b746a-9abc-482e-9214-5231e0dd75ca
-        panelId: 24
         noDataState: Alerting
         execErrState: Alerting
         for: 15m
         annotations:
-          __dashboardUid__: cb9b746a-9abc-482e-9214-5231e0dd75ca
-          __panelId__: "24"
           summary: Watcher {{ index $labels "instance" }} of group {{ index $labels "job" }} is falling behind external head by {{ index $values "diff" }}
         isPaused: false
 
@@ -865,14 +833,10 @@ groups:
               maxDataPoints: 43200
               refId: condition
               type: math
-        dashboardUid: cb9b746a-9abc-482e-9214-5231e0dd75ca
-        panelId: 24
         noDataState: Alerting
         execErrState: Alerting
         for: 15m
         annotations:
-          __dashboardUid__: cb9b746a-9abc-482e-9214-5231e0dd75ca
-          __panelId__: "24"
           summary: Watcher {{ index $labels "instance" }} of group {{ index $labels "job" }} is falling behind external head by {{ index $values "diff" }}
         isPaused: false
       - uid: merkl_sushiswap_diff_external
@@ -961,13 +925,9 @@ groups:
               maxDataPoints: 43200
               refId: condition
               type: math
-        dashboardUid: cb9b746a-9abc-482e-9214-5231e0dd75ca
-        panelId: 24
         noDataState: Alerting
         execErrState: Alerting
         for: 15m
         annotations:
-          __dashboardUid__: cb9b746a-9abc-482e-9214-5231e0dd75ca
-          __panelId__: "24"
           summary: Watcher {{ index $labels "instance" }} of group {{ index $labels "job" }} is falling behind external head by {{ index $values "diff" }}
         isPaused: false

stack_orchestrator/data/container-build/cerc-fixturenet-eth-genesis/Dockerfile

@@ -10,9 +10,10 @@ COPY genesis /opt/genesis
 COPY --from=ethgen /usr/local/bin/eth2-testnet-genesis /usr/local/bin/
 COPY --from=ethgen /usr/local/bin/eth2-val-tools /usr/local/bin/
 COPY --from=ethgen /apps /apps
-RUN cd /apps/el-gen && pip3 install -r requirements.txt
+RUN cd /apps/el-gen && pip3 install --break-system-packages -r requirements.txt
 # web3==5.24.0 used by el-gen is broken on python 3.11
-RUN pip3 install --upgrade "web3==6.5.0"
+RUN pip3 install --break-system-packages --upgrade "web3==6.5.0"
+RUN pip3 install --break-system-packages --upgrade "typing-extensions"
 
 # Build genesis config
 RUN apk add --no-cache make bash envsubst jq

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/data/container-build/cerc-nextjs-base/scripts/build-app.sh

@@ -21,6 +21,11 @@ WORK_DIR="${1:-/app}"
 
 cd "${WORK_DIR}" || exit 1
 
+# If this file doesn't exist at all, we'll get errors below.
+if [ ! -f "next.config.js" ]; then
+  touch next.config.js
+fi
+
 if [ ! -f "next.config.dist" ]; then
   cp next.config.js next.config.dist
 fi

stack_orchestrator/data/stacks/monitoring/README.md

@@ -140,6 +140,10 @@ Set the following env variables in the deployment env config file (`monitoring-d
   # External ETH RPC endpoint (filecoin)
   # (Optional, default: https://api.node.glif.io/rpc/v1)
   CERC_FIL_RPC_ENDPOINT=
+
+  # Grafana server host URL (used in various links in alerts, etc.)
+  # (Optional, default: http://localhost:3000)
+  GF_SERVER_ROOT_URL=
   ```
 
 ## Start the stack

stack_orchestrator/data/stacks/monitoring/monitoring-watchers.md

@@ -130,6 +130,10 @@ Set the following env variables in the deployment env config file (`monitoring-w
   ```bash
   # Infura key to be used
   CERC_INFURA_KEY=
+
+  # Grafana server host URL to be used
+  # (Optional, default: http://localhost:3000)
+  GF_SERVER_ROOT_URL=
   ```
 
 ## Start the stack

stack_orchestrator/deploy/deployment_create.py

@@ -54,19 +54,44 @@ def _get_ports(stack):
 
 def _get_named_volumes(stack):
     # Parse the compose files looking for named volumes
-    named_volumes = []
+    named_volumes = {
+        "rw": [],
+        "ro": []
+    }
     parsed_stack = get_parsed_stack_config(stack)
     pods = get_pod_list(parsed_stack)
     yaml = get_yaml()
+
+    def find_vol_usage(parsed_pod_file, vol):
+        ret = {}
+        if "services" in parsed_pod_file:
+            for svc_name, svc in parsed_pod_file["services"].items():
+                if "volumes" in svc:
+                    for svc_volume in svc["volumes"]:
+                        parts = svc_volume.split(":")
+                        if parts[0] == vol:
+                            ret[svc_name] = {
+                                "volume": parts[0],
+                                "mount": parts[1],
+                                "options": parts[2] if len(parts) == 3 else None
+                            }
+        return ret
+
     for pod in pods:
         pod_file_path = get_pod_file_path(parsed_stack, pod)
         parsed_pod_file = yaml.load(open(pod_file_path, "r"))
         if "volumes" in parsed_pod_file:
             volumes = parsed_pod_file["volumes"]
             for volume in volumes.keys():
-                # Volume definition looks like:
-                # 'laconicd-data': None
-                named_volumes.append(volume)
+                for vu in find_vol_usage(parsed_pod_file, volume).values():
+                    read_only = vu["options"] == "ro"
+                    if read_only:
+                        if vu["volume"] not in named_volumes["rw"] and vu["volume"] not in named_volumes["ro"]:
+                            named_volumes["ro"].append(vu["volume"])
+                    else:
+                        if vu["volume"] not in named_volumes["rw"]:
+                            named_volumes["rw"].append(vu["volume"])
+
     return named_volumes
 
 
@@ -98,12 +123,24 @@ def _fixup_pod_file(pod, spec, compose_dir):
                     _create_bind_dir_if_relative(volume, volume_spec, compose_dir)
                     new_volume_spec = {"driver": "local",
                                        "driver_opts": {
-                                          "type": "none",
-                                          "device": volume_spec_fixedup,
-                                          "o": "bind"
+                                           "type": "none",
+                                           "device": volume_spec_fixedup,
+                                           "o": "bind"
                                        }
                                        }
                     pod["volumes"][volume] = new_volume_spec
+
+    # Fix up configmaps
+    if "configmaps" in spec:
+        spec_cfgmaps = spec["configmaps"]
+        if "volumes" in pod:
+            pod_volumes = pod["volumes"]
+            for volume in pod_volumes.keys():
+                if volume in spec_cfgmaps:
+                    volume_cfg = spec_cfgmaps[volume]
+                    # Just make the dir (if necessary)
+                    _create_bind_dir_if_relative(volume, volume_cfg, compose_dir)
+
     # Fix up ports
     if "network" in spec and "ports" in spec["network"]:
         spec_ports = spec["network"]["ports"]
@@ -319,9 +356,19 @@ def init_operation(deploy_command_context, stack, deployer_type, config,
     named_volumes = _get_named_volumes(stack)
     if named_volumes:
         volume_descriptors = {}
-        for named_volume in named_volumes:
+        configmap_descriptors = {}
+        for named_volume in named_volumes["rw"]:
             volume_descriptors[named_volume] = f"./data/{named_volume}"
-        spec_file_content["volumes"] = volume_descriptors
+        for named_volume in named_volumes["ro"]:
+            if "k8s" in deployer_type:
+                if "config" in named_volume:
+                    configmap_descriptors[named_volume] = f"./data/{named_volume}"
+                else:
+                    volume_descriptors[named_volume] = f"./data/{named_volume}"
+        if volume_descriptors:
+            spec_file_content["volumes"] = volume_descriptors
+        if configmap_descriptors:
+            spec_file_content["configmaps"] = configmap_descriptors
 
     if opts.o.debug:
         print(f"Creating spec file for stack: {stack} with content: {spec_file_content}")

stack_orchestrator/deploy/images.py

@@ -31,7 +31,8 @@ def _image_needs_pushed(image: str):
 
 def remote_tag_for_image(image: str, remote_repo_url: str):
     # Turns image tags of the form: foo/bar:local into remote.repo/org/bar:deploy
-    (org, image_name_with_version) = image.split("/")
+    major_parts = image.split("/", 2)
+    image_name_with_version = major_parts[1] if 2 == len(major_parts) else major_parts[0]
     (image_name, image_version) = image_name_with_version.split(":")
     if image_version == "local":
         return f"{remote_repo_url}/{image_name}:deploy"

stack_orchestrator/deploy/k8s/cluster_info.py

@@ -13,6 +13,8 @@
 # You should have received a copy of the GNU Affero General Public License
 # along with this program.  If not, see <http:#www.gnu.org/licenses/>.
 
+import os
+
 from kubernetes import client
 from typing import Any, List, Set
 
@@ -112,9 +114,10 @@ class ClusterInfo:
             services = pod["services"]
             for service_name in services:
                 service_info = services[service_name]
-                port = int(service_info["ports"][0])
-                if opts.o.debug:
-                    print(f"service port: {port}")
+                if "ports" in service_info:
+                    port = int(service_info["ports"][0])
+                    if opts.o.debug:
+                        print(f"service port: {port}")
         service = client.V1Service(
             metadata=client.V1ObjectMeta(name=f"{self.app_name}-service"),
             spec=client.V1ServiceSpec(
@@ -130,30 +133,70 @@ class ClusterInfo:
 
     def get_pvcs(self):
         result = []
-        volumes = named_volumes_from_pod_files(self.parsed_pod_yaml_map)
+        spec_volumes = self.spec.get_volumes()
+        named_volumes = named_volumes_from_pod_files(self.parsed_pod_yaml_map)
         if opts.o.debug:
-            print(f"Volumes: {volumes}")
-        for volume_name in volumes:
+            print(f"Spec Volumes: {spec_volumes}")
+            print(f"Named Volumes: {named_volumes}")
+        for volume_name in spec_volumes:
+            if volume_name not in named_volumes:
+                if opts.o.debug:
+                    print(f"{volume_name} not in pod files")
+                continue
             spec = client.V1PersistentVolumeClaimSpec(
                 access_modes=["ReadWriteOnce"],
                 storage_class_name="manual",
                 resources=client.V1ResourceRequirements(
                     requests={"storage": "2Gi"}
                 ),
-                volume_name=volume_name
+                volume_name=f"{self.app_name}-{volume_name}"
             )
             pvc = client.V1PersistentVolumeClaim(
-                metadata=client.V1ObjectMeta(name=volume_name,
-                                             labels={"volume-label": volume_name}),
+                metadata=client.V1ObjectMeta(name=f"{self.app_name}-{volume_name}",
+                                             labels={"volume-label": f"{self.app_name}-{volume_name}"}),
                 spec=spec,
             )
             result.append(pvc)
         return result
 
+    def get_configmaps(self):
+        result = []
+        spec_configmaps = self.spec.get_configmaps()
+        named_volumes = named_volumes_from_pod_files(self.parsed_pod_yaml_map)
+        for cfg_map_name, cfg_map_path in spec_configmaps.items():
+            if cfg_map_name not in named_volumes:
+                if opts.o.debug:
+                    print(f"{cfg_map_name} not in pod files")
+                continue
+
+            if not cfg_map_path.startswith("/"):
+                cfg_map_path = os.path.join(os.path.dirname(self.spec.file_path), cfg_map_path)
+
+            # Read in all the files at a single-level of the directory.  This mimics the behavior
+            # of `kubectl create configmap foo --from-file=/path/to/dir`
+            data = {}
+            for f in os.listdir(cfg_map_path):
+                full_path = os.path.join(cfg_map_path, f)
+                if os.path.isfile(full_path):
+                    data[f] = open(full_path, 'rt').read()
+
+            spec = client.V1ConfigMap(
+                metadata=client.V1ObjectMeta(name=f"{self.app_name}-{cfg_map_name}",
+                                             labels={"configmap-label": cfg_map_name}),
+                data=data
+            )
+            result.append(spec)
+        return result
+
     def get_pvs(self):
         result = []
-        volumes = named_volumes_from_pod_files(self.parsed_pod_yaml_map)
-        for volume_name in volumes:
+        spec_volumes = self.spec.get_volumes()
+        named_volumes = named_volumes_from_pod_files(self.parsed_pod_yaml_map)
+        for volume_name in spec_volumes:
+            if volume_name not in named_volumes:
+                if opts.o.debug:
+                    print(f"{volume_name} not in pod files")
+                continue
             spec = client.V1PersistentVolumeSpec(
                 storage_class_name="manual",
                 access_modes=["ReadWriteOnce"],
@@ -161,15 +204,15 @@ class ClusterInfo:
                 host_path=client.V1HostPathVolumeSource(path=get_node_pv_mount_path(volume_name))
             )
             pv = client.V1PersistentVolume(
-                metadata=client.V1ObjectMeta(name=volume_name,
-                                             labels={"volume-label": volume_name}),
+                metadata=client.V1ObjectMeta(name=f"{self.app_name}-{volume_name}",
+                                             labels={"volume-label": f"{self.app_name}-{volume_name}"}),
                 spec=spec,
             )
             result.append(pv)
         return result
 
-    # to suit the deployment, and also annotate the container specs to point at said volumes
-    def get_deployment(self):
+    # TODO: put things like image pull policy into an object-scope struct
+    def get_deployment(self, image_pull_policy: str = None):
         containers = []
         for pod_name in self.parsed_pod_yaml_map:
             pod = self.parsed_pod_yaml_map[pod_name]
@@ -178,10 +221,11 @@ class ClusterInfo:
                 container_name = service_name
                 service_info = services[service_name]
                 image = service_info["image"]
-                port = int(service_info["ports"][0])
-                if opts.o.debug:
-                    print(f"image: {image}")
-                    print(f"service port: {port}")
+                if "ports" in service_info:
+                    port = int(service_info["ports"][0])
+                    if opts.o.debug:
+                        print(f"image: {image}")
+                        print(f"service port: {port}")
                 # Re-write the image tag for remote deployment
                 image_to_use = remote_tag_for_image(
                     image, self.spec.get_image_registry()) if self.spec.get_image_registry() is not None else image
@@ -189,17 +233,17 @@ class ClusterInfo:
                 container = client.V1Container(
                     name=container_name,
                     image=image_to_use,
-                    image_pull_policy="Always",
+                    image_pull_policy=image_pull_policy,
                     env=envs_from_environment_variables_map(self.environment_variables.map),
                     ports=[client.V1ContainerPort(container_port=port)],
                     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)
-        volumes = volumes_for_pod_files(self.parsed_pod_yaml_map)
+        volumes = volumes_for_pod_files(self.parsed_pod_yaml_map, self.spec)
         image_pull_secrets = [client.V1LocalObjectReference(name="laconic-registry")]
         template = client.V1PodTemplateSpec(
             metadata=client.V1ObjectMeta(labels={"app": self.app_name}),

stack_orchestrator/deploy/k8s/deploy_k8s.py

@@ -1,5 +1,4 @@
 # Copyright © 2023 Vulcanize
-
 # This program is free software: you can redistribute it and/or modify
 # it under the terms of the GNU Affero General Public License as published by
 # the Free Software Foundation, either version 3 of the License, or
@@ -110,8 +109,22 @@ class K8sDeployer(Deployer):
             if opts.o.debug:
                 print("PVCs created:")
                 print(f"{pvc_resp}")
+
+        # Figure out the ConfigMaps for this deployment
+        config_maps = self.cluster_info.get_configmaps()
+        for cfg_map in config_maps:
+            if opts.o.debug:
+                print(f"Sending this ConfigMap: {cfg_map}")
+            cfg_rsp = self.core_api.create_namespaced_config_map(
+                body=cfg_map,
+                namespace=self.k8s_namespace
+            )
+            if opts.o.debug:
+                print("ConfigMap created:")
+                print(f"{cfg_rsp}")
+
         # Process compose files into a Deployment
-        deployment = self.cluster_info.get_deployment()
+        deployment = self.cluster_info.get_deployment(image_pull_policy=None if self.is_kind() else "Always")
         # Create the k8s objects
         if opts.o.debug:
             print(f"Sending this deployment: {deployment}")
@@ -132,20 +145,24 @@ class K8sDeployer(Deployer):
             print("Service created:")
             print(f"{service_resp}")
 
-        # TODO: disable ingress for kind
-        ingress: client.V1Ingress = self.cluster_info.get_ingress()
+        if not self.is_kind():
+            ingress: client.V1Ingress = self.cluster_info.get_ingress()
 
-        if opts.o.debug:
-            print(f"Sending this ingress: {ingress}")
-        ingress_resp = self.networking_api.create_namespaced_ingress(
-            namespace=self.k8s_namespace,
-            body=ingress
-        )
-        if opts.o.debug:
-            print("Ingress created:")
-            print(f"{ingress_resp}")
+            if ingress:
+                if opts.o.debug:
+                    print(f"Sending this ingress: {ingress}")
+                ingress_resp = self.networking_api.create_namespaced_ingress(
+                    namespace=self.k8s_namespace,
+                    body=ingress
+                )
+                if opts.o.debug:
+                    print("Ingress created:")
+                    print(f"{ingress_resp}")
+            else:
+                if opts.o.debug:
+                    print("No ingress configured")
 
-    def down(self, timeout, volumes):
+    def down(self, timeout, volumes):  # noqa: C901
         self.connect_api()
         # Delete the k8s objects
         # Create the host-path-mounted PVs for this deployment
@@ -175,6 +192,22 @@ class K8sDeployer(Deployer):
                     print(f"{pvc_resp}")
             except client.exceptions.ApiException as e:
                 _check_delete_exception(e)
+
+        # Figure out the ConfigMaps for this deployment
+        cfg_maps = self.cluster_info.get_configmaps()
+        for cfg_map in cfg_maps:
+            if opts.o.debug:
+                print(f"Deleting this ConfigMap: {cfg_map}")
+            try:
+                cfg_map_resp = self.core_api.delete_namespaced_config_map(
+                    name=cfg_map.metadata.name, namespace=self.k8s_namespace
+                )
+                if opts.o.debug:
+                    print("ConfigMap deleted:")
+                    print(f"{cfg_map_resp}")
+            except client.exceptions.ApiException as e:
+                _check_delete_exception(e)
+
         deployment = self.cluster_info.get_deployment()
         if opts.o.debug:
             print(f"Deleting this deployment: {deployment}")
@@ -196,16 +229,20 @@ class K8sDeployer(Deployer):
         except client.exceptions.ApiException as e:
             _check_delete_exception(e)
 
-        # TODO: disable ingress for kind
-        ingress: client.V1Ingress = self.cluster_info.get_ingress()
-        if opts.o.debug:
-            print(f"Deleting this ingress: {ingress}")
-        try:
-            self.networking_api.delete_namespaced_ingress(
-                name=ingress.metadata.name, namespace=self.k8s_namespace
-            )
-        except client.exceptions.ApiException as e:
-            _check_delete_exception(e)
+        if not self.is_kind():
+            ingress: client.V1Ingress = self.cluster_info.get_ingress()
+            if ingress:
+                if opts.o.debug:
+                    print(f"Deleting this ingress: {ingress}")
+                try:
+                    self.networking_api.delete_namespaced_ingress(
+                        name=ingress.metadata.name, namespace=self.k8s_namespace
+                    )
+                except client.exceptions.ApiException as e:
+                    _check_delete_exception(e)
+            else:
+                if opts.o.debug:
+                    print("No ingress to delete")
 
         if self.is_kind():
             # Destroy the kind cluster
@@ -219,7 +256,7 @@ class K8sDeployer(Deployer):
 
         if all_pods.items:
             for p in all_pods.items:
-                if self.cluster_info.app_name in p.metadata.name:
+                if f"{self.cluster_info.app_name}-deployment" in p.metadata.name:
                     pods.append(p)
 
         if not pods:
@@ -266,7 +303,7 @@ class K8sDeployer(Deployer):
         ret = []
 
         for p in pods.items:
-            if self.cluster_info.app_name in p.metadata.name:
+            if f"{self.cluster_info.app_name}-deployment" in p.metadata.name:
                 pod_ip = p.status.pod_ip
                 ports = AttrDict()
                 for c in p.spec.containers:
@@ -299,11 +336,20 @@ class K8sDeployer(Deployer):
 
     def logs(self, services, tail, follow, stream):
         self.connect_api()
-        pods = pods_in_deployment(self.core_api, "test-deployment")
+        pods = pods_in_deployment(self.core_api, self.cluster_info.app_name)
         if len(pods) > 1:
             print("Warning: more than one pod in the deployment")
-        k8s_pod_name = pods[0]
-        log_data = self.core_api.read_namespaced_pod_log(k8s_pod_name, namespace="default", container="test")
+        if len(pods) == 0:
+            log_data = "******* Pods not running ********\n"
+        else:
+            k8s_pod_name = pods[0]
+            # If the pod is not yet started, the logs request below will throw an exception
+            try:
+                log_data = self.core_api.read_namespaced_pod_log(k8s_pod_name, namespace="default", container="test")
+            except client.exceptions.ApiException as e:
+                if opts.o.debug:
+                    print(f"Error from read_namespaced_pod_log: {e}")
+                log_data = "******* No logs available ********\n"
         return log_stream_from_string(log_data)
 
     def update(self):

stack_orchestrator/deploy/k8s/helpers.py

@@ -21,6 +21,7 @@ from typing import Set, Mapping, List
 
 from stack_orchestrator.opts import opts
 from stack_orchestrator.deploy.deploy_util import parsed_pod_files_map_from_file_names
+from stack_orchestrator.deploy.deployer import DeployerException
 
 
 def _run_command(command: str):
@@ -29,10 +30,13 @@ def _run_command(command: str):
     result = subprocess.run(command, shell=True)
     if opts.o.debug:
         print(f"Result: {result}")
+    return result
 
 
 def create_cluster(name: str, config_file: str):
-    _run_command(f"kind create cluster --name {name} --config {config_file}")
+    result = _run_command(f"kind create cluster --name {name} --config {config_file}")
+    if result.returncode != 0:
+        raise DeployerException(f"kind create cluster failed: {result}")
 
 
 def destroy_cluster(name: str):
@@ -41,12 +45,14 @@ def destroy_cluster(name: str):
 
 def load_images_into_kind(kind_cluster_name: str, image_set: Set[str]):
     for image in image_set:
-        _run_command(f"kind load docker-image {image} --name {kind_cluster_name}")
+        result = _run_command(f"kind load docker-image {image} --name {kind_cluster_name}")
+        if result.returncode != 0:
+            raise DeployerException(f"kind create cluster failed: {result}")
 
 
 def pods_in_deployment(core_api: client.CoreV1Api, deployment_name: str):
     pods = []
-    pod_response = core_api.list_namespaced_pod(namespace="default", label_selector="app=test-app")
+    pod_response = core_api.list_namespaced_pod(namespace="default", label_selector=f"app={deployment_name}")
     if opts.o.debug:
         print(f"pod_response: {pod_response}")
     for pod_info in pod_response.items:
@@ -67,7 +73,7 @@ def named_volumes_from_pod_files(parsed_pod_files):
         parsed_pod_file = parsed_pod_files[pod]
         if "volumes" in parsed_pod_file:
             volumes = parsed_pod_file["volumes"]
-            for volume in volumes.keys():
+            for volume, value in volumes.items():
                 # Volume definition looks like:
                 # 'laconicd-data': None
                 named_volumes.append(volume)
@@ -92,22 +98,31 @@ def volume_mounts_for_service(parsed_pod_files, service):
                         volumes = service_obj["volumes"]
                         for mount_string in volumes:
                             # Looks like: test-data:/data
-                            (volume_name, mount_path) = mount_string.split(":")
-                            volume_device = client.V1VolumeMount(mount_path=mount_path, name=volume_name)
+                            parts = mount_string.split(":")
+                            volume_name = parts[0]
+                            mount_path = parts[1]
+                            mount_options = parts[2] if len(parts) == 3 else None
+                            volume_device = client.V1VolumeMount(
+                                mount_path=mount_path, name=volume_name, read_only="ro" == mount_options)
                             result.append(volume_device)
     return result
 
 
-def volumes_for_pod_files(parsed_pod_files):
+def volumes_for_pod_files(parsed_pod_files, spec):
     result = []
     for pod in parsed_pod_files:
         parsed_pod_file = parsed_pod_files[pod]
         if "volumes" in parsed_pod_file:
             volumes = parsed_pod_file["volumes"]
             for volume_name in volumes.keys():
-                claim = client.V1PersistentVolumeClaimVolumeSource(claim_name=volume_name)
-                volume = client.V1Volume(name=volume_name, persistent_volume_claim=claim)
-                result.append(volume)
+                if volume_name in spec.get_configmaps():
+                    config_map = client.V1ConfigMapVolumeSource(name=volume_name)
+                    volume = client.V1Volume(name=volume_name, config_map=config_map)
+                    result.append(volume)
+                else:
+                    claim = client.V1PersistentVolumeClaimVolumeSource(claim_name=volume_name)
+                    volume = client.V1Volume(name=volume_name, persistent_volume_claim=claim)
+                    result.append(volume)
     return result
 
 
@@ -152,7 +167,7 @@ def _generate_kind_mounts(parsed_pod_files, deployment_dir):
                         volume_definitions.append(
                             f"  - hostPath: {_make_absolute_host_path(volume_host_path_map[volume_name], deployment_dir)}\n"
                             f"    containerPath: {get_node_pv_mount_path(volume_name)}"
-                            )
+                        )
     return (
         "" if len(volume_definitions) == 0 else (
             "  extraMounts:\n"

stack_orchestrator/deploy/spec.py

@@ -22,6 +22,7 @@ from stack_orchestrator import constants
 class Spec:
 
     obj: typing.Any
+    file_path: Path
 
     def __init__(self) -> None:
         pass
@@ -29,12 +30,23 @@ class Spec:
     def init_from_file(self, file_path: Path):
         with file_path:
             self.obj = get_yaml().load(open(file_path, "r"))
+            self.file_path = file_path
 
     def get_image_registry(self):
         return (self.obj[constants.image_resigtry_key]
                 if self.obj and constants.image_resigtry_key in self.obj
                 else None)
 
+    def get_volumes(self):
+        return (self.obj["volumes"]
+                if self.obj and "volumes" in self.obj
+                else {})
+
+    def get_configmaps(self):
+        return (self.obj["configmaps"]
+                if self.obj and "configmaps" in self.obj
+                else {})
+
     def get_http_proxy(self):
         return (self.obj[constants.network_key][constants.http_proxy_key]
                 if self.obj and constants.network_key in self.obj

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

@@ -1,14 +1,59 @@
 #!/usr/bin/env bash
 set -e
 if [ -n "$CERC_SCRIPT_DEBUG" ]; then
-  set -x
+    set -x
+    # Dump environment variables for debugging
+    echo "Environment variables:"
+    env
 fi
+
+# Helper functions: TODO move into a separate file
+wait_for_pods_started () {
+    for i in {1..50}
+    do
+        local ps_output=$( $TEST_TARGET_SO deployment --dir $test_deployment_dir ps )
+
+        if [[ "$ps_output" == *"Running containers:"* ]]; then
+            # if ready, return
+            return
+        else
+            # if not ready, wait
+            sleep 5
+        fi
+    done
+    # Timed out, error exit
+    echo "waiting for pods to start: FAILED"
+    delete_cluster_exit
+}
+
+wait_for_log_output () {
+    for i in {1..50}
+    do
+
+        local log_output=$( $TEST_TARGET_SO deployment --dir $test_deployment_dir logs )
+
+        if [[ ! -z "$log_output" ]]; then
+            # if ready, return
+            return
+        else
+            # if not ready, wait
+            sleep 5
+        fi
+    done
+    # Timed out, error exit
+    echo "waiting for pods log content: FAILED"
+    delete_cluster_exit
+}
+
+
+delete_cluster_exit () {
+    $TEST_TARGET_SO deployment --dir $test_deployment_dir stop --delete-volumes
+    exit 1
+}
+
 # Note: eventually this test should be folded into ../deploy/
 # but keeping it separate for now for convenience
 TEST_TARGET_SO=$( ls -t1 ./package/laconic-so* | head -1 )
-# Dump environment variables for debugging
-echo "Environment variables:"
-env
 # Set a non-default repo dir
 export CERC_REPO_BASE_DIR=~/stack-orchestrator-test/repo-base-dir
 echo "Testing this package: $TEST_TARGET_SO"
@@ -18,7 +63,9 @@ echo "Version reported is: ${reported_version_string}"
 echo "Cloning repositories into: $CERC_REPO_BASE_DIR"
 rm -rf $CERC_REPO_BASE_DIR
 mkdir -p $CERC_REPO_BASE_DIR
-# Test basic stack-orchestrator deploy
+$TEST_TARGET_SO --stack test setup-repositories
+$TEST_TARGET_SO --stack test build-containers
+# Test basic stack-orchestrator deploy to k8s
 test_deployment_dir=$CERC_REPO_BASE_DIR/test-deployment-dir
 test_deployment_spec=$CERC_REPO_BASE_DIR/test-deployment-spec.yml
 $TEST_TARGET_SO --stack test deploy --deploy-to k8s-kind init --output $test_deployment_spec --config CERC_TEST_PARAM_1=PASSED
@@ -53,23 +100,37 @@ fi
 echo "deploy create output file test: passed"
 # Try to start the deployment
 $TEST_TARGET_SO deployment --dir $test_deployment_dir start
-# TODO: add a check to see if the container is up
-# Sleep because k8s not up yet
-sleep 30
+wait_for_pods_started
 # Check logs command works
+wait_for_log_output
 log_output_3=$( $TEST_TARGET_SO deployment --dir $test_deployment_dir logs )
 if [[ "$log_output_3" == *"Filesystem is fresh"* ]]; then
     echo "deployment logs test: passed"
 else
     echo "deployment logs test: FAILED"
-    exit 1
+    delete_cluster_exit
 fi
 # Check the config variable CERC_TEST_PARAM_1 was passed correctly
 if [[ "$log_output_3" == *"Test-param-1: PASSED"* ]]; then
     echo "deployment config test: passed"
 else
     echo "deployment config test: FAILED"
-    exit 1
+    delete_cluster_exit
+fi
+# Stop then start again and check the volume was preserved
+$TEST_TARGET_SO deployment --dir $test_deployment_dir stop
+# Sleep a bit just in case
+# sleep for longer to check if that's why the subsequent create cluster fails
+sleep 20
+$TEST_TARGET_SO deployment --dir $test_deployment_dir start
+wait_for_pods_started
+wait_for_log_output
+log_output_4=$( $TEST_TARGET_SO deployment --dir $test_deployment_dir logs )
+if [[ "$log_output_4" == *"Filesystem is old"* ]]; then
+    echo "Retain volumes test: passed"
+else
+    echo "Retain volumes test: FAILED"
+    delete_cluster_exit
 fi
 # Stop and clean up
 $TEST_TARGET_SO deployment --dir $test_deployment_dir stop --delete-volumes

tests/scripts/install-kind.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# TODO: handle ARM
+curl --silent -Lo ./kind https://kind.sigs.k8s.io/dl/v0.20.0/kind-linux-amd64
+chmod +x ./kind
+mv ./kind /usr/local/bin

tests/scripts/install-kubectl.sh

@@ -0,0 +1,5 @@
+#!/usr/bin/env bash
+# TODO: handle ARM
+curl --silent -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl"
+chmod +x ./kubectl
+mv ./kubectl /usr/local/bin