Merge remote-tracking branch 'origin/master' into main-2440-global-in…

…structions

Dockerfile

@@ -26,8 +26,8 @@ ENV PATH="$VIRTUAL_ENV/bin:$PATH"
 RUN curl -fsSL https://pkgs.k8s.io/core:/stable:/v1.29/deb/Release.key -o Release.key
 
 # Set the architecture-specific kube lineage URLs
-ARG ARM_URL=https://github.com/Avi-Robusta/kube-lineage/releases/download/v2.1/kube-lineage-macos-latest-v2.1
-ARG AMD_URL=https://github.com/Avi-Robusta/kube-lineage/releases/download/v2.1/kube-lineage-ubuntu-latest-v2.1
+ARG ARM_URL=https://github.com/Avi-Robusta/kube-lineage/releases/download/v2.2.1/kube-lineage-macos-latest-v2.2.1
+ARG AMD_URL=https://github.com/Avi-Robusta/kube-lineage/releases/download/v2.2.1/kube-lineage-ubuntu-latest-v2.2.1
 # Define a build argument to identify the platform
 ARG TARGETPLATFORM
 # Conditional download based on the platform
@@ -41,6 +41,8 @@ RUN if [ "$TARGETPLATFORM" = "linux/arm64" ]; then \
 RUN chmod 777 kube-lineage
 RUN ./kube-lineage --version
 
+RUN curl -sSL -o argocd-linux-amd64 https://github.com/argoproj/argo-cd/releases/latest/download/argocd-linux-amd64
+
 # Set up poetry
 ARG PRIVATE_PACKAGE_REGISTRY="none"
 RUN if [ "${PRIVATE_PACKAGE_REGISTRY}" != "none" ]; then \
@@ -90,6 +92,10 @@ RUN apt-get install -y kubectl
 COPY --from=builder /app/kube-lineage /usr/local/bin
 RUN kube-lineage --version
 
+COPY --from=builder /app/argocd-linux-amd64 /usr/local/bin/argocd
+RUN chmod 555 /usr/local/bin/argocd
+RUN argocd --help
+
 ARG AWS_DEFAULT_PROFILE
 ARG AWS_DEFAULT_REGION
 ARG AWS_PROFILE

README.md

@@ -523,6 +523,15 @@ HolmesGPT can consult webpages containing runbooks or other relevant information
 HolmesGPT uses playwright to scrape webpages and requires playwright to be installed and working through `playwright install`.
 </details>
 
+<details>
+<summary>
+ArgoCD
+</summary>
+
+Holmes can use the `argocd` CLI to get details about the ArgoCD setup like the apps configuration and status, clusters and projects within ArgoCD.
+To enable ArgoCD, set the `ARGOCD_AUTH_TOKEN` environment variable as described in the [argocd documentation](https://argo-cd.readthedocs.io/en/latest/user-guide/commands/argocd_account_generate-token/).
+</details>
+
 ## More Use Cases
 
 HolmesGPT was designed for incident response, but it is a general DevOps assistant too. Here are some examples:

holmes/plugins/toolsets/argocd.yaml

@@ -0,0 +1,43 @@
+toolsets:
+  argocd/core:
+    description: "Set of tools to get argocd metadata like list of apps, repositories, projects, etc."
+    docs_url: "https://argo-cd.readthedocs.io/en/stable/user-guide/commands/argocd/"
+    icon_url: "https://argo-cd.readthedocs.io/en/stable/assets/logo.png"
+    tags:
+      - core
+    prerequisites:
+      - command: "argocd version"
+      - env:
+        - ARGOCD_AUTH_TOKEN
+    tools:
+      - name: "argocd_app_get"
+        description: "Retrieve information about an existing application, such as its status and configuration"
+        command: "argocd app get {{ app_name }} --show-operation -o yaml"
+
+      - name: "argocd_app_diff"
+        description: "Display the differences between the current state of an application and the desired state specified in its Git repository"
+        command: "argocd app diff {{ app_name }}"
+
+      - name: "argocd_app_list"
+        description: "List the applications in Argocd"
+        command: "argocd app list"
+
+      - name: "argocd_app_history"
+        description: "List the deployment history of an application in ArgoCD"
+        command: "argocd app history {{app_name}} --app-namespace {{namespace}}"
+
+      - name: "argocd_repo_list"
+        description: "List all the Git repositories that ArgoCD is currently managing"
+        command: "argocd repo list"
+
+      - name: "argocd_proj_list"
+        description: "List all available projects"
+        command: "argocd proj list"
+
+      - name: "argocd_proj_get"
+        description: "Retrieves information about an existing project, such as its applications and policies"
+        command: "argocd proj get {{ project_name }}"
+
+      - name: "argocd_cluster_list"
+        description: "List all known clusters"
+        command: "argocd cluster list"

holmes/utils/holmes_status.py

@@ -6,6 +6,11 @@
 
 def update_holmes_status_in_db(dal: SupabaseDal, config: Config):
     logging.info("Updating status of holmes")
+
+    if not config.cluster_name:
+        raise Exception("Cluster name is missing in the configuration. Please ensure 'CLUSTER_NAME' is defined in the environment variables, "
+        "or verify that a cluster name is provided in the Robusta configuration file.")
+
     dal.upsert_holmes_status(
         {
             "cluster_id": config.cluster_name,

holmes/utils/holmes_sync_toolsets.py

@@ -8,7 +8,7 @@
 from holmes.plugins.prompts import load_and_render_prompt
 from holmes.utils.definitions import CUSTOM_TOOLSET_LOCATION
 import logging
-from datetime import datetime
+from datetime import datetime 
 
 
 def load_custom_toolsets_config() -> list[ToolsetYamlFromConfig]:
@@ -96,11 +96,21 @@ def holmes_sync_toolsets_status(dal: SupabaseDal, config) -> None:
         toolsets_loaded_from_config, default_toolsets_by_name
     )
 
-    db_toolsets = []
-    updated_at = datetime.now().isoformat()
+    # we check every toolset and save to local config toolsets which have passed the checks
+    # before we try to upsert anything to db
     for toolset in toolsets_for_sync_by_name.values():
         if toolset.enabled:
             toolset.check_prerequisites()
+
+    config.enabled_toolsets_names = [toolset.name for toolset in toolsets_for_sync_by_name.values() if toolset.get_status() == ToolsetStatusEnum.ENABLED]
+
+    if not config.cluster_name:
+        raise Exception("Cluster name is missing in the configuration. Please ensure 'CLUSTER_NAME' is defined in the environment variables, "
+        "or verify that a cluster name is provided in the Robusta configuration file.")
+
+    db_toolsets = []
+    updated_at = datetime.now().isoformat()
+    for toolset in toolsets_for_sync_by_name.values():
         if not toolset.installation_instructions:
             is_default_toolset = bool(toolset.name in default_toolsets_by_name.keys())
             instructions = render_default_installation_instructions_for_toolset(
@@ -118,7 +128,6 @@ def holmes_sync_toolsets_status(dal: SupabaseDal, config) -> None:
                 updated_at=updated_at
             ).model_dump(exclude_none=True)
         )
-    config.enabled_toolsets_names = [toolset.name for toolset in toolsets_for_sync_by_name.values() if toolset.get_status() == ToolsetStatusEnum.ENABLED]
     dal.sync_toolsets(db_toolsets, config.cluster_name)
 
 

server.py

@@ -1,17 +1,15 @@
 import os
-from holmes.core import investigation
 from holmes.utils.cert_utils import add_custom_certificate
-from contextlib import asynccontextmanager
-from holmes.utils.holmes_status import update_holmes_status_in_db
 
 ADDITIONAL_CERTIFICATE: str = os.environ.get("CERTIFICATE", "")
 if add_custom_certificate(ADDITIONAL_CERTIFICATE):
     print("added custom certificate")
 
 # DO NOT ADD ANY IMPORTS OR CODE ABOVE THIS LINE
 # IMPORTING ABOVE MIGHT INITIALIZE AN HTTPS CLIENT THAT DOESN'T TRUST THE CUSTOM CERTIFICATE
-
-
+from holmes.core import investigation
+from contextlib import asynccontextmanager
+from holmes.utils.holmes_status import update_holmes_status_in_db
 import jinja2
 import logging
 import uvicorn
@@ -128,16 +126,12 @@ def workload_health_check(request: WorkloadHealthRequest):
             request.ask = f"{request.ask}\n My instructions for the investigation '''{nl.join(instructions)}'''"
 
         global_instructions = dal.get_global_instructions_for_account()
-        print("GLOBAL INSTRUCTIONS")
-        print(global_instructions)
         if global_instructions and global_instructions.instructions and len(global_instructions.instructions[0]) > 0:
             request.ask += f"\n\nGlobal Instructions (use only if relevant): {global_instructions.instructions[0]}\n"
 
-        print(request.ask)
-        system_prompt = load_and_render_prompt(request.prompt_template)
-        system_prompt = jinja2.Environment().from_string(system_prompt)
-        system_prompt = system_prompt.render(alerts=workload_alerts)
-        print(system_prompt)
+        system_prompt = load_and_render_prompt(request.prompt_template, context={'alerts': workload_alerts})
+
+
         ai = config.create_toolcalling_llm(console, dal=dal)
 
         structured_output = {"type": "json_object"}