Merge branch 'master' into claude-slack

Author: Avi-Robusta

docs/configuration/holmesgpt/builtin_toolsets.rst

@@ -18,7 +18,9 @@ Builtin Toolsets
    toolsets/internet
    toolsets/kafka
    toolsets/kubernetes
-   toolsets/opensearch
+   toolsets/notion
+   toolsets/opensearch_logs
+   toolsets/opensearch_status
    toolsets/prometheus
    toolsets/rabbitmq
    toolsets/robusta
@@ -98,10 +100,20 @@ by the user by providing credentials or API keys to external systems.
         :class-card: sd-bg-light sd-bg-text-light
         :link: toolsets/kubernetes
         :link-type: doc
+    
+    .. grid-item-card:: :octicon:`cpu;1em;` Notion
+        :class-card: sd-bg-light sd-bg-text-light
+        :link: toolsets/notion
+        :link-type: doc
+
+    .. grid-item-card:: :octicon:`cpu;1em;` OpenSearch logs
+        :class-card: sd-bg-light sd-bg-text-light
+        :link: toolsets/opensearch_logs
+        :link-type: doc
 
-    .. grid-item-card:: :octicon:`cpu;1em;` OpenSearch
+    .. grid-item-card:: :octicon:`cpu;1em;` OpenSearch status
         :class-card: sd-bg-light sd-bg-text-light
-        :link: toolsets/opensearch
+        :link: toolsets/opensearch_status
         :link-type: doc
 
     .. grid-item-card:: :octicon:`cpu;1em;` Prometheus

docs/configuration/holmesgpt/index.rst

@@ -10,12 +10,15 @@ AI Analysis
    builtin_toolsets
    permissions
    custom_toolsets
+   remote_mcp_servers
 
 Why use HolmesGPT?
 ^^^^^^^^^^^^^^^^^^^
 
 Robusta can integrate with `Holmes GPT <https://github.com/robusta-dev/holmesgpt>`_ to analyze health issues on your cluster, and to run AI based root cause analysis for alerts.
 
+This requires a Robusta SaaS account, and for the Robusta UI sink to be enabled. (We have plans to support HolmesGPT in a pure OSS mode in the near future. Stay tuned!)
+
 When available, AI based investigations can be launched in one of two ways:
 
 1. Click the ``Ask HolmesGPT`` button in Slack. The AI investigation will be sent back as a new message.
@@ -35,7 +38,7 @@ Add ``enableHolmesGPT: true`` to the Robusta Helm values, and then follow these
 
 1. Choose an AI model - we highly recommend using GPT-4o to get the most accurate results! Other models may work, but are not officially supported.
 2. :ref:`Configure your AI provider with the chosen model <Choosing and configuring an AI provider>`.
-3. :ref:`Optional: Configure HolmesGPT Access to SaaS Data <Configuring HolmesGPT Access to SaaS Data>`.
+3. :ref:`Configure HolmesGPT Access to SaaS Data <Configuring HolmesGPT Access to SaaS Data>`.
 
 Choosing and configuring an AI provider
 ----------------------------------------
@@ -362,7 +365,7 @@ Builtin toolsets
 
 Built-in toolsets cover essential areas like pod status inspection, node health analysis,
 application diagnostics, and resource utilization monitoring. These toolsets include access to
-Kubernetes events and logs, AWS, Grafana, Opensearch, etc. See the full list :doc:`here <builtin_toolsets>`.
+Kubernetes events and logs, AWS, Grafana, OpenSearch, etc. See the full list :doc:`here <builtin_toolsets>`.
 
 Custom toolsets
 ----------------
@@ -371,3 +374,14 @@ Custom toolsets
 
 Custom toolsets are created through your Helm values file and you can find instructions
 to :doc:`write your own toolsets here <builtin_toolsets>`.
+
+Remote MCP servers 
+---------------------
+
+.. warning::
+
+    Remote MCP servers are in **Tech Preview** stage.
+
+Remote MCP server connections are configured through your Helm values file.
+For detailed instructions, refer to the :doc:`Connecting to Remote MCP Servers <remote_mcp_servers>` guide.
+

docs/configuration/holmesgpt/remote_mcp_servers.rst

@@ -0,0 +1,46 @@
+Remote MCP Servers 
+====================
+
+.. warning::
+
+    Remote MCP servers are in **Tech Preview** stage.
+
+
+HolmesGPT can integrate with remote MCP servers using SSE mode.
+This capability enables HolmesGPT to access external data sources and tools in real time.
+This guide provides step-by-step instructions for configuring HolmesGPT to connect with remote MCP servers over SSE.
+
+Example : MCP server configuration
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+.. md-tab-set::
+
+  .. md-tab-item:: Robusta Helm Chart
+
+    **Helm Values:**
+
+    .. code-block:: yaml
+
+      holmes:
+        mcp_servers:
+          mcp_server_1:
+            # human-readable description of the mcp server (this is not seen by the AI model - its just for users)
+            description: "Remote mcp server"
+            url: "http://example.com:8000/sse"
+          
+          mcp_server_2:
+            description: "MCP server that runs in my cluster"
+            url: "http://<service-name>.<namespace>.svc.cluster.local:<service-port>"
+            config:
+              headers:
+                key: "{{ env.my_mcp_server_key }}" # You can use holmes environment variables as headers for the MCP server requests.
+        
+    Update your Helm values with the provided YAML configuration, then apply the changes with Helm upgrade:
+
+    .. code-block:: bash
+
+        helm upgrade robusta robusta/robusta --values=generated_values.yaml --set clusterName=<YOUR_CLUSTER_NAME>
+
+    
+
+

docs/configuration/holmesgpt/toolsets/_disable_default_logging_toolset.inc.rst

@@ -0,0 +1,31 @@
+
+Disabling the Default Logging Toolset
+*****************************************
+
+The default HolmesGPT logging tool **must** be disabled if you use a different datasource for logs.
+HolmesGPT may still use kubectl to fetch logs and never call your datasource if ``kubernetes/logs`` is not disabled. 
+To disable the default logging toolset, add the following to your holmes configuration:
+
+.. md-tab-set::
+
+  .. md-tab-item:: Robusta Helm Chart
+
+    .. code-block:: yaml
+
+      holmes:
+        toolsets:
+          kubernetes/logs:
+            enabled: false
+
+
+    .. include:: ./_toolset_configuration.inc.rst
+
+  .. md-tab-item:: Holmes CLI
+
+    Add the following to **~/.holmes/config.yaml**, creating the file if it doesn't exist:
+
+    .. code-block:: yaml
+
+      toolsets:
+        kubernetes/logs:
+          enabled: false

docs/configuration/holmesgpt/toolsets/_toolsets_that_provide_logging.inc.rst

@@ -1,5 +1,6 @@
 HolmesGPT provides several out-of-the-box alternatives for log access. You can select from these options:
 
-* :ref:`kubernetes/logs <toolset_kubernetes_logs>`: Access logs with ``kubectl logs`` commands. **This is the default toolset.**
+* :ref:`kubernetes/logs <toolset_kubernetes_logs>`: Access logs directly through Kubernetes. **This is the default toolset.**
 * :ref:`coralogix/logs <toolset_coralogix_logs>`: Access logs through Coralogix.
 * :ref:`grafana/loki <toolset_grafana_loki>`: Access Loki logs by proxying through a Grafana instance.
+* :ref:`opensearch/logs <toolset_opensearch_logs>`: Access logs through OpenSearch.

docs/configuration/holmesgpt/toolsets/coralogix_logs.rst

@@ -3,10 +3,10 @@
 Coralogix logs
 ==============
 
-By enabling this toolset, HolmesGPT will fetch node and pods logs from `Coralogix <https://coralogix.com/>`_.
+By enabling this toolset, HolmesGPT will fetch pod logs from `Coralogix <https://coralogix.com/>`_.
 
 You **should** enable this toolset to replace the default :ref:`kubernetes/logs <toolset_kubernetes_logs>`
-toolset if all your kubernetes/pod logs are consolidated inside Coralogix. It will make it easier for HolmesGPT
+toolset if all your kubernetes pod logs are consolidated inside Coralogix. It will make it easier for HolmesGPT
 to fetch incident logs, including the ability to precisely consult past logs.
 
 
@@ -43,7 +43,7 @@ Configuration
               team_hostname: my-team # Your team's hostname in coralogix, without the domain part
 
           kubernetes/logs:
-            enabled: false # Disable HolmesGPT's default logging mechanism
+            enabled: false # HolmesGPT's default logging mechanism MUST be disabled
 
 
     .. include:: ./_toolset_configuration.inc.rst
@@ -63,13 +63,13 @@ Configuration
             team_hostname: my-team # Your team's hostname in coralogix
 
         kubernetes/logs:
-          enabled: false # Disable HolmesGPT's default logging mechanism
+          enabled: false # HolmesGPT's default logging mechanism MUST be disabled
 
 Advanced Configuration
 ^^^^^^^^^^^^^^^^^^^^^^
 
 Frequent logs and archive
-****************************
+*************************
 
 By default, holmes fetched the logs from the `Frequent search <https://coralogix.com/docs/user-guides/account-management/tco-optimizer/logs/#frequent-search-data-high-priority>`_ 
 tier and only fetch logs from the `Archive` tier if the frequent search returned no result.
@@ -100,10 +100,10 @@ Here is a description of each possible log retrieval methodology:
 - **BOTH_FREQUENT_SEARCH_AND_ARCHIVE** Always use both the frequent search and the archive to fetch logs. The result contains merged data which is deduplicated and sorted by timestamp.
 
 Search labels
-***************
+*************
 
 You can tweak the labels used by the toolset to identify kubernetes resources. This is **optional** and only needed if your
-logs settings for ``pod``, ``namespace``, ``application`` and ``subsystem`` differ from the defaults in the example below.
+logs settings for ``pod`` and ``namespace`` differ from the defaults in the example below.
 
 .. code-block:: yaml
 
@@ -114,8 +114,6 @@ logs settings for ``pod``, ``namespace``, ``application`` and ``subsystem`` diff
         labels: # OPTIONAL: tweak the filters used by HolmesGPT if your coralogix configuration is non standard
           namespace: "kubernetes.namespace_name"
           pod: "kubernetes.pod_name"
-          application: "coralogix.metadata.applicationName"
-          subsystem: "coralogix.metadata.subsystemName"
         ...
 
 
@@ -126,19 +124,8 @@ You can verify what labels to use by attempting to run a query in the coralogix
   :align: center
 
 
-Disabling the default toolset
-*********************************
 
-If Coralogix is your primary datasource for logs, it is **advised** to disable the default HolmesGPT logging
-tool by disabling the ``kubernetes/logs`` toolset. Without this. HolmesGPT may still use kubectl to
-fetch logs instead of Coralogix.
-
-.. code-block:: yaml
-
-    holmes:
-        toolsets:
-            kubernetes/logs:
-                enabled: false
+.. include:: ./_disable_default_logging_toolset.inc.rst
 
 
 Capabilities
@@ -152,5 +139,5 @@ Capabilities
 
    * - Tool Name
      - Description
-   * - fetch_coralogix_logs_for_resource
+   * - fetch_pod_logs
      - Retrieve logs using coralogix

docs/configuration/holmesgpt/toolsets/grafanaloki.rst

@@ -3,7 +3,7 @@
 Loki
 ====
 
-By enabling this toolset, HolmesGPT will fetch node and pods logs from `Loki <https://grafana.com/oss/loki/>`_.
+By enabling this toolset, HolmesGPT will fetch pod logs from `Loki <https://grafana.com/oss/loki/>`_.
 Loki can be accessed directly or by proxying through a `Grafana <https://grafana.com/oss/grafana/>`_ instance.
 
 You **should** enable this toolset to replace the default :ref:`kubernetes/logs <toolset_kubernetes_logs>`
@@ -20,7 +20,7 @@ Proxying through Grafana
 This is the recommended approach because we intend to add more capabilities to the toolset that are only available with Grafana.
 
 Prerequisites
--------------
+*************
 
 A `Grafana service account token <https://grafana.com/docs/grafana/latest/administration/service-accounts/>`_
 with the following permissions:
@@ -79,7 +79,7 @@ A simple way to get the datasource UID is to access the Grafana API by running t
 
 
 Configuration (grafana proxy)
------------------------------
+*****************************
 
 
 .. md-tab-set::
@@ -98,7 +98,7 @@ Configuration (grafana proxy)
               grafana_datasource_uid: <the UID of the loki data source in Grafana>
 
           kubernetes/logs:
-            enabled: false # Disable HolmesGPT's default logging mechanism
+            enabled: false # HolmesGPT's default logging mechanism MUST be disabled
 
 
     .. include:: ./_toolset_configuration.inc.rst
@@ -118,7 +118,7 @@ Configuration (grafana proxy)
             grafana_datasource_uid: <the UID of the loki data source in Grafana>
 
         kubernetes/logs:
-          enabled: false # Disable HolmesGPT's default logging mechanism
+          enabled: false # HolmesGPT's default logging mechanism MUST be disabled
 
 Direct connection
 ^^^^^^^^^^^^^^^^^
@@ -128,7 +128,7 @@ This is done by not setting the ``grafana_datasource_uid`` field. Not setting th
 assume that it is directly connecting to Loki.
 
 Configuration (direct connection)
----------------------------------
+*********************************
 
 .. md-tab-set::
 
@@ -146,7 +146,7 @@ Configuration (direct connection)
                 X-Scope-OrgID: "<tenant id>" # Set the X-Scope-OrgID if loki multitenancy is enabled
 
           kubernetes/logs:
-            enabled: false # Disable HolmesGPT's default logging mechanism
+            enabled: false # HolmesGPT's default logging mechanism MUST be disabled
 
 
     .. include:: ./_toolset_configuration.inc.rst
@@ -166,13 +166,14 @@ Configuration (direct connection)
               X-Scope-OrgID: "<tenant id>" # Set the X-Scope-OrgID if loki multitenancy is enabled
 
         kubernetes/logs:
-          enabled: false # Disable HolmesGPT's default logging mechanism
+          enabled: false # HolmesGPT's default logging mechanism MUST be disabled
 
 
 Advanced configuration
 ^^^^^^^^^^^^^^^^^^^^^^
 
-**Search labels**
+Search labels
+*************
 
 You can tweak the labels used by the toolset to identify kubernetes resources. This is only needed if your
 Loki logs settings for ``pod``, and ``namespace`` differ from the defaults in the example above.
@@ -223,19 +224,7 @@ Use the following commands to list Loki's labels and determine which ones to use
     curl http://localhost:3100/loki/api/v1/labels
 
 
-**Disabling the default toolset**
-
-If Loki is your primary datasource for logs, it is **advised** to disable the default HolmesGPT logging
-tool by disabling the ``kubernetes/logs`` toolset. Without this. HolmesGPT may still use kubectl to
-fetch logs instead of Loki.
-
-.. code-block:: yaml
-
-    holmes:
-        toolsets:
-            kubernetes/logs:
-                enabled: false
-
+.. include:: ./_disable_default_logging_toolset.inc.rst
 
 Capabilities
 ^^^^^^^^^^^^
@@ -248,7 +237,5 @@ Capabilities
 
    * - Tool Name
      - Description
-   * - fetch_loki_logs_for_resource
-     - Fetches the Loki logs for a given kubernetes resource
-   * - fetch_loki_logs
-     - Fetches Loki logs from any query
+   * - fetch_pod_logs
+     - Fetches pod logs

docs/configuration/holmesgpt/toolsets/grafanatempo.rst

@@ -135,9 +135,6 @@ Configuration (direct connection)
               headers:
                 X-Scope-OrgID: "<tenant id>" # Set the X-Scope-OrgID if tempo multitenancy is enabled
 
-          kubernetes/logs:
-            enabled: false # Disable HolmesGPT's default logging mechanism
-
 
     .. include:: ./_toolset_configuration.inc.rst
 
@@ -155,9 +152,6 @@ Configuration (direct connection)
             headers:
               X-Scope-OrgID: "<tenant id>" # Set the X-Scope-OrgID if tempo multitenancy is enabled
 
-        kubernetes/logs:
-          enabled: false # Disable HolmesGPT's default logging mechanism
-
 
 Advanced configuration
 ^^^^^^^^^^^^^^^^^^^^^^