docs: rabbitmq toolset

Author: nherment

docs/configuration/holmesgpt/builtin_toolsets.rst

@@ -20,6 +20,7 @@ Builtin Toolsets
    toolsets/kubernetes
    toolsets/opensearch
    toolsets/prometheus
+   toolsets/rabbitmq
    toolsets/robusta
    toolsets/slab
 
@@ -108,6 +109,11 @@ by the user by providing credentials or API keys to external systems.
         :link: toolsets/prometheus
         :link-type: doc
 
+    .. grid-item-card:: :octicon:`cpu;1em;` RabbitMQ
+        :class-card: sd-bg-light sd-bg-text-light
+        :link: toolsets/rabbitmq
+        :link-type: doc
+
     .. grid-item-card:: :octicon:`cpu;1em;` Robusta
         :class-card: sd-bg-light sd-bg-text-light
         :link: toolsets/robusta

docs/configuration/holmesgpt/toolsets/rabbitmq.rst

@@ -0,0 +1,94 @@
+.. _toolset_prometheus:
+
+RabbitMQ
+========
+
+By enabling this toolset, HolmesGPT will be able to detect RabbitMQ partitions, memory alerts and disk alerts and suggest mitigations.
+
+This toolset follows a two step process to detect partition:
+
+1. The nodes and partitioning status is obtained by fetching information from the configured ``management_url``.
+2. If some nodes are reported as not-running, the toolset will try to contact these nodes individually and deduct any partitioning 
+   state for any node that is actually running.
+
+
+Configuration
+-------------
+
+.. md-tab-set::
+
+  .. md-tab-item:: Robusta Helm Chat
+
+    .. code-block:: yaml
+
+      holmes:
+        toolsets:
+          rabbitmq/core:
+            enabled: true
+            config:
+              clusters:
+                - id: rabbitmq # must be unique across all configured clusters
+                  username: <user>
+                  password: <password>
+                  management_url: <http://rabbitmq.rabbitmq:15672>
+
+
+    .. 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:
+        rabbitmq/core:
+          enabled: true
+          config:
+            clusters:
+              - id: rabbitmq # must be unique across all configured clusters
+                username: <user>
+                password: <password>
+                management_url: <http://rabbitmq.rabbitmq:15672>
+
+
+**Advanced configuration**
+
+Below is the full list of options for this toolset:
+
+.. code-block:: yaml
+
+  rabbitmq/core:
+    enabled: true
+    config:
+      clusters:
+        - id: rabbitmq # must be unique across all configured clusters
+          username: <user>
+          password: <password>
+          management_url: <http://rabbitmq.rabbitmq:15672>
+          request_timeout_seconds: 30 # timeout for the API requests to RabbitMQ
+          verify_certs: True # if the management API is https and the certificate is not valid
+
+
+- **id** A unique identifier for this cluster. Holmes uses it to figure out what cluster to check when answering a question or investigating an alert.
+- **username** A username to authenticate against the management API
+- **password** A pasword to authenticate against the management API
+- **management_url** The base URL for this RabbitMQ's cluster. Typically you would want to put the URL of the RabbitMQ service ``http://rabbitmq.rabbitmq:15672``.
+- **request_timeout_seconds** The timeout in seconds for the API requests to RabbitMQ.
+- **verify_certs** Set to False if RabbitMQ exposes the API through self signed certificates.
+
+
+Capabilities
+------------
+.. include:: ./_toolset_capabilities.inc.rst
+
+.. list-table::
+   :header-rows: 1
+   :widths: 30 70
+
+   * - Tool Name
+     - Description
+   * - list_configured_clusters
+     - List all configured clusters. Useful to get the id of a configured cluster (cluster_id) and pass as argument to other rabbitmq tool calls.
+   * - get_rabbitmq_cluster_status
+     - Fetches the overall status of the RabbitMQ cluster, including node information, listeners, and partition details. Crucial for detecting split-brain scenarios (network partitions).