MongoDB

Connect HolmesGPT to MongoDB databases to query data, investigate slow queries, analyze collection schemas, and diagnose performance issues like connection exhaustion, lock contention, and replication lag.

You can configure multiple MongoDB instances with different names (e.g., prod-mongo, analytics-mongo, staging-mongo).

Creating a Read-Only User

use admin
db.createUser({
  user: "holmes_readonly",
  pwd: "your_secure_password",
  roles: [
    { role: "readAnyDatabase", db: "admin" },
    { role: "clusterMonitor", db: "admin" }
  ]
})

The readAnyDatabase role grants read access to all databases. The clusterMonitor role enables the serverStatus and currentOp diagnostics tools.

Configuration

Holmes CLIHolmes Helm ChartRobusta Helm Chart

~/.holmes/config.yaml:

toolsets:
  prod-mongo:
    type: mongodb
    config:
      connection_url: "{{ env.PROD_MONGO_URL }}"
    llm_instructions: "Production MongoDB with customer and order data"

  analytics-mongo:
    type: mongodb
    config:
      connection_url: "{{ env.ANALYTICS_MONGO_URL }}"
    llm_instructions: "Analytics MongoDB for reporting"

After making changes to your configuration, run:

holmes toolset refresh

Connection URL format:

mongodb://[username]:[password]@[host]:[port]/[database]
mongodb+srv://[username]:[password]@[cluster].mongodb.net/[database]

Step 1: Create secret with credentials

kubectl create secret generic mongodb-credentials \
  --from-literal=url='mongodb://holmes_readonly:your_secure_password@mongo.example.com:27017/mydb' \
  -n holmes

Step 2: Configure in values.yaml

additionalEnvVars:
  - name: MONGO_URL
    valueFrom:
      secretKeyRef:
        name: mongodb-credentials
        key: url

toolsets:
  prod-mongo:
    type: mongodb
    config:
      connection_url: "{{ env.MONGO_URL }}"
    llm_instructions: "Production MongoDB database"

Multiple instances:

additionalEnvVars:
  - name: PROD_MONGO_URL
    valueFrom:
      secretKeyRef:
        name: mongo-prod
        key: url
  - name: ANALYTICS_MONGO_URL
    valueFrom:
      secretKeyRef:
        name: mongo-analytics
        key: url

toolsets:
  prod-mongo:
    type: mongodb
    config:
      connection_url: "{{ env.PROD_MONGO_URL }}"

  analytics-mongo:
    type: mongodb
    config:
      connection_url: "{{ env.ANALYTICS_MONGO_URL }}"

Step 1: Create secret with credentials

kubectl create secret generic mongodb-credentials \
  --from-literal=url='mongodb://holmes_readonly:your_secure_password@mongo.example.com:27017/mydb' \
  -n default

Step 2: Configure in values.yaml

holmes:
  additionalEnvVars:
    - name: MONGO_URL
      valueFrom:
        secretKeyRef:
          name: mongodb-credentials
          key: url

  toolsets:
    prod-mongo:
      type: mongodb
      config:
        connection_url: "{{ env.MONGO_URL }}"
      llm_instructions: "Production MongoDB database"

Multiple instances:

holmes:
  additionalEnvVars:
    - name: PROD_MONGO_URL
      valueFrom:
        secretKeyRef:
          name: mongo-prod
          key: url
    - name: ANALYTICS_MONGO_URL
      valueFrom:
        secretKeyRef:
          name: mongo-analytics
          key: url

  toolsets:
    prod-mongo:
      type: mongodb
      config:
        connection_url: "{{ env.PROD_MONGO_URL }}"

    analytics-mongo:
      type: mongodb
      config:
        connection_url: "{{ env.ANALYTICS_MONGO_URL }}"

Update your Helm values and run a Helm upgrade:

helm upgrade robusta robusta/robusta --values=generated_values.yaml --set clusterName=<YOUR_CLUSTER_NAME>

Configuration Options

• connection_url (required): MongoDB connection string (mongodb:// or mongodb+srv://)
• default_database: Database to use when not specified per-query (also read from URL)
• read_only (default: true): Only allow read operations (find, aggregate without $out/$merge)
• verify_ssl (default: true): Verify SSL certificates
• max_rows (default: 200): Maximum documents to return per query (1-10000)
• timeout_seconds (default: 30): Connection and operation timeout
• llm_instructions: Context about this database for the LLM

Tools

Tool	Description
{prefix}_find	Query documents with filter, projection, sort, and limit
{prefix}_aggregate	Run aggregation pipelines ($match, $group, $sort, $lookup, etc.)
{prefix}_list_databases	List all databases with their sizes
{prefix}_list_collections	List collections in a database
{prefix}_collection_schema	Infer field types from sample documents and show indexes
{prefix}_server_status	Server diagnostics: connections, memory, opcounters, locks, WiredTiger cache
{prefix}_current_op	Find slow or blocked operations, filterable by duration

{prefix} is derived from the instance name (e.g., prod_mongo_find for prod-mongo).

Common Use Cases

"List all collections in the orders database and show their schemas"

"Find the 10 most recent error documents in the events collection"

"How many orders per status are there in the last 24 hours?"

"Are there any slow queries running on the MongoDB server?"

"Check the MongoDB server health — connections, memory, and lock statistics"