Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
AMBARI_HOSTNoHost address for the Ambari server
AMBARI_PASSNoPassword for Ambari server authentication (e.g., "admin")
AMBARI_PORTNoPort number for the Ambari server8080
AMBARI_USERNoUsername for Ambari server authentication (e.g., "admin")
FASTMCP_HOSTNoHost address for HTTP transport
FASTMCP_PORTNoPort number for HTTP transport (also enables streamable-http mode when set)
FASTMCP_TYPENoTransport type (stdio or streamable-http)
AMBARI_LOG_LEVELNoLogging level for the MCP server (DEBUG, INFO, WARNING, ERROR)
AMBARI_CLUSTER_NAMENoName of the target Ambari cluster (e.g., "TEST-AMBARI")

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

Apache Ambari (Hadoop) cluster operations through the Ambari REST API: cluster, service, component, and host status; service and component start/stop/restart; configuration lookup; request and task tracking; users; current and historical alerts; Ambari Metrics (AMS) time series; and HDFS capacity reports.

Scope

  • Every tool call is a real request to the configured Ambari server and cluster (AMBARI_CLUSTER_NAME unless a tool accepts cluster_name). Use tools for facts; never guess, simulate, or invent results, and do not tell the user to check the Ambari UI or run curl instead.

  • The server only reads live Ambari state and runs service operations. Generic Hadoop theory, tuning advice, log analysis, installs, and data deletion are out of scope; say so and offer a concrete status query.

  • Call only the tools needed; avoid repeating the same lookup.

Reporting results

  • Report what the tool returned. Never claim success after an error, an empty response, or a partial result; quote the error.

  • Start/stop/restart tools submit an Ambari request; accepted is not completed. Show the returned request ID near the top (Request ID: <id>, or Unknown if missing; never fabricate one) and point to get_request_status / get_request_tasks for progress.

Changing state

  • Run start/stop/restart only when the user explicitly asks for that action on a named service, component, or host. If the target is ambiguous ("restart it"), ask first.

  • *_all_services tools affect the whole cluster. Use them only when the user clearly asks for all services, confirm before running, and warn that live cluster state will change and may take several minutes.

  • If a required value (service, host, component, request ID) is missing or unclear, ask or look it up with a read tool (get_cluster_services, list_hosts, get_active_requests) instead of guessing.

Time ranges

  • For relative times ("yesterday", "last week"), call get_alerts_history with include_time_context=true, compute epoch-millisecond from_timestamp/to_timestamp from the returned current time, then call again with those values.

Metrics

  • query_ambari_metrics needs an explicit app_id and exact metric names. If unsure, find them first with list_ambari_metric_apps, list_common_metrics_catalog (search=), or list_ambari_metrics_metadata.

  • app_id mapping: Ambari Server=ambari_server, NameNode=namenode, DataNode=datanode, NodeManager=nodemanager, ResourceManager=resourcemanager, host-level=HOST. hostnames is optional.

  • Keep numbers distinct: configured values (dump_configurations), capacity (hdfs_dfadmin_report), and measured time series (query_ambari_metrics) are different things. State units, time window, and hosts. If no datapoints are returned, say so; never estimate.

Large results

  • Narrow before listing everything: dump_configurations with config_type, filter, service_filter, or summarize=true; get_alerts_history with filters, limit, page_size, and start_page; get_request_tasks with status_filter (e.g. "not:COMPLETED", "FAILED") and host_filter. Say when output is truncated or only one page was shown.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}
logging
{}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}
extensions
{
  "io.modelcontextprotocol/ui": {}
}
experimental
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
dump_configurationsC

Unified configuration introspection tool (supersedes get_configurations & list_configurations & dump_all_configurations).

get_cluster_infoA

Retrieves basic information for an Ambari cluster.

[Tool Role]: Dedicated tool for real-time retrieval of overall status and basic information for an Ambari cluster.

[Core Functions]:

  • Retrieve cluster name, version, provisioning state, and security type

  • Provide formatted output for LLM automation and cluster monitoring

[Required Usage Scenarios]:

  • When users request cluster info, status, or summary

  • When monitoring cluster health or auditing cluster properties

  • When users mention cluster overview, Ambari cluster, or cluster details

Returns: Cluster basic information (success: formatted info, failure: English error message)

get_active_requestsA

Retrieves currently active (in progress) requests/operations in an Ambari cluster. Shows running operations, in-progress tasks, pending requests.

[Tool Role]: Dedicated tool for monitoring currently running Ambari operations

[Core Functions]:

  • Retrieve active/running Ambari operations (IN_PROGRESS, PENDING status)

  • Show real-time progress of ongoing operations

  • Monitor current cluster activity

[Required Usage Scenarios]:

  • When users ask for "active requests", "running operations", "current requests"

  • When users ask for "request list", "operation list", "task list"

  • When users want to see "current tasks", "running tasks", "in progress operations"

  • When users mention "running", "in progress", "current activity"

  • When users ask about Ambari requests, operations, or tasks

  • When checking if any operations are currently running

Returns: Active requests information (success: active request list, failure: error message)

get_cluster_servicesA

Retrieves the list of services with status in an Ambari cluster.

[Tool Role]: Dedicated tool for real-time retrieval of all running services and basic status information in an Ambari cluster

[Core Functions]:

  • Retrieve cluster service list with status via Ambari REST API

  • Provide service names, current state, and cluster information

  • Include detailed link information for each service

  • Display visual indicators for service status

[Required Usage Scenarios]:

  • When users mention "service list", "cluster services", "Ambari services"

  • When cluster status check is needed

  • When service management requires current status overview

  • When real-time cluster information is absolutely necessary

[Absolutely Prohibited Scenarios]:

  • General Hadoop knowledge questions

  • Service installation or configuration changes

  • Log viewing or performance monitoring

  • Requests belonging to other cluster management tools

Returns: Cluster service list with status information (success: service list with status, failure: error message)

get_service_statusA

Retrieves the status information for a specific service in an Ambari cluster.

[Tool Role]: Dedicated tool for real-time retrieval of specific service status and state information

[Core Functions]:

  • Retrieve specific service status via Ambari REST API

  • Provide detailed service state information (STARTED, STOPPED, INSTALLING, etc.)

  • Include service configuration and component information

[Required Usage Scenarios]:

  • When users ask about specific service status (e.g., "HDFS status", "YARN state")

  • When troubleshooting service issues

  • When monitoring specific service health

get_service_componentsB

Retrieves detailed components information for a specific service in the Ambari cluster.

[Tool Role]: Dedicated tool for retrieving service component details and host assignments.

[Core Functions]:

  • List all components for a service, including state and category

  • Show host assignments and instance counts

  • Provide formatted output for LLM automation and troubleshooting

[Required Usage Scenarios]:

  • When users request service component details or host info

  • When troubleshooting service health or scaling

  • When users mention component list, host assignments, or service breakdown

get_service_detailsA

Retrieves detailed status and configuration information for a specific service in the Ambari cluster.

[Tool Role]: Dedicated tool for retrieving comprehensive service details, including state, components, and configuration.

[Core Functions]:

  • Retrieve service state, component list, and configuration availability

  • Provide formatted output for LLM automation and troubleshooting

[Required Usage Scenarios]:

  • When users request detailed service info or breakdown

  • When troubleshooting service health or auditing service setup

  • When users mention service details, service summary, or configuration status

start_all_servicesA

Starts all services in an Ambari cluster (equivalent to "Start All" in Ambari Web UI).

[Tool Role]: Dedicated tool for bulk starting all services in the cluster, automating mass startup.

[Core Functions]:

  • Start all installed services simultaneously

  • Return request information for progress tracking

  • Provide clear success or error message for LLM automation

[Required Usage Scenarios]:

  • When users request to "start all services", "start everything", "cluster startup"

  • When recovering cluster after maintenance or outage

  • When users mention mass startup, bulk start, or cluster bring-up

Returns: Start operation result (success: request info, failure: English error message)

stop_all_servicesA

Stops all services in an Ambari cluster (equivalent to "Stop All" in Ambari Web UI).

[Tool Role]: Dedicated tool for bulk stopping all services in the cluster, automating mass shutdown.

[Core Functions]:

  • Stop all running services simultaneously

  • Return request information for progress tracking

  • Provide clear success or error message for LLM automation

[Required Usage Scenarios]:

  • When users request to "stop all services", "stop everything", "cluster shutdown"

  • When cluster maintenance or troubleshooting requires mass shutdown

  • When users mention mass shutdown, bulk stop, or cluster halt

Returns: Stop operation result (success: request info, failure: English error message)

start_serviceA

Starts a specific service in the Ambari cluster.

[Tool Role]: Dedicated tool for automated start of Ambari services, ensuring safe and monitored startup.

[Core Functions]:

  • Start the specified service and initiate Ambari request

  • Return request information for progress tracking

  • Provide clear success or error message for LLM automation

[Required Usage Scenarios]:

  • When users request to "start" a service (e.g., "start HDFS", "start YARN")

  • When recovering stopped services

  • When maintenance or configuration changes require a service start

  • When users mention service start, bring up service, or automated start

stop_serviceA

Stops a specific service in the Ambari cluster.

[Tool Role]: Dedicated tool for automated stop of Ambari services, ensuring safe and monitored shutdown.

[Core Functions]:

  • Stop the specified service and initiate Ambari request

  • Return request information for progress tracking

  • Provide clear success or error message for LLM automation

[Required Usage Scenarios]:

  • When users request to "stop" a service (e.g., "stop HDFS", "stop YARN")

  • When maintenance or troubleshooting requires a service shutdown

  • When users mention service stop, shutdown, or automated stop

get_request_statusA

Retrieves the status and progress of a specific Ambari request operation.

[Tool Role]: Dedicated tool for real-time tracking and reporting of Ambari request status.

[Core Functions]:

  • Query the status, progress, and context of a request by its ID

  • Provide detailed status (PENDING, IN_PROGRESS, COMPLETED, FAILED, etc.)

  • Show progress percentage and timing information

  • Return actionable status for automation and LLM integration

[Required Usage Scenarios]:

  • When users ask for the status or progress of a specific operation/request

  • When monitoring or troubleshooting Ambari operations

  • When tracking bulk or individual service actions

  • When users mention request ID, operation status, or progress

get_request_tasksA

Retrieves task-level details for a specific Ambari request (host-by-host breakdown).

[Tool Role]: Provides granular, per-host task information for an Ambari request, going beyond the summary level offered by get_request_status.

[Core Functions]:

  • List all tasks (host/role/status/timing) belonging to a request

  • Filter tasks by status (e.g. FAILED, IN_PROGRESS) or exclude a status (not:COMPLETED)

  • Filter tasks by hostname to inspect a specific host

  • Identify which hosts/roles are failing, pending, or still running

[Required Usage Scenarios]:

  • When users want to know which hosts or roles failed/are pending in an operation

  • When debugging a partially-failed bulk operation (e.g. "Start all services")

  • When users ask for task details, host-level progress, or remaining tasks

  • When request progress is near 100% but not complete and user wants to know what's left

restart_serviceA

Restarts a specific service in an Ambari cluster (stop then start).

[Tool Role]: Dedicated tool for automated restart of Ambari services, ensuring safe stop and start sequence.

[Core Functions]:

  • Stop the specified service and wait for completion

  • Start the service and wait for completion

  • Return clear success or error message for LLM automation

[Required Usage Scenarios]:

  • When users request to "restart" a service (e.g., "restart HDFS", "restart YARN")

  • When troubleshooting or recovering service issues

  • When maintenance or configuration changes require a restart

  • When users mention service restart, safe restart, or automated restart

start_host_componentA

Starts a specific component on a specific host in the Ambari cluster.

[Tool Role]: Dedicated tool for starting individual host-level components, enabling fine-grained control without affecting other hosts or the overall service.

[Core Functions]:

  • Start the specified component on the given host

  • Skip if the component is already in STARTED state

  • Return request information for progress tracking

[Required Usage Scenarios]:

  • When a specific host's component is STOPPED but the overall service is STARTED

  • When users request to start a component on a specific host (e.g., "start DataNode on host-A")

  • When recovering a single component instance without restarting the entire service

stop_host_componentA

Stops a specific component on a specific host in the Ambari cluster.

[Tool Role]: Dedicated tool for stopping individual host-level components, enabling fine-grained control without affecting other hosts or the overall service.

[Core Functions]:

  • Stop the specified component on the given host

  • Skip if the component is already in INSTALLED (stopped) state

  • Return request information for progress tracking

[Required Usage Scenarios]:

  • When users request to stop a component on a specific host (e.g., "stop DataNode on host-A")

  • When decommissioning or isolating a specific node's component

  • When troubleshooting a single component instance

restart_host_componentA

Restarts a specific component on a specific host in the Ambari cluster (stop then start).

[Tool Role]: Dedicated tool for restarting individual host-level components, enabling fine-grained control without affecting other hosts or the overall service.

[Core Functions]:

  • Stop the specified component on the given host and wait for completion

  • Start the component and return request information

  • Return clear success or error message for LLM automation

[Required Usage Scenarios]:

  • When users request to restart a component on a specific host (e.g., "restart DataNode on host-A")

  • When recovering a stuck or malfunctioning single component instance

  • When applying config changes that require a component-level restart

restart_all_servicesA

Restarts all services in the Ambari cluster (stop all, then start all).

[Tool Role]: Dedicated tool for automated bulk restart of all Ambari services, ensuring safe stop and start sequence.

[Core Functions]:

  • Stop all running services and wait for completion

  • Start all services and wait for completion

  • Return clear success or error message for LLM automation

[Required Usage Scenarios]:

  • When users request to "restart all services", "bulk restart", "cluster-wide restart"

  • When troubleshooting or recovering cluster-wide issues

  • When maintenance or configuration changes require a full restart

Returns: Bulk restart operation result (success: English completion message, failure: English error message) - Success: "All services restart operation completed successfully." - Failure: "Error: ..." with details

list_hostsA

Retrieves the list of hosts in the Ambari cluster.

[Tool Role]: Dedicated tool for listing all hosts registered in the Ambari cluster.

[Core Functions]:

  • Query Ambari REST API for host list

  • Return host names and API links

  • Provide formatted output for LLM automation and cluster management

[Required Usage Scenarios]:

  • When users request cluster host list or host details

  • When auditing or monitoring cluster nodes

Returns: List of hosts (success: formatted list, failure: error message)

get_host_detailsA

Retrieves detailed information for a specific host or all hosts in the Ambari cluster.

[Tool Role]: Dedicated tool for retrieving comprehensive host details including metrics, hardware info, and components.

[Core Functions]:

  • If host_name provided: Query specific host information

  • If host_name not provided: Query all hosts and their detailed information

  • Return host hardware specs, state, metrics, and assigned components

  • Provide formatted output for LLM automation and cluster management

[Required Usage Scenarios]:

  • When users request specific host details or host status

  • When users request all hosts details or cluster-wide host information

  • When auditing or monitoring individual or all cluster nodes

  • When troubleshooting host-specific issues

list_usersA

List all users in the Ambari system.

Returns a formatted list of all users with their basic information.

get_userB

Get detailed information about a specific user.

get_alerts_historyA

Retrieve current alerts or alert history from Ambari cluster.

[Tool Role]: Unified tool for retrieving both current/active alert status and historical alert events from Ambari cluster

[Core Functions]:

  • Current mode: Retrieve current alerts for entire cluster, specific service, or specific host

  • History mode: Retrieve alert history for entire cluster, specific service, or specific host

  • Support filtering by alert state (CRITICAL, WARNING, OK, UNKNOWN)

  • Support filtering by definition name

  • Current mode: Support filtering by maintenance state (ON, OFF)

  • History mode: Support filtering by time range with from_timestamp/to_timestamp

  • Support different output formats (detailed, summary, compact, groupedSummary for current)

  • History mode: Provide pagination support for large datasets

  • Provide current time context for LLM natural language time calculations

[Required Usage Scenarios]:

  • Current mode: When users request current alerts, active alerts, or alert status

  • Current mode: When monitoring immediate cluster health

  • Current mode: When investigating current issues or troubleshooting active problems

  • History mode: When users request alert history, past alerts, or historical alert data

  • History mode: When monitoring alert trends or analyzing alert patterns

  • History mode: When investigating past alert incidents or troubleshooting

  • When users mention alert status, current problems, cluster health, alert events, alert timeline, or alert logs

get_current_alertsA

Retrieve current alerts from Ambari cluster.

DEPRECATED: This function is deprecated in favor of get_alerts_history with mode="current". This wrapper is maintained for backward compatibility.

list_ambari_metrics_metadataC

Retrieve metric metadata from Ambari Metrics service with optional filters.

list_common_metrics_catalogC

List supported metrics per appId (exact names only).

list_ambari_metric_appsC

Return discovered AMS appIds, optionally with metric counts.

hdfs_dfadmin_reportC

Produce a DFSAdmin-style capacity and DataNode report using Ambari metrics.

query_ambari_metricsC

Fetch time-series metrics (exact metric names only) from Ambari Metrics.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

TDQS

B3.2/5.0

Scored across 29 tools

Disambiguation3/5

Several tools have overlapping purposes: get_service_details already includes state and components, blurring the line with get_service_status and get_service_components; get_host_details can retrieve all hosts with details, overlapping list_hosts; and the deprecated get_current_alerts remains alongside get_alerts_history. Descriptions help, but an agent could misselect, especially for service/host introspection and alert retrieval.

Naming Consistency4/5

Nearly all tools follow snake_case with a verb_noun pattern (get_, list_, start_, stop_, restart_, dump_). Exceptions like hdfs_dfadmin_report (noun phrase) and mixed get_/list_ for retrieval are minor deviations that don't hinder readability.

Tool Count2/5

29 tools is above the 25-tool threshold for 'too many' given the guideline; some consolidation is possible (e.g., deprecating get_current_alerts, merging service status/components into details). While Ambari is complex, the surface feels heavy with redundant read operations.

Completeness3/5

Core lifecycle operations for services and components (start/stop/restart) and read access for hosts, requests, alerts, metrics, and users are present. However, notable gaps exist: no configuration update (dump_configurations only reads), no user CRUD beyond list/get, and no host add/remove or service installation—operations likely needed for full cluster management.

Maintenance

ActivityActive
ResponsivenessNo recent community issues