Skip to main content

​
Mongodbatlas Receiver

Status Available in: contrib Maintainers: @justinianvoss22, @dyl10s, @ishleenk17 Source: opentelemetry-collector-contrib

​
Supported Telemetry

Logs Metrics

​
Overview

​
Getting Started

The MongoDB Atlas receiver takes the following parameters. public_key and private_key are the only two required values to receive metrics and logs and are obtained via the “API Keys” tab of the MongoDB Atlas Project Access Manager. In the example below both values are being pulled from the environment. In order to collect logs, at least one project must be specified. By default, logs for all clusters within a project will be collected. Clusters can be limited using either the include_clusters or exclude_clusters setting. In order to collect project events, the requesting API key needs the appropriate permission which at minimum is the Project Read Only role. Project events are specific to a single project. In order to collect organization events, the requesting API key needs the appropriate permission which at minimum is the Organization Member role. Organization events are collected across all the projects hosted on Atlas within the organization. These events are not associated with a project. In order to collect access logs, the requesting API key needs the appropriate permission which requires either the Project Owner or Organization Owner role. Access logs are specific to each cluster. MongoDB Atlas Documentation recommends a polling interval of 5 minutes.
  • base_url (default https://cloud.mongodb.com/) set the base URL to connect to Atlas Cloud
  • public_key (required for metrics, logs, or alerts in poll mode)
  • private_key (required for metrics, logs, or alerts in poll mode)
  • granularity (default PT1M - See MongoDB Atlas Documentation)
  • collection_interval (default 3m) This receiver collects metrics on an interval. Valid time units are ns, us (or µs), ms, s, m, h.
  • storage (optional) The component ID of a storage extension which can be used when polling for alerts or events . The storage extension prevents duplication of data after a collector restart by remembering which data were previously collected.
  • projects (optional for metrics) a slice of projects this receiver collects metrics from instead of all projects in an organization
    • name Name of the project to discover metrics from
    • include_clusters (default empty, exclusive with exclude_clusters)
    • exclude_clusters (default empty, exclusive with include_clusters)
      • If both include_clusters and exclude_clusters are empty, then all clusters in the project will be included
  • retry_on_failure
    • enabled (default true)
    • initial_interval (default 5s)
    • max_interval (default 30s)
    • max_elapsed_time (default 5m)
  • alerts
    • enabled (default false)
    • mode (default listen. Options are poll or listen)
    • secret (required if using listen mode)
    • endpoint (required if using listen mode)
    • poll_interval (default 5m, only relevant using poll mode)
    • page_size (default 100)
      • When in poll mode, this is the number of alerts that will be processed per request to the MongoDB Atlas API.
    • max_pages (default 10)
      • When in poll mode, this will limit how many pages of alerts the receiver will request for each project.
    • projects (required if using poll mode)
      • name (required if using poll mode)
      • include_clusters (default empty, exclusive with exclude_clusters)
      • exclude_clusters (default empty, exclusive with include_clusters)
        • If both include_clusters and exclude_clusters are empty, then all clusters in the project will be included
    • tls (relevant only for listen mode)
      • key_file
      • cert_file
  • logs
    • enabled (default false)
    • projects (required if enabled)
      • name (required if enabled)
      • collect_host_logs (default true)
      • collect_audit_logs (default false)
      • access_logs
        • enabled (default true, if the access_logs parameter is defined)
        • auth_result
          • If specified, will limit the access logs queried to successful accesses (true) or failed accesses (false). If not specified, all will be collected
        • page_size (default 20000)
          • This is the number of access logs that will be processed per request to the MongoDB Atlas API. The maximum value is 20000.
        • max_pages (default 10)
          • This will limit how many pages of access logs the receiver will request from the MongoDB Atlas API for a project.
        • poll_interval (default 5m)
          • This will define how frequently the MongoDB Atlas API is queried for Access Logs for the given project.
      • include_clusters (default empty)
      • exclude_clusters (default empty)
  • events
    • projects
      • name Name of the Project to discover events from.
    • organizations
      • id ID of the Organization to discover events from.
    • poll_interval (default 1m)
      • How often the receiver will poll the Events API for new events.
    • page_size (default 100)
      • This is the number of events that will be processed per request to the MongoDB Atlas API.
    • max_pages (default 25)
      • This will limit how many pages of events the receiver will request from the MongoDB Atlas API for each project.
    • types (defaults to all types of events)
      • This is a list of event types that the receiver will request from the API. If specified, the receiver will collect only the indicated types of events.
Examples: Receive metrics: 
receivers:
  mongodb_atlas:
    public_key: ${env:MONGODB_ATLAS_PUBLIC_KEY}
    private_key: ${env:MONGODB_ATLAS_PRIVATE_KEY}
Listen for alerts (default mode): 
receivers:
  mongodb_atlas:
    alerts:
      enabled: true
      secret: "some_secret"
      endpoint: "0.0.0.0:7706"
Poll alerts from API: 
receivers:
  mongodb_atlas:
    public_key: <redacted>
    private_key: <redacted>
    alerts:
      enabled: true
      mode: poll
      projects:
      - name: Project 0
        include_clusters: [Cluster0]
      poll_interval: 1m
    # use of a storage extension is recommended to reduce chance of duplicated alerts
    storage: file_storage
Receive logs: 
receivers:
  mongodb_atlas:
    logs:
      enabled: true
      projects: 
        - name: "project 1"
          collect_audit_logs: true
          collect_host_logs: true
Receive events: 
receivers:
  mongodb_atlas:
    events:
      projects:
        - name: "project 1"
      organizations:
        - id: "5b478b3afc4625789ce616a3"
      poll_interval: 1m
      page_size: 100
      max_pages: 25
    # use of a storage extension is recommended to reduce chance of duplicated events
    storage: file_storage
Poll Access Logs from API: 
receivers:
  mongodb_atlas:
    public_key: <redacted>
    private_key: <redacted>
    logs:
      enabled: true
      projects:
      - name: Project 0
        include_clusters: [Cluster0]
        access_logs:
          page_size: 20000
          max_pages: 10
          poll_interval: 5m
    # use of a storage extension is recommended to reduce chance of duplicated access logs
    storage: file_storage

​
Metrics

Metric NameDescriptionUnitTypeAttributes
✅ mongodbatlas.db.countsDatabase feature size{objects}Gaugeobject_type
✅ mongodbatlas.db.sizeDatabase feature sizeByGaugeobject_type
✅ mongodbatlas.disk.partition.iops.averageDisk partition iops{ops}/sGaugedisk_direction
✅ mongodbatlas.disk.partition.iops.maxDisk partition iops{ops}/sGaugedisk_direction
✅ mongodbatlas.disk.partition.latency.averageDisk partition latencymsGaugedisk_direction
✅ mongodbatlas.disk.partition.latency.maxDisk partition latencymsGaugedisk_direction
❌ mongodbatlas.disk.partition.queue.depthDisk queue depth1Gauge
✅ mongodbatlas.disk.partition.space.averageDisk partition spaceByGaugedisk_status
✅ mongodbatlas.disk.partition.space.maxDisk partition spaceByGaugedisk_status
❌ mongodbatlas.disk.partition.throughputDisk throughputBy/sGaugedisk_direction
✅ mongodbatlas.disk.partition.usage.averageDisk partition usage (%)1Gaugedisk_status
✅ mongodbatlas.disk.partition.usage.maxDisk partition usage (%)1Gaugedisk_status
✅ mongodbatlas.disk.partition.utilization.averageThe percentage of time during which requests are being issued to and serviced by the partition.1Gauge
✅ mongodbatlas.disk.partition.utilization.maxThe maximum percentage of time during which requests are being issued to and serviced by the partition.1Gauge
✅ mongodbatlas.process.assertsNumber of assertions per second{assertions}/sGaugeassert_type
✅ mongodbatlas.process.background_flushAmount of data flushed in the background1Gauge
✅ mongodbatlas.process.cache.ioCache throughput (per second)ByGaugecache_direction
❌ mongodbatlas.process.cache.ratioCache ratios represented as (%)%Gaugecache_ratio_type
✅ mongodbatlas.process.cache.sizeCache sizesByUpDownCountercache_status
✅ mongodbatlas.process.connectionsNumber of current connections{connections}UpDownCounter
✅ mongodbatlas.process.cpu.children.normalized.usage.averageCPU Usage for child processes, normalized to pct1Gaugecpu_state
✅ mongodbatlas.process.cpu.children.normalized.usage.maxCPU Usage for child processes, normalized to pct1Gaugecpu_state
✅ mongodbatlas.process.cpu.children.usage.averageCPU Usage for child processes (%)1Gaugecpu_state
✅ mongodbatlas.process.cpu.children.usage.maxCPU Usage for child processes (%)1Gaugecpu_state
✅ mongodbatlas.process.cpu.normalized.usage.averageCPU Usage, normalized to pct1Gaugecpu_state
✅ mongodbatlas.process.cpu.normalized.usage.maxCPU Usage, normalized to pct1Gaugecpu_state
✅ mongodbatlas.process.cpu.usage.averageCPU Usage (%)1Gaugecpu_state
✅ mongodbatlas.process.cpu.usage.maxCPU Usage (%)1Gaugecpu_state
✅ mongodbatlas.process.cursorsNumber of cursors{cursors}Gaugecursor_state
✅ mongodbatlas.process.db.document.rateDocument access rates{documents}/sGaugedocument_status
✅ mongodbatlas.process.db.operations.rateDB Operation Rates{operations}/sGaugeoperation, cluster_role
✅ mongodbatlas.process.db.operations.timeDB Operation TimesmsCounterexecution_type
✅ mongodbatlas.process.db.query_executor.scannedScanned objects{objects}/sGaugescanned_type
✅ mongodbatlas.process.db.query_targeting.scanned_per_returnedScanned objects per returned{scanned}/{returned}Gaugescanned_type
✅ mongodbatlas.process.db.storageStorage used by the databaseByGaugestorage_status
✅ mongodbatlas.process.global_lockNumber and status of locks{locks}Gaugeglobal_lock_state
✅ mongodbatlas.process.index.btree_miss_ratioIndex miss ratio (%)1Gauge
✅ mongodbatlas.process.index.countersIndexes{indexes}Gaugebtree_counter_type
✅ mongodbatlas.process.journaling.commitsJournaling commits{commits}Gauge
✅ mongodbatlas.process.journaling.data_filesData file sizesMiByGauge
✅ mongodbatlas.process.journaling.writtenJournals writtenMiByGauge
✅ mongodbatlas.process.memory.usageMemory UsageByGaugememory_state
✅ mongodbatlas.process.network.ioNetwork IOBy/sGaugedirection
✅ mongodbatlas.process.network.requestsNetwork requests{requests}Counter
✅ mongodbatlas.process.oplog.rateExecution rate by operationGiBy/hGauge
✅ mongodbatlas.process.oplog.timeExecution time by operationsGaugeoplog_type
✅ mongodbatlas.process.page_faultsPage faults{faults}/sGaugememory_issue_type
✅ mongodbatlas.process.restartsRestarts in last hour{restarts}/hGauge
✅ mongodbatlas.process.ticketsTickets{tickets}Gaugeticket_type
✅ mongodbatlas.system.cpu.normalized.usage.averageSystem CPU Normalized to pct1Gaugecpu_state
✅ mongodbatlas.system.cpu.normalized.usage.maxSystem CPU Normalized to pct1Gaugecpu_state
✅ mongodbatlas.system.cpu.usage.averageSystem CPU Usage (%)1Gaugecpu_state
✅ mongodbatlas.system.cpu.usage.maxSystem CPU Usage (%)1Gaugecpu_state
✅ mongodbatlas.system.fts.cpu.normalized.usageFull text search disk usage (%)1Gaugecpu_state
✅ mongodbatlas.system.fts.cpu.usageFull-text search (%)1Gaugecpu_state
✅ mongodbatlas.system.fts.disk.usedFull text search disk usageByGauge
✅ mongodbatlas.system.fts.memory.usageFull-text searchMiByCountermemory_state
✅ mongodbatlas.system.memory.usage.averageSystem Memory UsageKiByGaugememory_status
✅ mongodbatlas.system.memory.usage.maxSystem Memory UsageKiByGaugememory_status
✅ mongodbatlas.system.network.io.averageSystem Network IOBy/sGaugedirection
✅ mongodbatlas.system.network.io.maxSystem Network IOBy/sGaugedirection
✅ mongodbatlas.system.paging.io.averageSwap IO{pages}/sGaugedirection
✅ mongodbatlas.system.paging.io.maxSwap IO{pages}/sGaugedirection
✅ mongodbatlas.system.paging.usage.averageSwap usageKiByGaugememory_state
✅ mongodbatlas.system.paging.usage.maxSwap usageKiByGaugememory_state

​
Attributes

Attribute NameDescriptionTypeValues
assert_typeMongoDB assertion typestringregular, warning, msg, user
btree_counter_typeDatabase index effectivenessstringaccesses, hits, misses
cache_directionWhether read into or written fromstringread_into, written_from
cache_ratio_typeCache ratio typestringcache_fill, dirty_fill
cache_statusCache statusstringdirty, used
cluster_roleWhether process is acting as replica or primarystringprimary, replica
cpu_stateCPU statestringkernel, user, nice, iowait, irq, softirq, guest, steal
cursor_stateWhether cursor is open or timed outstringtimed_out, open
directionNetwork traffic directionstringreceive, transmit
disk_directionMeasurement type for disk operationstringread, write, total
disk_statusDisk measurement typestringfree, used
document_statusStatus of documents in the databasestringreturned, inserted, updated, deleted
execution_typeType of commandstringreads, writes, commands
global_lock_stateWhich queue is lockedstringcurrent_queue_total, current_queue_readers, current_queue_writers
memory_issue_typeType of memory issue encounteredstringextra_info, global_accesses_not_in_memory, exceptions_thrown
memory_stateMemory usage typestringresident, virtual, mapped, computed, shared, free, used
memory_statusMemory measurement typestringavailable, buffers, cached, free, shared, used
object_typeMongoDB object typestringcollection, index, extent, object, view, storage, data
operationType of database operationstringcmd, query, update, delete, getmore, insert, scan_and_order, ttl_deleted
oplog_typeOplog typestringslave_lag_master_time, master_time, master_lag_time_diff
scanned_typeObjects or indexes scanned during querystringindex_items, objects
storage_statusViews on database sizestringtotal, data_size, index_size, data_size_wo_system
ticket_typeType of ticket availablestringavailable_reads, available_writes

​
Resource Attributes

Attribute NameDescriptionTypeEnabled
mongodb_atlas.cluster.nameCluster Namestring❌
mongodb_atlas.db.nameName of the Databasestring✅
mongodb_atlas.disk.partitionName of a disk partitionstring✅
mongodb_atlas.host.nameHostname of the processstring✅
mongodb_atlas.org_nameOrganization Namestring✅
mongodb_atlas.process.idID of the processstring✅
mongodb_atlas.process.portPort process is bound tostring✅
mongodb_atlas.process.type_nameProcess typestring✅
mongodb_atlas.project.idProject IDstring✅
mongodb_atlas.project.nameProject Namestring✅
mongodb_atlas.provider.nameProvider Namestring❌
mongodb_atlas.region.nameRegion Namestring❌
mongodb_atlas.user.aliasUser-friendly hostname of the cluster nodestring❌

​
Configuration

​
Example Configuration

mongodb_atlas:
  public_key: "my-public-key"
  private_key: "my-private-key"
  alerts:
    enabled: true
    mode: poll
    projects:
    - name: Project 0
      include_clusters: [Cluster0]
    poll_interval: 1m
  logs:
    enabled: true
    projects:
    - name: Project 0
      access_logs:
        enabled: true
        auth_result: true
        poll_interval: 1m
      collect_audit_logs: true
  events:
    projects:
      - name: Project 0
    organizations:
      - id: 5b478b3afc4625789ce616a3
    poll_interval: 1m
    page_size: 100
    max_pages: 25
Last generated: 2026-04-13