Atlas Profiler Parsed Logged Document

sample snippet from the parsed log document:

  "appName": "MongoDB Automation Agent v10.16.5.6520 (git: e4fef549c7036b5a31e046b5692a259467d6f68e)",
  "command": {
    "aggregate": "dispatchjobs",
    "maxTimeMS": 15000,
    "pipeline": [
        "$sample": {
          "size": 772
    "cursor": {},
    "lsid": {
      "id": {
        "$binary": "HJjZ1xLjRRuVloMZ01zw7A==",
        "$type": "03"
    "$clusterTime": {
      "clusterTime": {
        "$timestamp": {
          "t": 1599032548,
          "i": 83
      "signature": {
        "hash": {
          "$binary": "PHh4eHh4eD4=",
          "$type": "00"
        "keyId": {
          "$numberLong": "6818169589222866945"
    "$db": "openfleet-core",
    "$readPreference": {
      "mode": "primaryPreferred"
  "planSummary": [
      "MULTI_ITERATOR": {}

identify queries causing alerts

1 what is “MongoDB Automation Agent”?
2 why is it running a $sample at regular intervals?

i cannot find detailed info on parsed log document

any links or guidance will be appreciated as we are having performance issues and in the process of culling unused indexes and tuning.

thank you in advance


Hi Mark - Thank you for reaching out. I lead one of the product teams that is responsible for this query.

The specific query you see here is invoked by our “Schema Advisor” tool in Atlas, which recommends potential schema changes you can make to optimize performance. The “automation agent” is part of the code which does that. It’s a very low-level implementation detail, which probably shouldn’t be exposed in the profiler - so apologize for the confusion it has caused you.

The query also runs only when you use Schema Advisor. Can you confirm that you are not viewing or using the product regularly, either via the UI or API? If so and if we are still seeing it run on a regular interval, we will investigate it as a potential bug.

I am happy to work with you to diagnose performance issues, specially pertaining to culling unused indexes. We also have a recommendation product that will do it automatically for you that I can speak about. Please feel free to reach out to me at!



1 Like

Hi Rez,

what a great response, thank you.

So that makes sense and thanks for the useful info which i can report back to my team.

I have now tuned some indexes and alerts gone.

However whenever I try to run the performance/schema advisor, i constantly receive (despite time of day its run) the message “Unexpected Error Occurred: return to clusters”. i dont even get a list of collections to choose from. could there be an issue with the advisor?

Is guessing this should be a new ticket or related to the service agent message being logged?