GCP Instance Groups doesn't work

When creating any Instance group except Unmanaged one, it creates, alive for a few minutes, creates instances,and then just terminates and writes Unknown error

I am following the configuration from the official GCP tutorial but changing all the unsupported services to the ones that are supported by playground, still it gives this error

Please check, project clgcporg8-016

I’ve found some logs in Logging that can be helpful. Here is the error which arises when I create Instance group

{
  "insertId": "116wx6kf5r5dg9",
  "jsonPayload": {
    "message": "Error reloading service: Failed to reload-or-restart sshd.service: Unit sshd.service not found..",
    "localTimestamp": "2024-01-15T12:41:55.5188Z",
    "omitempty": null
  },
  "resource": {
    "type": "gce_instance",
    "labels": {
      "instance_id": "3318733743996836097",
      "project_id": "clgcporg8-064",
      "zone": "us-central1-c"
    }
  },
  "timestamp": "2024-01-15T12:41:55.518923532Z",
  "severity": "ERROR",
  "labels": {
    "instance_name": "instance-group-global-2m1m"
  },
  "logName": "projects/clgcporg8-064/logs/GCEGuestAgent",
  "sourceLocation": {
    "file": "oslogin.go",
    "line": "131",
    "function": "main.(*osloginMgr).Set"
  },
  "receiveTimestamp": "2024-01-15T12:41:55.969540632Z"
}

Here is the log entry just before

{
  "insertId": "116wx6kf5r5dg8",
  "jsonPayload": {
    "omitempty": null,
    "localTimestamp": "2024-01-15T12:41:54.8724Z",
    "message": "added: [now 2024-01-15 12:41:54.872436663 +0000 UTC entry 1 next 2024-01-16 12:41:54 +0000 UTC]"
  },
  "resource": {
    "type": "gce_instance",
    "labels": {
      "project_id": "clgcporg8-064",
      "zone": "us-central1-c",
      "instance_id": "3318733743996836097"
    }
  },
  "timestamp": "2024-01-15T12:41:54.872480965Z",
  "severity": "INFO",
  "labels": {
    "instance_name": "instance-group-global-2m1m"
  },
  "logName": "projects/clgcporg8-064/logs/GCEGuestAgent",
  "sourceLocation": {
    "file": "logger.go",
    "line": "24",
    "function": "github.com/GoogleCloudPlatform/guest-agent/google_guest_agent/scheduler.(*cronLogger).Info"
  },
  "receiveTimestamp": "2024-01-15T12:41:55.969540632Z"
}

I’ve also noticed that it worked fine when I choose multiple zones instead of the single one in the Instance groups

Hi @danylo.hotvianskyi,

How many instances are you deploying? Can you please check the allowed resource limits from the link below:

GCP Playground: Support Services (kodekloud.com)

Regards,

Hello @Tej-Singh-Rana, 1-2 instances. Not more

Actually, I might have resolved the issue by creating google_compute_region_instance_group_manager instead of google_compute_instance_group_manager

It would be useful to mention this in limitations too, if this is the case

Thanks for your input. I will ask the team to check from their end.

Regards,