Failed: j.n.ConnectException: Connection refused (Connection refused)

Gatling Enterprise
1

We have been unable to run simulation even with single injector please help

[14:13:05,427] All injectors could be started. Proceeding with Running.
[14:13:05,427] Run 6560a46b-a65b-4098-9758-4dcdd593ee6f: starting injection
[14:17:55,444] Collecting stats from instance=West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] at offset=290 for offsets=([289, 290]) failed: j.n.ConnectException: Connection refused (Connection refused). Retrying.
[14:17:56,442] Lag detected: Run 6560a46b-a65b-4098-9758-4dcdd593ee6f: requesting West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] for offsets=([289, 291])
[14:17:56,443] Collecting stats from instance=West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] at offset=291 for offsets=([289, 291]) failed: j.n.ConnectException: Connection refused (Connection refused). Retrying.
[14:17:57,442] Lag detected: Run 6560a46b-a65b-4098-9758-4dcdd593ee6f: requesting West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] for offsets=([289, 292])
[14:17:57,443] Collecting stats from instance=West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] at offset=292 for offsets=([289, 292]) failed: j.n.ConnectException: Connection refused (Connection refused). Retrying.
[14:17:58,442] Lag detected: Run 6560a46b-a65b-4098-9758-4dcdd593ee6f: requesting West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] for offsets=([289, 293])
[14:17:58,443] Collecting stats from instance=West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] at offset=293 for offsets=([289, 293]) failed: j.n.ConnectException: Connection refused (Connection refused). Retrying.
[14:17:59,442] Lag detected: Run 6560a46b-a65b-4098-9758-4dcdd593ee6f: requesting West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] for offsets=([289, 294])
[14:17:59,443] Collecting stats from instance=West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] at offset=294 for offsets=([289, 294]) failed: j.n.ConnectException: Connection refused (Connection refused). Retrying.
[14:18:00,441] Lag detected: Run 6560a46b-a65b-4098-9758-4dcdd593ee6f: requesting West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] for offsets=([289, 295])
[14:18:00,443] Collecting stats from instance=West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] at offset=295 for offsets=([289, 295]) failed: j.n.ConnectException: Connection refused (Connection refused). Retrying.
[14:18:01,442] Lag detected: Run 6560a46b-a65b-4098-9758-4dcdd593ee6f: requesting West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] for offsets=([289, 296])
[14:18:01,443] Collecting stats from instance=West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] at offset=296 for offsets=([289, 296]) failed: j.n.ConnectException: Connection refused (Connection refused). Retrying.
[14:18:02,441] Lag detected: Run 6560a46b-a65b-4098-9758-4dcdd593ee6f: requesting West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] for offsets=([289, 297])
[14:18:03,442] Lag detected: Run 6560a46b-a65b-4098-9758-4dcdd593ee6f: requesting West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] for offsets=([289, 298])
[14:18:04,442] Lag detected: Run 6560a46b-a65b-4098-9758-4dcdd593ee6f: requesting West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] for offsets=([289, 299])
[14:18:05,442] At offset=299, failed to received stats for injectors List(West-Az2-L/10.84.40.158 [public IP: 10.84.40.158]): timeout. Stopping
[14:18:05,442] Injector on instance West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] crashed. Gatling Enterprise will now try to check if the injector process is still running and kill it.
[14:18:15,445] Failed to access crashed instance West-Az2-L/10.84.40.158 [public IP: 10.84.40.158] to get logs: j.u.c.TimeoutException: Task timed out after 10000ms
[14:18:15,771] West-Az2-L: instances successfully stopped
2

Hi,

It looks that the connectivity between the controller and the load generator was definitely lost, despite numerous delays and retries. Even connecting to the instance over SSH fails.

Possible explanations:

  • your test completely saturating IO on the load generator because you’re trying to generate a load that’s way too large for your single instance => check for high CPU and network segments usage in the “Injector” tab and for bandwidth usage in the “Connection” one. If you’re maxing CPU or IO, you have to distribute your test on multiple load generators, not just one.
  • Azure network issue => => to be investigated by your Azure support
  • some rule in your Azure cloud configuration causing the load generator instance to be quarantined => to be investigated by your ops people
  • some rule in your Azure cloud configuration causing your load generator instance to be terminated => to be investigated by your ops people

Also, are you aware that as a Gatling Enterprise Self-Hosted customer, you are entitled to a private support portal?