Intermittent Webhook Loss (~50%) - Self-Hosted n8n & Evolution API on EasyPanel (DNS Issue?)

Questions
1

Hello everyone,

I’m facing a persistent issue with my self-hosted setup and I’m hoping to find the correct architectural solution.

My Setup

  • n8n: Self-hosted via EasyPanel (v1.119.1)

  • Webhook Source: Evolution API (also on the same VPS via EasyPanel)

  • VPS Provider: Hostinger

  • n8n URL: Publicly accessible (e.g., https://n8n.mydomain.com)

The Problem

I am losing approximately 50% of my incoming leads. These leads are coming directly from paid traffic (Meta Ads), so this 50% failure rate is critical for the business.

The flow is:

  1. A WhatsApp message is sent to my number.

  2. The message arrives on my phone.

  3. However, for about 50% of these attempts, the execution never starts in n8n. It does not appear in the “Executions” list (not pending, not failed—it simply never arrives).

What I’ve Investigated

  1. Tested Evolution API Versions: The problem persists. I have tried this with Evolution API v2.3.0 and have also tested the latest v2.3.6. The intermittent failure rate is the same.

  2. DNS Instability: When I test my n8n domain (https://n8n.mydomain.com) on a site like whatsmydns.net, I see constant, intermittent DNS propagation failures (the “red X’s”) from multiple servers around the world.

  3. Fresh Install: I completely formatted my VPS and did a fresh, clean installation of EasyPanel, n8n, and the Evolution API. The 50% loss problem immediately returned, which confirms this is likely an environmental/networking issue.

Has anyone else running n8n and an API on the same EasyPanel/VPS server experienced this? What is the recommended best practice to solve this webhook delivery failure?

PS: THE WORKFLOW WAS WORKING JUST FINE FOR ABOUT 6 MONTHS, AND THIS HAPPENED AFTER THE LAST EVOLUTION API AND N8N UPDATES, WITH NO CHANGES ON THE WORKFLOW ITSELF.

Thank you!