[BUG] OpenIM has periodic multi-minute outages despite normal server resources

[elijahdev0]

OpenIM Server Version

openim/openim-server:v3.8.3-patch.15

Operating System and CPU Architecture

Linux (AMD)

Deployment Method

Docker Deployment

Bug Description and Steps to Reproduce

I'm running OpenIM and Appwrite on the same VPS. Appwrite has zero issues, and the host itself is healthy (Ping and DNS monitors are 100% up for the last 24h, CPU/RAM are fine).

But three OpenIM monitors keep dropping:

• OpenIM WebSocket (WSS handshake to /msg_gateway every 20s, 30s timeout) — ~87% uptime
• OpenIM WS Gateway Reachable (HTTP GET to /msg_gateway every 20s, 30s timeout) — ~86% uptime
• OpenIM API Admin Token (HTTP POST to /api/auth/get_admin_token every 15s, 30s timeout) — ~99% uptime

The WebSocket and Gateway monitors fall over together in long blocks (~50 minutes at a time, e.g. 21:01–21:54 UTC and 04:01–04:55 UTC).

The admin token endpoint has shorter blips during some of those same windows. Everything recovers on its own.

Since the server is fine and Appwrite doesn't blink, I'm guessing this is something inside OpenIM's own processes — maybe health check gating, connection draining, or a component restarting. Any idea what could cause this pattern?

Screenshots Link

[elijahdev0]

I added monitoring for appwrite realtime too since they are on the same server. Everything is 100% uptime besides OpenIM websocket connection and get token