[HELP] Server running but unreachable via Web GUI & no ping response – post v7.2.1 upgrade

Hi all,

I'm hoping to get some community insight on an issue I've been experiencing a few months after upgrading to Unraid v7.2.1.

The Problem:

My server appears to still be running (drives spinning, lights on, no apparent crash), but it becomes completely unreachable — no response to ping and the Web GUI is inaccessible. A reboot restores access temporarily, but the issue eventually returns.

What I've checked so far:

- Server physically appears powered on and operational

- No response to ping on local network

- Web GUI (both IP and tower.local) times out

- NIC has been verified as working fine (no hardware issue)

- HDMI no response

What I'm uncertain about:

- Whether this is a software/driver hang, a network stack issue, or something triggered by a specific Unraid process

- Whether v7.2.1 introduced anything that could cause this

Questions for the community:

1. Has anyone experienced similar symptoms after upgrading to v7.2.1?

2. What logs should I pull after a reboot to help pinpoint the cause? (syslog, diagnostics zip?)

3. Could this be related to Docker networking, a runaway process, or something else on the software side?

I'll attach a diagnostics zip in a follow-up reply once I can capture one after the next occurrence. Any pointers on what to look for would be greatly appreciated!

Thanks in advance.

I will reboot the server after this post.

Do you have an attached monitor and keyboard?

UPDATE - Issue Investigation & Fixes Applied

After extensive troubleshooting, I wanted to share my findings in case anyone else experiences similar symptoms. The issue turned out to be caused by multiple factors rather than a single root cause.

Root Causes Identified:

1. Dynamix S3 Sleep plugin (2018.02.04) — This outdated plugin was causing the server to enter sleep/suspend state and fail to properly restore network connectivity on wake. It is marked as "Unknown to Community Applications" and is not compatible with Unraid v7.2.1. Removing this plugin immediately improved stability.

2. Docker MacVLAN network configuration — All three of my Docker containers (Prowlarr, Sonarr, Transmission) were running on MacVLAN/Bridge network mode. This was causing Docker to manipulate the host's network interfaces on startup, resulting in complete network loss. Switching all containers to Host network mode and disabling Host access to custom networks in Docker settings resolved this.

3. Degraded Realtek RTL8168g NIC — My onboard Realtek NIC was only negotiating at 100Mbps instead of 1Gbps despite being a Gigabit NIC, a Cat5e cable, and a Gigabit router. This was causing intermittent network drops under load, especially during Plex streaming. As a temporary fix, I disabled auto-negotiation and locked the NIC to 100Mbps Full Duplex via ethtool. A replacement Intel NIC (i210/i211) is on order.

Fixes Applied:

- Removed Dynamix S3 Sleep plugin

- Switched all Docker containers to Host network mode

- Disabled Host access to custom networks in Docker settings

- Locked NIC to 100Mbps Full Duplex (temporary fix)

- Fixed appdata share pointing to missing Cache pool

Current Status:

Server is significantly more stable. Still monitoring while waiting for replacement Intel NIC to arrive.

Additional findings worth noting:

- Three of my four data disks are still formatted as ReiserFS, which is now deprecated and scheduled for kernel removal. Planning to migrate to XFS.

- No swap space configured — worth considering if running memory-intensive workloads like Plex transcoding.