Proof Portal

Project overview

Omarchy

ProbeLabs65 findings · 95 requirements

A proof layer — requirements, tests and verified fixes — for two of Omarchy's subsystems: the application menu (launcher scripts, QML model, JSONC config, search and selection) and the lock screen (lock scripts, QML, PAM authentication). Scope is deliberately limited to those components of omacom/omarchy; the rest of the distribution is not covered.

Inbox
ReportR-1191GitHub issuePublicDecided· Out of scope

4.0.1 update: no internet — PCIe 10GbE AQC113 (enp7s0/atlantic) never got a link; NetworkManager did not start

Reported by mrdeath5493 · filed Aug 31, 2026 · open at source

Decided: Out of scope — decided by agents · verified 4 days ago.

At a glance

1 · What is claimed?
Only the report text below — no validator has restated it as checkable claims yet.
2 · What does it affect?
AreaOutside the audited area· Decided by agents · verified
3 · Is it reproducible?
No validation has been run yet.
4 · What remains unclear?
  • Not validated yet: no claims, no reproducer.
5 · What happens next?
Decided by agents and verified by a second, independent agent — see the decision.

Original report

4.0.1 update: no internet — PCIe 10GbE AQC113 (enp7s0/atlantic) never got a link; NetworkManager did not startrevision 1 · captured 2026-09-28 21:24 UTC via poll

System details

  • Machine: MSI PRO B650-P WIFI (MS-7D78) desktop
  • CPU: AMD Ryzen 7 7800X3D
  • GPU: AMD Radeon RX 7900 XT (amdgpu) + Raphael iGPU
  • Desktop: Hyprland 0.56.2, SDDM, Omarchy
  • At time of failure: Omarchy 4.0.1-1, kernel 7.1.9-arch1-2, NetworkManager 1.58.1-1
  • Now (after snapshot restore): Omarchy 4.0.0-1, kernel 7.1.8-arch1-3, NetworkManager 1.58.0-1

omarchy-debug output: https://gist.github.com/mrdeath5493/4138ac998193d5a4e03f5894feba7010

Network adapters

The internet path I actually use is the PCIe 10GbE card, not Wi-Fi.

InterfaceRolePCIHardwareDriverMACCurrent state
`enp7s0`Primary (PCIe 10G)07:00.0 [1d6a:04c0]Aquantia AQtion AQC113 10G (add-in PCIe)atlantic (fw ATL2FW 1050026)XX:XX:XX:XX:XX:01UP, 10G, DHCP 192.168.1.84 (after rollback)
`enp14s0`Onboard 2.5G, unused0e:00.0 [10ec:8125]Realtek RTL8125B (MSI PRO B650-P WIFI)r8169XX:XX:XX:XX:XX:02UP but NO-CARRIER (nothing plugged in)
(none)Onboard Wi-Fi0f:00.0 [14c3:0616]MediaTek MT7922mt7921e—hardware init failed; WIFI-HW: missing

NetworkManager profiles: Wired connection 2 → enp7s0, Wired connection 1 → enp14s0.

What's wrong?

After omarchy update 4.0.0 → 4.0.1 and reboot onto linux 7.1.9-arch1-2, the machine had no internet. I was not relying on Wi-Fi. The PCIe 10GbE AQC113 (enp7s0 / atlantic) failed as well.

On that boot:

  1. `atlantic` probed and renamed `eth0` → `enp7s0`, but never brought a link up. There is no atlantic: link change old 0 new 10000 for the entire 8-minute 7.1.9 session. On working 7.1.8 boots that event appears ~10s after probe.
  2. NetworkManager never started. Zero NetworkManager journal lines. systemd still printed Reached target Network / Reached target Network is Online.
  3. systemd deleted `dbus.socket/start` to break an ordering cycle involving a CIFS automount that After=network-online.target:
sockets.target: Found ordering cycle: dbus.socket/start after sysinit.target/start after systemd-binfmt.service/start after local-fs.target/start after mnt-nas-projects.automount/start after network-online.target/start after network.target/start after NetworkManager.service/start after basic.target/start after sockets.target/start - after dbus.socket
sockets.target: Job dbus.socket/start deleted to break ordering cycle starting with sockets.target/start
  1. CIFS to the NAS then failed with ENETUNREACH (cifs_mount failed w/return code = -101) — consistent with no usable IP on enp7s0.
  2. Onboard enp14s0 (RTL8125) was probed (RTL8125B … renamed from eth0) but has no cable, so it could not have been a fallback.

After restoring snapper snapshot 2 (back to 4.0.0 / 7.1.8), NetworkManager did start, enp7s0 got link change old 0 new 10000, and DHCP returned 192.168.1.84. The same NAS automount cycle is still present on 4.0.0, but systemd broke it by dropping systemd-binfmt / local-fs.target instead of dbus.socket, so NM came up.

Packages pulled in by the 4.0.1 update (2026-08-30 20:40 CDT)

linux              7.1.8.arch1-3 -> 7.1.9.arch1-2
networkmanager     1.58.0-1      -> 1.58.1-1
omarchy            4.0.0-1       -> 4.0.1-1
omarchy-settings   4.0.0-1       -> 4.0.1-1

Working vs failing boots (10GbE)

BootKernelenp7s0 (atlantic)NetworkManager
2026-08-27 (pre-update)7.1.8-arch1-3link change old 0 new 10000Started
2026-08-30 20:43 (post-update)7.1.9-arch1-2probed/renamed, no link eventnever started
2026-08-30 20:53 (snapshot 2 restore)7.1.8-arch1-3link change old 0 new 10000Started

Kernel log — 10GbE on the failed 7.1.9 boot

atlantic: Detect ATL2FW 1050026
atlantic 0000:07:00.0 enp7s0: renamed from eth0
# no further atlantic/enp7s0 link messages for the rest of the boot
CIFS: VFS: Error connecting to socket. Aborting operation.
CIFS: VFS: cifs_mount failed w/return code = -101

Onboard 2.5G on the same boot (no cable, so expected down):

r8169 0000:0e:00.0 eth0: RTL8125B, XX:XX:XX:XX:XX:02, XID 641, IRQ 118
r8169 0000:0e:00.0 enp14s0: renamed from eth0

Kernel log — 10GbE on a working 7.1.8 boot

atlantic: Detect ATL2FW 1050026
atlantic 0000:07:00.0 enp7s0: renamed from eth0   # or eth1
atlantic 0000:07:00.0 enp7s0: atlantic: link change old 0 new 10000

After rollback, NM:

device (enp7s0): carrier: link connected
policy: auto-activating connection 'Wired connection 2'
dhcp4 (enp7s0): state changed new lease, address=192.168.1.84
device (enp7s0): Activation: successful, device activated.
DEVICE   TYPE      STATE                   CONNECTION
enp7s0   ethernet  connected               Wired connection 2
lo       loopback  connected (externally)  lo
enp14s0  ethernet  unavailable             --

Also broken: onboard MT7922 Wi-Fi (secondary)

I was not using Wi-Fi for internet, but the onboard MT7922 failed independently on the same 7.1.9 boot (mt7921e: Failed to start patch / Failed to release patch semaphore / hardware init failed). Snapshot restore did not recover Wi-Fi (chip still wedged). Bluetooth on the same chip also failed (hci0: Failed to send wmt patch dwnld (-110)).

Steps to reproduce

  1. Omarchy 4.0.0-1, internet via PCIe AQC113 enp7s0 (atlantic), 10G link.
  2. Run omarchy update (4.0.1-1, linux 7.1.9-arch1-2, NetworkManager 1.58.1-1).
  3. Reboot.
  4. No internet: atlantic never reports 10G link; NetworkManager does not start; CIFS returns ENETUNREACH.

Expected

enp7s0 comes up at 10G and NetworkManager starts after the update, as it did on 4.0.0 / 7.1.8.

Notes

  • PCI retraining failed on empty AMD 600-series switch ports 06:08.0 / 06:09.0 happens on working and failing boots — not the cause.
  • The mnt-nas-projects.automount unit has After=network-online.target. That cycle exists on 4.0.0 too, but only the 4.0.1/7.1.9 boot dropped dbus.socket and skipped NetworkManager. Happy to test with that automount disabled.
  • Related historical Wi-Fi report: #1829 (older kernel/firmware pair). Not the iwd credential-drop case (#8996).
View raw text
### System details

- **Machine:** MSI PRO B650-P WIFI (MS-7D78) desktop
- **CPU:** AMD Ryzen 7 7800X3D
- **GPU:** AMD Radeon RX 7900 XT (amdgpu) + Raphael iGPU
- **Desktop:** Hyprland 0.56.2, SDDM, Omarchy
- **At time of failure:** Omarchy **4.0.1-1**, kernel **7.1.9-arch1-2**, NetworkManager **1.58.1-1**
- **Now (after snapshot restore):** Omarchy **4.0.0-1**, kernel **7.1.8-arch1-3**, NetworkManager **1.58.0-1**

`omarchy-debug` output: https://gist.github.com/mrdeath5493/4138ac998193d5a4e03f5894feba7010

### Network adapters

The internet path I actually use is the **PCIe 10GbE card**, not Wi-Fi.

| Interface | Role | PCI | Hardware | Driver | MAC | Current state |
|-----------|------|-----|----------|--------|-----|---------------|
| **`enp7s0`** | **Primary (PCIe 10G)** | `07:00.0` `[1d6a:04c0]` | Aquantia AQtion **AQC113** 10G (add-in PCIe) | `atlantic` (fw ATL2FW 1050026) | `XX:XX:XX:XX:XX:01` | UP, 10G, DHCP `192.168.1.84` (after rollback) |
| **`enp14s0`** | Onboard 2.5G, unused | `0e:00.0` `[10ec:8125]` | Realtek **RTL8125B** (MSI PRO B650-P WIFI) | `r8169` | `XX:XX:XX:XX:XX:02` | UP but **NO-CARRIER** (nothing plugged in) |
| *(none)* | Onboard Wi-Fi | `0f:00.0` `[14c3:0616]` | MediaTek **MT7922** | `mt7921e` | — | hardware init failed; `WIFI-HW: missing` |

NetworkManager profiles: `Wired connection 2` → `enp7s0`, `Wired connection 1` → `enp14s0`.

### What's wrong?

After `omarchy update` 4.0.0 → 4.0.1 and reboot onto **linux 7.1.9-arch1-2**, **the machine had no internet**. I was not relying on Wi-Fi. The PCIe 10GbE AQC113 (`enp7s0` / `atlantic`) failed as well.

On that boot:

1. **`atlantic` probed and renamed `eth0` → `enp7s0`, but never brought a link up.** There is no `atlantic: link change old 0 new 10000` for the entire 8-minute 7.1.9 session. On working 7.1.8 boots that event appears ~10s after probe.
2. **NetworkManager never started.** Zero `NetworkManager` journal lines. systemd still printed `Reached target Network` / `Reached target Network is Online`.
3. **systemd deleted `dbus.socket/start` to break an ordering cycle** involving a CIFS automount that `After=network-online.target`:

```
sockets.target: Found ordering cycle: dbus.socket/start after sysinit.target/start after systemd-binfmt.service/start after local-fs.target/start after mnt-nas-projects.automount/start after network-online.target/start after network.target/start after NetworkManager.service/start after basic.target/start after sockets.target/start - after dbus.socket
sockets.target: Job dbus.socket/start deleted to break ordering cycle starting with sockets.target/start
```

4. CIFS to the NAS then failed with **ENETUNREACH** (`cifs_mount failed w/return code = -101`) — consistent with no usable IP on `enp7s0`.
5. Onboard `enp14s0` (RTL8125) was probed (`RTL8125B … renamed from eth0`) but has no cable, so it could not have been a fallback.

After restoring snapper snapshot 2 (back to 4.0.0 / 7.1.8), NetworkManager **did** start, `enp7s0` got `link change old 0 new 10000`, and DHCP returned `192.168.1.84`. The same NAS automount cycle is still present on 4.0.0, but systemd broke it by dropping `systemd-binfmt` / `local-fs.target` instead of `dbus.socket`, so NM came up.

#### Packages pulled in by the 4.0.1 update (2026-08-30 20:40 CDT)

```
linux              7.1.8.arch1-3 -> 7.1.9.arch1-2
networkmanager     1.58.0-1      -> 1.58.1-1
omarchy            4.0.0-1       -> 4.0.1-1
omarchy-settings   4.0.0-1       -> 4.0.1-1
```

#### Working vs failing boots (10GbE)

| Boot | Kernel | `enp7s0` (`atlantic`) | NetworkManager |
|------|--------|------------------------|----------------|
| 2026-08-27 (pre-update) | 7.1.8-arch1-3 | `link change old 0 new 10000` | Started |
| 2026-08-30 20:43 (post-update) | **7.1.9-arch1-2** | probed/renamed, **no link event** | **never started** |
| 2026-08-30 20:53 (snapshot 2 restore) | 7.1.8-arch1-3 | `link change old 0 new 10000` | Started |

#### Kernel log — 10GbE on the failed 7.1.9 boot

```
atlantic: Detect ATL2FW 1050026
atlantic 0000:07:00.0 enp7s0: renamed from eth0
# no further atlantic/enp7s0 link messages for the rest of the boot
CIFS: VFS: Error connecting to socket. Aborting operation.
CIFS: VFS: cifs_mount failed w/return code = -101
```

Onboard 2.5G on the same boot (no cable, so expected down):

```
r8169 0000:0e:00.0 eth0: RTL8125B, XX:XX:XX:XX:XX:02, XID 641, IRQ 118
r8169 0000:0e:00.0 enp14s0: renamed from eth0
```

#### Kernel log — 10GbE on a working 7.1.8 boot

```
atlantic: Detect ATL2FW 1050026
atlantic 0000:07:00.0 enp7s0: renamed from eth0   # or eth1
atlantic 0000:07:00.0 enp7s0: atlantic: link change old 0 new 10000
```

After rollback, NM:

```
device (enp7s0): carrier: link connected
policy: auto-activating connection 'Wired connection 2'
dhcp4 (enp7s0): state changed new lease, address=192.168.1.84
device (enp7s0): Activation: successful, device activated.
```

```
DEVICE   TYPE      STATE                   CONNECTION
enp7s0   ethernet  connected               Wired connection 2
lo       loopback  connected (externally)  lo
enp14s0  ethernet  unavailable             --
```

### Also broken: onboard MT7922 Wi-Fi (secondary)

I was not using Wi-Fi for internet, but the onboard MT7922 failed independently on the same 7.1.9 boot (`mt7921e`: `Failed to start patch` / `Failed to release patch semaphore` / `hardware init failed`). Snapshot restore did **not** recover Wi-Fi (chip still wedged). Bluetooth on the same chip also failed (`hci0: Failed to send wmt patch dwnld (-110)`).

### Steps to reproduce

1. Omarchy 4.0.0-1, internet via PCIe AQC113 `enp7s0` (`atlantic`), 10G link.
2. Run `omarchy update` (4.0.1-1, linux 7.1.9-arch1-2, NetworkManager 1.58.1-1).
3. Reboot.
4. No internet: `atlantic` never reports 10G link; NetworkManager does not start; CIFS returns ENETUNREACH.

### Expected

`enp7s0` comes up at 10G and NetworkManager starts after the update, as it did on 4.0.0 / 7.1.8.

### Notes

- PCI `retraining failed` on empty AMD 600-series switch ports `06:08.0` / `06:09.0` happens on **working and failing** boots — not the cause.
- The `mnt-nas-projects.automount` unit has `After=network-online.target`. That cycle exists on 4.0.0 too, but only the 4.0.1/7.1.9 boot dropped `dbus.socket` and skipped NetworkManager. Happy to test with that automount disabled.
- Related historical Wi-Fi report: #1829 (older kernel/firmware pair). Not the iwd credential-drop case (#8996).

Validation

Not checked yet. A validator will restate this report as checkable claims, map it to the project's requirements and write a reproducer that the proof engine runs.

Linked

Not linked to any known issue, other report or change yet.

Decision

Decisions are recorded by our staff — or, where this project allows it, by agents: checked by a second, independent agent, or by a single agent where the project's policy says so. Any decision reached so far is shown below.

  • Out of scopeDecided by agents · verified · 2026-10-09 10:13 UTC