avi/Orion
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Orion vault — clean initial history

Browse source 
Knowledge vault (Orion/PARA) migrated from the pre-Orion 484vault on
2026-10-01. Deliberately orphaned: prior history contained a plaintext
password and stays local-only on branch archive/pre-boilerplate-history.
Secrets and live Hermes state are gitignored.
This commit is contained in:
Avi 2026-10-02 08:34:04 -05:00
commit a66996ac10
233 changed files with 103810 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

28
2. Areas/Home Personal/knowledge/monitor-not-detected-framework-hdmi-card-latch.md Normal file
View file

@ -0,0 +1,28 @@
# External monitor not detected after projector swap (Framework HDMI card latch)
**Date:** 2026-09-11 · **Machine:** Framework Laptop (Omarchy, i915, Hyprland)
## Symptom
LG 32MR50C-B monitor plugged into the Framework HDMI expansion card showed nothing. `hyprctl monitors` listed only eDP-1. Kernel showed **all** `/sys/class/drm/card1-DP-*/status` = `disconnected` with zero AUX/hotplug log lines even after replugging cable, reseating the card, and power-cycling the monitor.
## Diagnosis path (do in this order — saves an hour)
1. `hyprctl monitors` — is the output even listed?
2. `cat /sys/class/drm/card1-DP-*/status` — **if the kernel says disconnected, stop all compositor-level debugging; it's hardware.**
3. `journalctl -k | grep -iE "dp|hpd|aux|over-current"` — no hotplug event = no cable-detect signal reaching GPU.
## Root cause
Yesterday a misbehaving Epson projector (bad EDID negotiation) was plugged into the **same** HDMI expansion card and latched its port controller so it stopped asserting cable-detect entirely. A normal reboot did not clear it.
## Fixes (in order)
1. Full power-off reset: shut down → unplug charger → hold power button 15s → wait 30s → boot with monitor connected.
2. **Swap in a different HDMI expansion card (power off first)** — this is what actually worked. Framework cards present to Linux as `DP-1..4`; the HDMI card owns `DP-1`.
3. If replacement card also dead: suspect cable, monitor input latch, or the slot.
## Update 2026-09-22 incident (resolved 2026-09-25)
After a desk move the LG FHD HDMI went fully dead again (all card1-DP-* `disconnected`, zero hotplug). Card swap + hard reset did NOT revive it same-day, but a plain reboot the next morning brought it back at 1920x1080@60 on the swapped card (slot moved to usb 3-1). **Lesson: the latched-card state CAN clear on a cold boot once standby power fully drains overnight — try one overnight-off + reboot cycle before escalating to cable/monitor replacement.** Forced DP-1 mode rule remains in `~/.config/hypr/monitors.lua`.
## Notes
- Framework officially supports hot-swapping cards, but with a suspected latched controller do it powered off, or the new card inherits the bad state and gives a false negative.
- Hyprland config (`hyprmoncfg-monitors.lua`) forces DP-1 to 1920x1080@59.94 — projector workaround that is also the LG's native mode; harmless for both.
- Transient `over-current condition` on usb3 ports during reseating was the phone cable (flaky, error -71), unrelated to the display.
- Related: [[Omarchy Commands]]