The troubleshooting method
Every space-specific troubleshooting guide on this site assumes you have worked through this page first. The point is to stop guessing: identify the symptom, then walk down the signal chain one layer at a time until you find the layer that is actually broken.
1. Before you touch anything
Ask three questions, in this order:
- Is there an audience in the room right now? If yes, you are not troubleshooting — you are doing damage control. Go to the "if something goes wrong during a show" section of the relevant operator manual, get the room to a presentable state, and do the real diagnosis afterwards.
- What changed? Almost every fault follows a change: a driver update, a moved cable, a new show file, a power event, someone borrowing a computer. Ask whoever was last in the room.
- Is it one channel or all channels? This single question splits the problem in half. One channel out points at a projector, decoder or single GPU output. All channels out points at the playback machine, the control system or power.
2. The layer ladder
Work from the content outwards. Do not skip a layer because you "know" it is fine.
| # | Layer | What to check | Reference |
|---|---|---|---|
| 1 | Content / show file | Is the right show or file loaded? Does another piece of content work? | Space operator manual |
| 2 | Playback machine | Is the node powered and logged in? Is the render app running? Are all nodes in sync? | The space's cluster/computer page |
| 3 | Signal distribution | Is the encoder/decoder pair linked? Is the matrix routed correctly? | Barco ENA-340, Matrixes |
| 4 | Display device | Is the projector on, unshuttered, on the right input, and free of alarms? | The space's projector page |
| 5 | Control system | Does the Showmaster tech panel respond? Is the iPad connected? | Showmaster and Medialon |
| 6 | Power | Is the PDU outlet actually on? Any UPS alarm? | Power (PDU/UPS) |
| 7 | Network | Can you reach the device by DNS name and by IP? | Common web systems |
A quick way to jump layers: if the projector shows its own on-screen menu but no image, layers 4–6 are fine and the fault is at layer 1–3.
3. Status pages to check first
These give you a facility-wide view in seconds and are almost always faster than walking to the room:
- Barco decoder/encoder status — barco-status.c.itn.liu.se (internal: barco-status-int.c.itn.liu.se). A decoder missing from this page is a signal-chain fault, not a projector fault.
- Showmaster tech panels — showmaster1.c.itn.liu.se (Dome) and showmaster2.c.itn.liu.se (VR-Arena), tab TechPanel.
- PDU web interfaces — linked from each space's PDU reference page.
- Speedtest — speedtest.c.itn.liu.se when something "feels slow".
The full list is on Common web systems.
4. Restarting, in the right order
Restarting is a legitimate fix, but the order matters and a wrong order can make things worse — most often by leaving a cluster out of genlock or leaving a decoder bound to a stream that no longer exists.
General rules that apply everywhere at C:
- Restart the least disruptive thing first. Application → operating system → device power cycle. Never start with the PDU.
- On a render cluster, restart the sync master last. The other nodes need something to lock to while they come back. For the Dome's ATLAS cluster the sync master is ATLAS-01 — see ATLAS cluster.
- Prefer the Showmaster tech panel to the PDU. The panel sequences equipment properly; the PDU just cuts power.
- Give projectors their cooldown. Cutting power to a lamp/laser projector mid-cooldown shortens its life and can trigger a fault on the next start.
To power cycle from a PDU: open the PDU page, click Login (no credentials needed), open the Control Panel tab, select the device by name and press Reset — power drops and returns after the configured reset duration, normally 60 seconds. Full instructions on Simple instructions.
PDU outlets are labelled per device, but labels drift as equipment is moved. Confirm the device name on the outlet against the space's PDU reference page before pressing anything. Cutting the wrong outlet in the Dome can take down a whole cluster mid-show.
5. When to swap hardware
Swap rather than repair when all of the following are true: the fault follows the device (not the slot or cable), you have a tested spare, and the space is needed soon.
- Spare PDUs are listed on Spare PDU units.
- Encoders and decoders are interchangeable within the ENA-340 fleet; note the unit number from the status page before swapping so the routing can be reapplied.
- Projectors are not field-swappable in the Dome, VR-Arena or Decision Arena — the warp and blend calibration is per-position. Escalate instead.
Always record what you swapped, from where, and what the replaced unit's symptom was.
6. When to stop and escalate
Escalate immediately, without further experimentation, if:
- There is smoke, a burning smell, water, or a tripped breaker.
- A UPS is alarming and you do not know what it feeds.
- The fault is inside a projector's optical path or requires opening a projector.
- A show starts in under an hour and you do not yet know which layer is broken.
- You would have to change calibration, warping or blending to "fix" it.
7. Close the loop
Before you walk away:
- Put the room back into its standby state (see the operator manual for that space).
- Verify the fix with real content, not a test pattern.
- Write down what happened. If the symptom is not already in the space's troubleshooting page, add it — that page is the reason the next person will not spend an hour on this.