ERRORS - West TN Linked Digital Communications System

Hidden troubleshooting page — public but intentionally unlinked.
This page contains only redacted timing/state metadata. It does not display passwords, tokens, passphrases, IPSC authentication keys, Authorization headers, literal IP addresses, or raw DMR voice payloads.

After reproducing a problem, press Refresh Now, then Copy Report and paste the entire box back into ChatGPT. The diagnostic code is a quick classification, while the evidence lines let us confirm or reject that classification.

Test	What it observes	Example evidence	What that would prove / disprove
A. Hotspot/HBlink stream duration	Recent EXTERNAL-ACCESS stream start/end events copied from the ACTIVE worker to WTN01.	`stream_end duration=3.8`	If HBLink itself records a ~3–4 second end while the hotspot PTT remains held, the failure is upstream/on the worker side. It is not the 119-second limit.
B. WTN01 HBP→IPSC starvation	Filtered ipsc2hbpc journal signatures on WTN01.	`stream timeout ... 34 consecutive silence bursts ... VOICE_TERM`	Shows WTN01 received more than ~2.04 seconds of continuous missing HBP voice and intentionally terminated the Motorola-side call. Shorter inter-node jitter is concealed and must not unkey the repeater. This points between the ACTIVE worker and translator.
C. Long source call	Whether ACTIVE-worker telemetry continues for at least 10 seconds.	`OK-SOURCE-LONG`	Disproves a 3–4 second source/HBlink timeout. If the repeater still drops, focus on WTN01 translator/IPSC/repeater handling.
D. Service stability	Director, dashboard, Nginx and every ipsc2hbpc instance state/restart count.	`state=failed` or increasing `restarts`	Shows a service restart/crash could be ending calls independently of hotspot PTT.
E. 119-second bridge policy	Completed stream durations reported by the bridge.	`duration≈119`	Confirms the intended hard bridge limit. A 3–4 second failure cannot be explained by this policy unless evidence actually reports an early bridge-generated end.
F. TGIF isolation	NODEv223 runs TGIF in a separate HBLink sidecar instead of as a member of the core TS2 conference bridge.	`E-TGIF-SIDECAR-RETURN`	Shows TGIF return traffic was caught inside the isolated edge. The main hotspot↔repeater call state is not modified by the TGIF process.
G. Hotspot HBP session/end reason	Recent EXTERNAL-ACCESS events and retained historical signatures identify real terminators, session closes, source timeouts, and stream replacements.	`E-HOTSPOT-VOICE-TERMINATOR`, `E-HOTSPOT-DMRD-STOPPED`, or `E-HBP-STREAM-RESTART`	Shows whether the main core itself is losing/replacing the hotspot stream after the TGIF topology rollback.
H. Core regression topology	The worker release contract requires the core TS2 conference bridge to contain only INTERNAL-REPEATERS and EXTERNAL-ACCESS. TGIF must be absent.	`NODEv223 core=v212 topology`	Proves the first NODEv213 topology delta has been removed rather than masked by another packet-flow compatibility patch.

Test

What it observes

Example evidence

What that would prove / disprove

A. Hotspot/HBlink stream duration

Recent EXTERNAL-ACCESS stream start/end events copied from the ACTIVE worker to WTN01.

stream_end duration=3.8

If HBLink itself records a ~3–4 second end while the hotspot PTT remains held, the failure is upstream/on the worker side. It is not the 119-second limit.

B. WTN01 HBP→IPSC starvation

Filtered ipsc2hbpc journal signatures on WTN01.

stream timeout ... 34 consecutive silence bursts ... VOICE_TERM

Shows WTN01 received more than ~2.04 seconds of continuous missing HBP voice and intentionally terminated the Motorola-side call. Shorter inter-node jitter is concealed and must not unkey the repeater. This points between the ACTIVE worker and translator.

C. Long source call

Whether ACTIVE-worker telemetry continues for at least 10 seconds.

OK-SOURCE-LONG

Disproves a 3–4 second source/HBlink timeout. If the repeater still drops, focus on WTN01 translator/IPSC/repeater handling.

D. Service stability

Director, dashboard, Nginx and every ipsc2hbpc instance state/restart count.

state=failed or increasing restarts

Shows a service restart/crash could be ending calls independently of hotspot PTT.

E. 119-second bridge policy

Completed stream durations reported by the bridge.

duration≈119

Confirms the intended hard bridge limit. A 3–4 second failure cannot be explained by this policy unless evidence actually reports an early bridge-generated end.

F. TGIF isolation

NODEv223 runs TGIF in a separate HBLink sidecar instead of as a member of the core TS2 conference bridge.

E-TGIF-SIDECAR-RETURN

Shows TGIF return traffic was caught inside the isolated edge. The main hotspot↔repeater call state is not modified by the TGIF process.

G. Hotspot HBP session/end reason

Recent EXTERNAL-ACCESS events and retained historical signatures identify real terminators, session closes, source timeouts, and stream replacements.

E-HOTSPOT-VOICE-TERMINATOR, E-HOTSPOT-DMRD-STOPPED, or E-HBP-STREAM-RESTART

Shows whether the main core itself is losing/replacing the hotspot stream after the TGIF topology rollback.

H. Core regression topology

The worker release contract requires the core TS2 conference bridge to contain only INTERNAL-REPEATERS and EXTERNAL-ACCESS. TGIF must be absent.

NODEv223 core=v212 topology

Proves the first NODEv213 topology delta has been removed rather than masked by another packet-flow compatibility patch.

ERRORS - West TN Linked Digital Communications System

Copy/paste diagnostic report

Current interpretation

What these tests reveal

Recent bridge events

WTN01 translator signatures

Service state