chore: release v0.22.9

wiki/Troubleshooting.md

@@ -2,57 +2,77 @@
 
 ## Where's the log file?
 
-OpsLog writes a diagnostic log. The path is shown in Settings (and can be opened
-from there). When reporting a problem, include the relevant lines — they're
-prefixed by subsystem (`Flex:`, `icom net:`, `icom scope`, `antgenius:`,
-`cluster:`, `qslmgr:log`, `audio:`…).
+OpsLog writes a diagnostic log. The path is shown in Settings (and can be
+opened from there). When reporting a problem, include the relevant lines —
+they're prefixed by subsystem (`cat:`, `Flex:`, `icom net:`, `antgenius:`,
+`cluster:`, `qslmgr:log`, `audio:`…). With an SMTP server configured, **Help →
+Send log to F4BPO** e-mails the logs straight to the developer.
+
+A crash no longer leaves a blank window: the error is shown on screen,
+selectable, with a copy button.
 
 ## CAT
 
 - **Radio not connecting** — check the backend fields (COM port + baud for USB;
-  IP/credentials for network). OpsLog connects non-blocking, so it won't freeze —
-  watch the CAT status.
-- **Icom: no CAT at all over USB** — the CI-V address must match the radio; pick
-  your **model** from the dropdown (it sets the address) and set *CI-V USB Echo
-  Back* **OFF** on the rig.
-- **Icom scope blank on an IC-7300** — this was a bug where the scope decoder was
-  chosen from the configured CI-V address; it now uses the model detected from the
-  CI-V ID, so a single-scope rig run at address `0x98` decodes correctly. Update
-  to the latest build.
-- **Icom attenuator button does nothing** — the dB steps are model-specific; make
-  sure the right model is selected (IC-7300 = 20 dB, IC-7610 = 6/12/18 dB).
+  IP/credentials for network). The CAT log line carries the error explaining a
+  disconnection.
+- **"Serial port busy" in the log** — another program holds the COM port
+  (WSJT-X, MSHV, another logger configured for direct CAT). Only one program
+  can own the port: let OpsLog hold it and point the others at **Hamlib NET
+  rigctl 127.0.0.1:4532** (Settings → CAT → Share CAT). Two masters on one
+  CI-V bus also cause garbled frames and erratic behaviour.
+- **The radio transmits as soon as OpsLog connects** — your interface reads DTR
+  or RTS as PTT. Xiegu drops the lines automatically; on Yaesu and Kenwood
+  enable the *lower DTR/RTS on connect* option. (It is off by default because
+  other interfaces stop transmitting with the lines low.)
+- **WSJT-X won't key a Xiegu G90 through OmniRig** — a G90 ignores the CI-V PTT
+  command. Use the native Xiegu backend (which keys RTS/DTR) + rigctl sharing,
+  or set the PTT line inside OmniRig. See [[Kenwood and Xiegu]].
+- **OmniRig frequency frozen / wrong VFO** — set **VFO to read** (A or B) in
+  the OmniRig backend settings; some rig files declare the wrong VFO.
+- **"OmniRig not found" with OmniRig running** — one of the two runs as
+  administrator and the other doesn't. Start both the same way.
+- **Icom: no CAT at all over USB** — the CI-V address must match the radio;
+  pick your **model** from the dropdown (it sets the address) and set *CI-V USB
+  Echo Back* **OFF** on the rig.
+- **A radio that answers oddly** — enable the **protocol trace** (Settings →
+  CAT): every frame to and from the rig lands in the log as hex, which turns a
+  guess into a diagnosis.
 
 ## Remote Icom
 
 - **Login rejected** → wrong Network User1 name/password.
 - **Can't bind local port** → the Icom **Remote Utility is still running** —
   close it.
-- **Link drops with the scope on** → receive-side retransmit should hold it; if
-  not, check network quality and the `icom net:` log lines. See
+- **Frequency frozen after another program took the CI-V session** — OpsLog
+  reconnects on its own; if it doesn't, check the `icom net:` log lines. See
   [[Remote Icom over the Internet]].
 
 ## FlexRadio
 
-- **A control reads 0 / stale at startup** — some SmartSDR status fields use short
-  names (e.g. the TX filter is reported as `lo`/`hi`, not `filter_low/high`). If a
-  value never populates, capture the `Flex: … status` log line and report it.
-- **Antenna wrong band on a spot click** — fixed: `SetFrequency`/`SetMode` update
-  the slice cache immediately so the panel and the Ultrabeam follow don't chase a
-  stale frequency.
+- **A control reads 0 / stale at startup** — some SmartSDR status fields use
+  short names. If a value never populates, capture the `Flex: … status` log
+  line and report it.
+- **SmartSDR CAT keeps disconnecting while OpsLog is open** — fixed: OpsLog now
+  binds only to the actual SmartSDR/Maestro GUI client. Update.
+- **"Interlock is preventing transmission" with a SteppIR** — see the tunable
+  range setting in [[Maps and Antennas]].
 
 ## Multi-op live status PHP
 
 `Connection refused` from the PHP page is almost always MySQL `bind-address =
-127.0.0.1` or a firewall/grant issue — see
-[[Multi-Operator Live Status]].
+127.0.0.1` or a firewall/grant issue — see [[Multi-Operator Live Status]].
 
 ## DVK / PTT
 
-- **Test PTT does nothing on CAT/OmniRig** — the key is held ~1.5 s so the async
-  OmniRig write is actually sent; if it still doesn't key, the rig's OmniRig
-  profile may not expose CAT TX keying — use RTS/DTR or VOX.
+- **Test PTT does nothing on CAT/OmniRig** — the key is held ~1.5 s so the
+  async OmniRig write is actually sent; if it still doesn't key, the rig's
+  OmniRig profile may not expose CAT TX keying — use RTS/DTR or VOX.
+- **A voice message transmits almost nothing** — raise the message level
+  (Settings → Audio) and check the rig's rear-input setting (FTDX10: SSB MOD
+  SOURCE = REAR).
 
 ## Still stuck?
 
-Open an issue with: what you did, what happened, your radio/model, and the
-relevant log lines.
+Open an issue (or *Send log to F4BPO*) with: what you did, what happened, your
+radio/model, and the relevant log lines.