Block a user
9e5868b839
fix: raise MySQL pool 8->16 — 8 starved the instance's own live-sync polling (revision + grid refresh + live_status + chat), so the grid stopped showing other operators' QSOs; 16 avoids the stall and still fits ~15 ops under max_connections=300
40d0ca57c3
diag: log slow QSO inserts (>2s) with call/band/mode; rotate the app log to .1 at 10MB instead of deleting it, so the previous session's diagnostics survive
aa537f9eea
fix: log returns as soon as the QSO row is inserted; award_refs/recording/uploads/eQSL now run off the critical path (both manual and UDP) — a busy multi-op MySQL no longer freezes the entry form ~40s waiting on the award_refs UPDATE
ea1bd2a66d
feat: sync the host CW keyer speed (wkWpm) to the FlexRadio's actual CW speed — including changes made on the radio/SmartSDR, not just the panel slider — so CW-length estimates (auto-call gap) match reality when using Flex CWX
1f54656256
fix: <LOGQSO> logs immediately at its position instead of waiting the full CW-duration estimate first (buffered keyers keep sending; logging never interrupts them) — was delaying the log up to ~10s; only the final segment still waits, for auto-call timing
54dd109288
docs: changelog entry for 0.20.7 (multi-op MySQL connection fix, By-Continent stats total)
2c1d7235f0
fix: shrink MySQL connection pool (50->8) so a multi-op event doesn't exhaust the shared server's max_connections (Error 1040: Too many connections), which was silently failing the on-air widget/award_refs/grid reads and making the UI look frozen
adf9844d1b
fix: stats By Continent donut totals now match the QSO count (bucket blank/unknown continents as 'Unknown' instead of dropping them; keep the Continents metric counting real continents only)
311ea8341f
docs: changelog entry for 0.20.6 (amp card in Flex panel, faster CAT+MySQL startup, no lost QSOs, instant stations-on-air, LOGQSO position, relay fixes)
a2958d458e
fix: 'stations on air' widget updates instantly after a QSO (UDP/FT8 path now emits qso:logged; publishLiveStatus emits livestatus:updated so the widget re-reads exactly when the shared row changes) — was lagging ~15-45s behind the ON-AIR badge
d9b7e48e83
fix: UDP-logged QSOs fall back to the offline outbox on DB failure (were silently lost); speed up MySQL connect (batch migration checks into 1 query, connect direct-to-DB first) — was ~40s on a high-latency link
3e9ebdb89a
fix: serialise UDP QSO logging (lookup+dedup+insert) so multi-stream MSHV (4 QSOs at once) queues instead of firing concurrent lookups/DB writes that got 'too many requests'
be66ac1e19
fix: auto-call gap respected again (wkSend already waits for the send; runAutoCall no longer double-counts it, so a 4s gap is 4s not ~13s)
6aab4a6989
fix: start CAT before the MySQL logbook connect — Flex/CAT no longer waits ~30s behind a slow remote-MySQL dial at launch
9c62bf0152
fix: cleanly stop CAT (and release relay handles) on shutdown — FlexRadio kept our API client registered, making the next launch take ~30s to reconnect
408791ddf5
fix: <LOGQSO> now logs at its position in the macro (splits the macro there), and ESC/Stop before it completes cancels the pending log