pstn-trunk: drop redundant container restart from update-in-place path

install_pstn-trunk's "update" branch unconditionally restarted the
asterisk container, then immediately called _pstn_ensure_live_includes,
which already patches the live pjsip.conf/extensions.conf #include lines
and reloads res_pjsip/dialplan — the restart accomplished nothing the
reload doesn't, while dropping any calls already in progress. Confirmed
live as a real double-restart when this runs chained from
asterisk-digital-ocean.sh, which had just rebuilt/restarted the same
container moments earlier.

Fresh installs keep their existing (still restart-first) apply step
unchanged — only "update in place" loses the redundant restart.
This commit is contained in:
Claude
2026-07-24 14:24:45 +00:00
parent 261ed3dc74
commit dd25782ef1
+17 -6
View File
@@ -1458,9 +1458,11 @@ install_pstn-trunk() {
echo "[DRY-RUN] pstn-permissions.conf, pstn-limits.conf, or the kill-switch trip state)"
echo "[DRY-RUN] instead of a fresh install if already configured; the international-calling"
echo "[DRY-RUN] review/change question is still asked every run either way"
echo "[DRY-RUN] Would restart the asterisk container to apply, AND directly patch the live"
echo "[DRY-RUN] pjsip.conf/extensions.conf with the #include lines regardless — Easy Asterisk"
echo "[DRY-RUN] only regenerates those files if they don't already exist, so an existing"
echo "[DRY-RUN] Fresh install: would prompt to restart the asterisk container to apply."
echo "[DRY-RUN] Update in place: applies live instead — no restart, no dropped calls."
echo "[DRY-RUN] Either way, directly patches the live pjsip.conf/extensions.conf with the"
echo "[DRY-RUN] #include lines and reloads res_pjsip/dialplan — Easy Asterisk only"
echo "[DRY-RUN] regenerates those files if they don't already exist, so an existing"
echo "[DRY-RUN] install (the common case) would otherwise never actually load the trunk config"
return 0
fi
@@ -1528,10 +1530,19 @@ install_pstn-trunk() {
"$TRUNK_SERVER" "${TRUNK_SERVER_IPS:-}" "$TRUNK_DID" \
"${RING_EXTS:-}" "${NTFY_URL:-}" "${RATE_PER_MIN:-0.01}" \
"${MONTH_THRESHOLD:-10}" "${BURST_THRESHOLD:-10}" "${PROVIDER_NAME:-unknown}" "${MAX_MONTHLY_SPEND:-0}" "${CONTAINER_NAME:-easy-asterisk-do}" || return 1
( cd "$EA_DIR" && docker compose restart asterisk ) \
&& log_success "Updated — settings unchanged (server $TRUNK_SERVER, DID $TRUNK_DID, ring exts: $RING_EXTS)." \
|| log_warning "Restart failed — check: docker compose -f $EA_DIR/docker-compose.yml logs asterisk"
# No container restart here — _pstn_ensure_live_includes's
# module/dialplan reload below already applies everything
# _pstn_apply_settings just wrote (trunk pjsip/dialplan
# includes, the group-ring script) live, without dropping
# any calls already in progress. A full restart doesn't
# accomplish anything the reload doesn't, and this is the
# "update in place" path — it shouldn't be more disruptive
# than it has to be. Confirmed live: this used to restart
# unconditionally here, redundant with (and disruptive on
# top of) the rebuild+restart asterisk[-digital-ocean].sh
# already just ran when this runs chained from there.
_pstn_ensure_live_includes "$ASTERISK_DIR" "$CONTAINER_NAME"
log_success "Updated — settings unchanged (server $TRUNK_SERVER, DID $TRUNK_DID, ring exts: $RING_EXTS)."
log_info "pstn-permissions.conf and pstn-limits.conf were NOT touched — edit them"
log_info "directly, via the Security Dashboard, or choose FRESH reinstall to reset them."
# Always asked, every run, update mode included — see