omarchy-pkgs/bin/timers

#!/bin/bash
# Show the release timers: schedule, last run and its result, what is queued,
# and what is running right now.
#
# Through bin/repo this forwards to the repository host, so the build box's
# state is one command away from any machine.

set -e

BUILD_ROOT=$(realpath "${BASH_SOURCE[0]%/*}/..")
source "$BUILD_ROOT/helpers/message-helpers.sh"
source "$BUILD_ROOT/helpers/paths.sh"

STATE_DIR="${OMARCHY_STATE_DIR:-/root/.state}"

if [[ "${1:-}" == "-h" || "${1:-}" == "--help" ]]; then
  echo "Usage: $0"
  echo ""
  echo "Reports each omarchy release timer's schedule, last run and result,"
  echo "any queued builds, and whether a release is running now."
  exit 0
fi

print_header "Omarchy Release Timers"

if ! command -v systemctl >/dev/null 2>&1; then
  print_warning "systemctl not found — timers live on the repository host"
  echo "Run this against the host: bin/repo timers (with OMARCHY_REPO_HOST or .repo-host set)"
  exit 0
fi

# Discover units from the repo so this never drifts from what setup installs.
UNITS=()
for unit in "$BUILD_ROOT"/systemd/*.timer; do
  [[ -e "$unit" ]] || continue
  UNITS+=("$(basename "$unit" .timer)")
done

if [[ ${#UNITS[@]} -eq 0 ]]; then
  print_warning "No timer units found in $BUILD_ROOT/systemd"
  exit 0
fi

# --- schedule ----------------------------------------------------------------

print_info "Schedule"
if ! systemctl list-timers --all 'omarchy-*' --no-pager 2>/dev/null | sed '/^$/d;s/^/  /'; then
  echo "  (could not read timers)"
fi
echo ""

# --- per-unit state ----------------------------------------------------------

print_info "Units"
for unit in "${UNITS[@]}"; do
  # systemctl prints "not-found" AND exits nonzero, so a || fallback would
  # append a second line and wreck the alignment. Take its output as-is.
  enabled=$(systemctl is-enabled "$unit.timer" 2>/dev/null || true)
  enabled=${enabled:-unknown}
  service_state=$(systemctl is-active "$unit.service" 2>/dev/null || true)
  result=$(systemctl show "$unit.service" -p Result --value 2>/dev/null || true)
  exit_status=$(systemctl show "$unit.service" -p ExecMainStatus --value 2>/dev/null || true)
  finished=$(systemctl show "$unit.service" -p ExecMainExitTimestamp --value 2>/dev/null || true)

  case "$enabled" in
  enabled) mark="✓" ;;
  not-found | unknown) mark="✗" ;;
  *) mark="⚠" ;;
  esac

  printf '  %s %-28s timer:%-14s' "$mark" "$unit" "$enabled"

  if [[ "$service_state" == "activating" || "$service_state" == "active" ]]; then
    printf 'RUNNING NOW\n'
  elif [[ -z "$finished" ]]; then
    printf 'never run\n'
  elif [[ "$result" == "success" && "$exit_status" == "0" ]]; then
    printf 'last: ok (%s)\n' "$finished"
  else
    printf 'last: FAILED [%s/exit %s] (%s)\n' "${result:-unknown}" "${exit_status:-?}" "$finished"
  fi
done
echo ""

# --- queued work -------------------------------------------------------------

print_info "Queued builds (state files in $STATE_DIR)"
queued=false
for channel in edge rc stable; do
  state_file="$STATE_DIR/.sync-needed-$channel"
  if [[ -f "$state_file" ]]; then
    queued=true
    since=$(date -r "$state_file" '+%Y-%m-%d %H:%M:%S' 2>/dev/null || echo "unknown")
    printf '  • %-7s queued since %s\n' "$channel" "$since"
  fi
done
[[ "$queued" == false ]] && echo "  (nothing queued — all channels up to date)"
echo ""

# A channel in backoff looks identical to an idle one from the outside, so
# say so plainly: it is queued but deliberately not being retried yet.
paused=false
for channel in edge rc stable; do
  fail_file="$STATE_DIR/.build-failed-$channel"
  [[ -f "$fail_file" ]] || continue
  if [[ "$paused" == false ]]; then
    print_error "Failing builds (backoff active)"
    paused=true
  fi
  FAILURE_COUNT=0 FAILURE_AT=0 FAILURE_FINGERPRINT=""
  # shellcheck disable=SC1090
  source "$fail_file" 2>/dev/null || true
  delay=600
  for ((i = 1; i < FAILURE_COUNT; i++)); do
    delay=$((delay * 2))
    ((delay >= 21600)) && { delay=21600; break; }
  done
  retry_at=$((FAILURE_AT + delay))
  now=$(date +%s)
  if ((now < retry_at)); then
    when="retries at $(date -d "@$retry_at" '+%H:%M:%S' 2>/dev/null || echo "+$((retry_at - now))s")"
  else
    when="retries on the next tick"
  fi
  printf '  ✗ %-7s %s consecutive failure(s), %s\n' "$channel" "$FAILURE_COUNT" "$when"
  printf '      last attempt %s on commit %s\n' \
    "$(date -d "@$FAILURE_AT" '+%Y-%m-%d %H:%M:%S' 2>/dev/null || echo "$FAILURE_AT")" \
    "${FAILURE_FINGERPRINT:0:12}"
done
if [[ "$paused" == true ]]; then
  echo "  Any new commit clears the backoff; or: rm $STATE_DIR/.build-failed-<channel>"
  echo ""
fi

# --- release lock ------------------------------------------------------------

print_info "Release lock"
lock_file="$REPO_ROOT/.release.lock"
if [[ -s "$lock_file" ]]; then
  holder=$(tail -1 "$lock_file" 2>/dev/null)
  # A recorded holder whose process is gone is a leftover, not a live release.
  pid=$(sed -n 's/^pid \([0-9]\+\).*/\1/p' <<<"$holder")
  if [[ -n "$pid" ]] && kill -0 "$pid" 2>/dev/null; then
    print_warning "HELD — $holder"
  else
    echo "  free (last holder: $holder)"
  fi
else
  echo "  free"
fi
echo ""

# --- recent failures ---------------------------------------------------------

failed=$(systemctl list-units --failed --no-legend 'omarchy-*' 2>/dev/null | sed 's/^/  /')
if [[ -n "$failed" ]]; then
  print_error "Failed units"
  echo "$failed"
  echo ""
  echo "  Logs: journalctl -u <unit> -n 50"
fi

print_info "Full logs: journalctl -u omarchy-auto-release-<channel> -n 50"