fix(ha/cluster-init): fix DRBD metadata skip condition and StandAlone grace period

The metadata creation check used grep -q "UpToDate" which matches when the
resource is DOWN — drbdadm dstate reads metadata directly and returns just
"UpToDate" (no slash) for a down-but-previously-synced resource.  This caused
metadata creation to be silently skipped, leaving mismatched UUIDs from an
interrupted sync which cause DRBD to go WFConnection→StandAlone immediately.

Fix: require exact "UpToDate/UpToDate" (with slash, only possible when the
resource is UP and both disks are confirmed synced) before skipping.

Also give the StandAlone detection a 5-iteration (15 s) grace period before
dying, to avoid false-positive on the first few ticks while the peer is still
coming up.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

scripts/ha/cluster-init.sh

@@ -215,14 +215,19 @@ log "Initialising DRBD metadata on $NODE1..."
 # "stdin not a TTY, not waiting for confirmation" → exit 20.
 # Calling drbdmeta --force directly bypasses the exclusive-open confirmation on
 # both steps without needing a TTY, regardless of whether the device is busy.
-if ! drbdadm dstate ha-data 2>/dev/null | grep -q "UpToDate"; then
+# Skip metadata creation only if DRBD is UP and fully synced (UpToDate/UpToDate).
+# When the resource is down, drbdadm dstate reads metadata and returns just
+# "UpToDate" (no slash) — that must not be treated as "already synced".
+# Mismatched UUIDs from an interrupted sync cause instant WFConnection→StandAlone,
+# so we always recreate metadata unless the sync is genuinely complete.
+if [[ "$(drbdadm dstate ha-data 2>/dev/null)" != "UpToDate/UpToDate" ]]; then
   UUID1=$(_rand_uuid)
   drbdmeta --force 0 v08 "${DRBD_DISK}" internal create-md
   drbdmeta --force 0 v08 "${DRBD_DISK}" internal write-dev-uuid "$UUID1"
 fi
 
 log "Initialising DRBD metadata on $NODE2..."
-if ! n2_ssh "drbdadm dstate ha-data 2>/dev/null | grep -q UpToDate" 2>/dev/null; then
+if [[ "$(n2_ssh "drbdadm dstate ha-data 2>/dev/null" 2>/dev/null)" != "UpToDate/UpToDate" ]]; then
   UUID2=$(n2_ssh "cat /proc/sys/kernel/random/uuid 2>/dev/null | tr -d '-' | cut -c1-16 | tr '[:lower:]' '[:upper:]'")
   n2_ssh "drbdmeta --force 0 v08 ${DRBD_DISK} internal create-md"
   n2_ssh "drbdmeta --force 0 v08 ${DRBD_DISK} internal write-dev-uuid ${UUID2}"
@@ -259,9 +264,9 @@ while true; do
   _spd=$(grep -oE "speed:[[:space:]]+[0-9,]+" /proc/drbd 2>/dev/null | grep -oE "[0-9,]+$" | head -1 || echo "")
   _sync_iter=$(( _sync_iter + 1 ))
   _sc="${_sync_chars[$_sync_iter % 4]}"
-  if [[ "$_cs" == "StandAlone" ]]; then
+  if [[ "$_cs" == "StandAlone" && $_sync_iter -gt 5 ]]; then
     printf "\r%-80s\r" ""
-    die "DRBD is StandAlone — peer connection lost (dstate: $_dstate). " \
+    die "DRBD is StandAlone after 15 s — peer connection lost (dstate: $_dstate). " \
         "Check corosync/network and re-run cluster-init."
   elif [[ -n "$_pct" ]]; then
     printf "\r  [%s] syncing: %s%% done — ETA %s @ %s K/s     " \