argus/src/sys/debug/scripts/03_wait_ready.sh

85 lines
2.2 KiB
Bash
Executable File

#!/usr/bin/env bash
set -euo pipefail
# shellcheck source=common.sh
source "$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)/common.sh"
ensure_env_file
ensure_paths_defined
service_id() {
compose ps -q "$1"
}
wait_http() {
local url="$1"; local attempts="${2:-120}"; local i=1
while (( i <= attempts )); do
if curl -fsS "$url" >/dev/null 2>&1; then
return 0
fi
echo "[..] waiting $url ($i/$attempts)"
sleep 5
((i++))
done
echo "[ERR] Timeout waiting for $url" >&2
return 1
}
log "Waiting for ES/Kibana/Master/Fluent Bit/Bind"
attempt=1; max=120
while (( attempt <= max )); do
if curl -fsS "http://localhost:9200/_cluster/health?wait_for_status=yellow&timeout=1s" >/dev/null 2>&1; then
break
fi
echo "[..] waiting ES ($attempt/$max)"
sleep 5
((attempt++))
done
if (( attempt > max )); then
echo "[ERR] ES not ready" >&2
exit 1
fi
log "Applying relaxed ES disk watermarks for debug"
curl -fsS -XPUT "http://localhost:9200/_cluster/settings" \
-H 'Content-Type: application/json' \
-d '{
"transient": {
"cluster.routing.allocation.disk.watermark.low": "99%",
"cluster.routing.allocation.disk.watermark.high": "99%",
"cluster.routing.allocation.disk.watermark.flood_stage": "99%"
}
}' >/dev/null || echo "[WARN] Failed to adjust ES watermarks"
log "Waiting for Kibana to be available (HTTP 200)"
kb_attempt=1; kb_max=180
while (( kb_attempt <= kb_max )); do
body=$(curl -sS "http://localhost:5601/api/status" 2>/dev/null || true)
code=$(curl -s -o /dev/null -w "%{http_code}" "http://localhost:5601/api/status" || echo 000)
if [[ "$code" == "200" ]] && echo "$body" | grep -q '"level":"available"'; then
log "Kibana available"
break
fi
echo "[..] waiting kibana 200 ($kb_attempt/$kb_max), last_code=$code"
sleep 5
((kb_attempt++))
done
if (( kb_attempt > kb_max )); then
echo "[ERR] Kibana did not reach HTTP 200" >&2
exit 1
fi
wait_http "http://localhost:32300/readyz" 120
wait_http "http://localhost:2020/api/v2/metrics" 120
wait_http "http://localhost:2021/api/v2/metrics" 120
BIND_ID="$(service_id bind)"
if [[ -n "$BIND_ID" ]]; then
docker exec "$BIND_ID" named-checkconf >/dev/null
else
echo "[WARN] bind container id not found" >&2
fi
log "All services are ready"