Overview
A block that only exists on the host is a block nobody can audit. Access rules, the smart firewall, the IDS, the WAF, threat intelligence, CAPTCHA and rate limits all emit the same structured event, so one query answers everything we blocked, and why.
Synapse emits; it does not store. Retention and search belong to your collector or SIEM.
When you see this
- The SOC asks why an address was blocked, and nobody can answer from the SIEM.
- Blocks happen on hosts that nobody logs into.
- You need to show which rule or feed drove enforcement.
Steps
- 01
Read one decision
layer is who enforced, action is what was done, and source is what decided. signals lists every family a fused rule used; source is only the first of them.
One event { "ts": "2026-09-08T11:04:22.481Z", "layer": "smart_firewall", "action": "drop", "source": "fingerprint_rule", "signals": ["fingerprint_rule", "threat_intel"], "rule_id": "sf-scanner-ja4t-01", "rule_msg": "known scanner TCP fingerprint", "ban_secs": 3600, "src_ip": "203.0.113.42", "dst_ip": "10.0.0.5", "dst_port": 443 } - 02
Answer why it was blocked
The last query is the useful daily one: a sudden shift in which source dominates is usually a feed change or a misfiring rule, not an attack.
From the event log # Everything dropped, newest first jq -c 'select(.action == "drop") | {ts, layer, source, src_ip, rule_msg}' \ /var/log/synapse/events.log | tail -20 # Which sources drive enforcement jq -r '.source // "unknown"' /var/log/synapse/events.log | sort | uniq -c | sort -rn - 03
Ship it as CEF
CEF carries three streams, and only block events are on by default. dvchost is on every record, so the SIEM can attribute each event to an agent.
/etc/synapse/config.yaml logging: cef: enabled: true file: "/var/log/synapse/cef.log" max_size: 104857600 keep: 10 streams: block_events: true ids_alerts: true access_logs: false device_host: "" - 04
Decide about access logs
Block events fire per decision; access logs fire per proxied request, which on a busy proxy is orders of magnitude more records.
DecisionIs your SIEM licensed by ingest volume?Yes Leave access_logs off. Block events and IDS alerts are the security picture.
No Turn it on deliberately; it is one record per request.
- 05
Or push over OTLP
OTLP over HTTP carries the event stream as log records and a set of metrics, so one collector gets both what was blocked and how much is being blocked. JSON Lines suits a shipper you already run: Fluent Bit, Vector or an OpenTelemetry Collector.
Common mistakes
- Attributing a block by source alone when the rule fused several signals.
- Turning on access_logs without checking how the SIEM is priced.
- Reading ban_secs as a countdown. It is the duration declared at install.
Escalate when
- A sudden shift in which source drives enforcement.
- Blocks on addresses in your own ranges.
Where this stops working
- Export is push or file, not a query API: Synapse keeps no searchable history.
- OTLP batches, so there is a short delay between a decision and its record.
- A shipper is your responsibility. The JSON Lines sink writes a file; something has to move it.