Ensure Centralized Audit Log Collection Is Configured

中等級 1人工審核tdsql

說明

TDSQL can use managed Filebeat instances to collect logs from compute nodes, data nodes, and Agents and send them to Kafka or Elasticsearch. The destination may be DBbrain full-link audit analysis or an equivalent enterprise SIEM. Central collection must cover all production instances and all security-relevant log sources.

檢查方式

稽核方法: 人工審核

此項建議無法自動檢查。當您傳入 --include-manual 時，掃描結果會列出此項目並附帶這些說明。

Verify the Filebeat process and management-platform configuration on every production instance: ps aux | grep '[f]ilebeat' # Inspect the effective Filebeat inputs and outputs. grep -E 'type:|paths:|output\.kafka|output\.elasticsearch|hosts:|topic:' \ /path/to/filebeat.yml Confirm that the following required sources are enabled where present: 1. Compute-node interface logs. 2. Compute-node SQL logs. 3. Data-node DB audit logs. 4. Data-node slow query logs. 5. Data-node error logs. 6. Agent sys_log, back_log, and dump_log. 7.

修補方式

修補方法: 操作指南或基礎設施修補檔

In the TDSQL management platform, open the instance's Filebeat audit-log settings and enable the required compute-node, data-node, and Agent log inputs. Configure an approved Kafka or Elasticsearch destination and use consistent topics and partition rules for equivalent logs across instances. Where DBbrain high-level analysis is licensed, Kafka output may be connected to DBbrain for full-link analysis. DBbrain is optional;

屬性

作業系統: 不限
標籤: tdsqlloggingaudit
參考資料: CIS Tencent TDSQL 10.3.22 Benchmark v1.0.0 - 6.8 Ensure Centralized Audit Log Collection Is Configured