feat(ops): 添加 count_tokens 错误过滤功能
功能特性: - 自动识别并标记 count_tokens 请求的错误 - 支持配置是否在统计中忽略 count_tokens 错误 - 错误数据完整保留,仅在统计时动态过滤 技术实现: - ops_error_logger.go: 自动标记 count_tokens 请求 - ops_repo.go: INSERT 语句添加 is_count_tokens 字段 - ops_repo_dashboard.go: buildErrorWhere 核心过滤函数 - ops_repo_preagg.go: 预聚合统计中添加过滤 - ops_repo_trends.go: 趋势统计查询添加过滤(2 处) - ops_settings_models.go: 添加 ignore_count_tokens_errors 配置 - ops_settings.go: 配置验证和默认值设置 - ops_port.go: 错误日志模型添加 IsCountTokens 字段 业务价值: - count_tokens 是探测性请求,其错误不影响真实业务 SLA - 用户可根据需求灵活控制是否计入统计 - 提升错误率、告警等运维指标的准确性 影响范围: - Dashboard 概览统计 - 错误趋势图表 - 告警规则评估 - 预聚合指标(hourly/daily) - 健康分数计算
This commit is contained in:
@@ -46,6 +46,7 @@ INSERT INTO ops_error_logs (
|
||||
severity,
|
||||
status_code,
|
||||
is_business_limited,
|
||||
is_count_tokens,
|
||||
error_message,
|
||||
error_body,
|
||||
error_source,
|
||||
@@ -64,7 +65,7 @@ INSERT INTO ops_error_logs (
|
||||
retry_count,
|
||||
created_at
|
||||
) VALUES (
|
||||
$1,$2,$3,$4,$5,$6,$7,$8,$9,$10,$11,$12,$13,$14,$15,$16,$17,$18,$19,$20,$21,$22,$23,$24,$25,$26,$27,$28,$29,$30,$31,$32,$33,$34
|
||||
$1,$2,$3,$4,$5,$6,$7,$8,$9,$10,$11,$12,$13,$14,$15,$16,$17,$18,$19,$20,$21,$22,$23,$24,$25,$26,$27,$28,$29,$30,$31,$32,$33,$34,$35
|
||||
) RETURNING id`
|
||||
|
||||
var id int64
|
||||
@@ -88,6 +89,7 @@ INSERT INTO ops_error_logs (
|
||||
opsNullString(input.Severity),
|
||||
opsNullInt(input.StatusCode),
|
||||
input.IsBusinessLimited,
|
||||
input.IsCountTokens,
|
||||
opsNullString(input.ErrorMessage),
|
||||
opsNullString(input.ErrorBody),
|
||||
opsNullString(input.ErrorSource),
|
||||
|
||||
@@ -964,8 +964,8 @@ func buildErrorWhere(filter *service.OpsDashboardFilter, start, end time.Time, s
|
||||
}
|
||||
|
||||
idx := startIndex
|
||||
clauses := make([]string, 0, 4)
|
||||
args = make([]any, 0, 4)
|
||||
clauses := make([]string, 0, 5)
|
||||
args = make([]any, 0, 5)
|
||||
|
||||
args = append(args, start)
|
||||
clauses = append(clauses, fmt.Sprintf("created_at >= $%d", idx))
|
||||
@@ -974,6 +974,8 @@ func buildErrorWhere(filter *service.OpsDashboardFilter, start, end time.Time, s
|
||||
clauses = append(clauses, fmt.Sprintf("created_at < $%d", idx))
|
||||
idx++
|
||||
|
||||
clauses = append(clauses, "is_count_tokens = FALSE")
|
||||
|
||||
if groupID != nil && *groupID > 0 {
|
||||
args = append(args, *groupID)
|
||||
clauses = append(clauses, fmt.Sprintf("group_id = $%d", idx))
|
||||
|
||||
@@ -78,7 +78,9 @@ error_base AS (
|
||||
status_code AS client_status_code,
|
||||
COALESCE(upstream_status_code, status_code, 0) AS effective_status_code
|
||||
FROM ops_error_logs
|
||||
-- Exclude count_tokens requests from error metrics as they are informational probes
|
||||
WHERE created_at >= $1 AND created_at < $2
|
||||
AND is_count_tokens = FALSE
|
||||
),
|
||||
error_agg AS (
|
||||
SELECT
|
||||
|
||||
@@ -170,6 +170,7 @@ error_totals AS (
|
||||
FROM ops_error_logs
|
||||
WHERE created_at >= $1 AND created_at < $2
|
||||
AND COALESCE(status_code, 0) >= 400
|
||||
AND is_count_tokens = FALSE -- 排除 count_tokens 请求的错误
|
||||
GROUP BY 1
|
||||
),
|
||||
combined AS (
|
||||
@@ -243,6 +244,7 @@ error_totals AS (
|
||||
AND platform = $3
|
||||
AND group_id IS NOT NULL
|
||||
AND COALESCE(status_code, 0) >= 400
|
||||
AND is_count_tokens = FALSE -- 排除 count_tokens 请求的错误
|
||||
GROUP BY 1
|
||||
),
|
||||
combined AS (
|
||||
|
||||
Reference in New Issue
Block a user