feat(platform): add quality notification plan

This commit is contained in:
lingniu
2026-07-04 13:31:36 +08:00
parent 817590c104
commit ff833bc065
9 changed files with 586 additions and 9 deletions

View File

@@ -76,6 +76,58 @@ func asClientError(err error) (clientError, bool) {
var canonicalVehicleProtocols = []string{"GB32960", "JT808", "YUTONG_MQTT"}
var qualityAlertRuleTemplates = []QualityAlertRule{
{
IssueType: "NO_SOURCE",
Title: "无数据来源",
Level: "P0",
Owner: "平台接入",
Trigger: "绑定车辆连续无 GB32960、JT808、MQTT 来源证据",
Notify: "立即通知接入运维30 分钟未恢复升级给业务责任人",
SLA: "30 分钟确认",
},
{
IssueType: "VIN_MISSING",
Title: "VIN 缺失",
Level: "P0",
Owner: "车辆档案",
Trigger: "来源有数据但无法归并到 VIN",
Notify: "通知档案维护人补齐车牌、手机号和 VIN 映射",
SLA: "2 小时修复",
},
{
IssueType: "LINK_GAP",
Title: "链路中断",
Level: "P1",
Owner: "链路运维",
Trigger: "车辆或平台来源上报间断、Redis 在线状态过期",
Notify: "通知平台转发和网关值班人,持续异常进入日报",
SLA: "1 小时恢复",
},
{
IssueType: "FIELD_MISSING",
Title: "字段缺失",
Level: "P1",
Owner: "协议解析",
Trigger: "核心字段缺失导致定位、里程或统计不可用",
Notify: "通知解析负责人核对字段映射和 RAW 样本",
SLA: "当日闭环",
},
}
var qualityNotificationPolicies = []QualityNotificationPolicy{
{Name: "P0 实时中断", Target: "接入运维 + 业务责任人", Channel: "站内告警 / 邮件 / 企业微信", Condition: "无来源、VIN 缺失、存储不可写"},
{Name: "P1 数据质量", Target: "协议解析 + 数据治理", Channel: "站内告警 / 每日汇总邮件", Condition: "字段缺失、链路间断、容量风险"},
{Name: "P2 趋势关注", Target: "数智中心", Channel: "周报 / 趋势看板", Condition: "单源车辆、档案缺项、来源覆盖下降"},
}
var qualityIssuePriorityWeight = map[string]int{
"NO_SOURCE": 10,
"VIN_MISSING": 9,
"LINK_GAP": 8,
"FIELD_MISSING": 7,
}
func completeProtocolStats(stats []ProtocolStat) []ProtocolStat {
byProtocol := make(map[string]ProtocolStat, len(stats)+len(canonicalVehicleProtocols))
for _, stat := range stats {
@@ -806,6 +858,46 @@ func (s *Service) QualityIssues(ctx context.Context, query url.Values) (Page[Qua
return s.store.QualityIssues(ctx, query)
}
func (s *Service) QualityNotificationPlan(ctx context.Context, query url.Values) (QualityNotificationPlan, error) {
summary, err := s.store.QualitySummary(ctx, query)
if err != nil {
return QualityNotificationPlan{}, err
}
issueQuery := cloneValues(query)
if issueQuery.Get("limit") == "" {
issueQuery.Set("limit", "20")
}
issues, err := s.store.QualityIssues(ctx, issueQuery)
if err != nil {
return QualityNotificationPlan{}, err
}
health, err := s.store.OpsHealth(ctx)
if err != nil {
return QualityNotificationPlan{}, err
}
health.Runtime = s.runtime
rules := qualityNotificationRules(summary, health)
activeRuleCount := 0
p0RuleCount := 0
for _, rule := range rules {
if rule.Count <= 0 {
continue
}
activeRuleCount++
if rule.Level == "P0" {
p0RuleCount++
}
}
return QualityNotificationPlan{
Summary: summary,
Rules: rules,
Policies: append([]QualityNotificationPolicy(nil), qualityNotificationPolicies...),
PriorityIssues: qualityPriorityIssues(issues.Items),
ActiveRuleCount: activeRuleCount,
P0RuleCount: p0RuleCount,
}, nil
}
func (s *Service) OpsHealth(ctx context.Context) (OpsHealth, error) {
health, err := s.store.OpsHealth(ctx)
if err != nil {
@@ -815,6 +907,219 @@ func (s *Service) OpsHealth(ctx context.Context) (OpsHealth, error) {
return health, nil
}
func qualityNotificationRules(summary QualitySummary, health OpsHealth) []QualityAlertRule {
rows := make([]QualityAlertRule, 0, len(qualityAlertRuleTemplates)+1)
for _, item := range qualityAlertRuleTemplates {
item.Count = qualityIssueCount(summary, item.IssueType)
rows = append(rows, item)
}
storageWritable := health.TDengineWritable && health.MySQLWritable
capacityCount := len(health.CapacityFindings)
capacityRule := QualityAlertRule{
IssueType: "CAPACITY_RISK",
Title: "容量与存储风险",
Level: "P2",
Owner: "基础设施",
Trigger: "Kafka Lag、连接容量、Redis 在线 Key 或存储读写异常",
Notify: "容量风险进入运维日报,超过阈值升级",
SLA: "当日评估",
Count: capacityCount,
}
if !storageWritable {
capacityRule.Level = "P0"
capacityRule.Notify = "立即通知基础设施值班人处理存储不可写"
capacityRule.SLA = "15 分钟恢复"
capacityRule.Count++
}
rows = append(rows, capacityRule)
return rows
}
func qualityIssueCount(summary QualitySummary, issueType string) int {
for _, item := range summary.IssueTypes {
if item.Name == issueType {
return item.Count
}
}
return 0
}
func qualityPriorityIssues(rows []QualityIssueRow) []QualityPriorityIssue {
out := make([]QualityPriorityIssue, 0, len(rows))
for _, row := range rows {
out = append(out, buildQualityPriorityIssue(row))
}
sort.Slice(out, func(i, j int) bool {
if out[i].Priority != out[j].Priority {
return out[i].Priority == "P0"
}
weightDelta := qualityIssuePriorityWeight[out[j].IssueType] - qualityIssuePriorityWeight[out[i].IssueType]
if weightDelta != 0 {
return weightDelta < 0
}
return out[i].LastSeen > out[j].LastSeen
})
if len(out) > 5 {
out = out[:5]
}
return out
}
func buildQualityPriorityIssue(row QualityIssueRow) QualityPriorityIssue {
actionLabel, actionDetail := qualityAction(row)
priority := "P1"
if row.Severity == "error" || row.IssueType == "NO_SOURCE" {
priority = "P0"
}
lookup := qualityIssueLookup(row)
dateFrom := evidenceDate(row.LastSeen)
dateTo := nextEvidenceDate(dateFrom)
realtimeHash := buildHash("realtime", lookup, row.Protocol, nil)
historyFilters := map[string]string{}
if dateFrom != "" {
historyFilters["dateFrom"] = dateFrom
}
if dateTo != "" {
historyFilters["dateTo"] = dateTo
}
historyHash := buildHash("history", lookup, row.Protocol, historyFilters)
rawFilters := cloneStringMap(historyFilters)
rawFilters["tab"] = "raw"
rawFilters["includeFields"] = "true"
rawHash := buildHash("history", lookup, row.Protocol, rawFilters)
vehicleHash := buildHash("detail", lookup, row.Protocol, nil)
item := QualityPriorityIssue{
QualityIssueRow: row,
Priority: priority,
ActionLabel: actionLabel,
ActionDetail: actionDetail,
SLA: qualityIssueSLA(row.IssueType),
VehicleLabel: qualityVehicleLabel(row),
RealtimeHash: realtimeHash,
HistoryHash: historyHash,
RawHash: rawHash,
VehicleHash: vehicleHash,
}
item.NotificationText = qualityNotificationText(item)
return item
}
func qualityAction(row QualityIssueRow) (string, string) {
switch {
case row.IssueType == "NO_SOURCE":
return "确认平台转发", "车辆已绑定但没有任何来源证据,先确认平台转发、端口和订阅。"
case row.IssueType == "VIN_MISSING" || strings.TrimSpace(row.VIN) == "":
return "维护身份绑定", "数据已有来源但无法归并到 VIN优先用车牌/手机号补齐绑定。"
case row.IssueType == "LINK_GAP":
return "排查来源链路", "来源存在上报间断,优先检查平台转发、网络和消费延迟。"
case row.IssueType == "FIELD_MISSING":
return "核对解析字段", "字段缺失会影响统计和展示,优先核对解析映射和原始 RAW。"
default:
return "查看车辆服务", "进入车辆服务详情,结合来源证据继续排查。"
}
}
func qualityIssueSLA(issueType string) string {
for _, item := range qualityAlertRuleTemplates {
if item.IssueType == issueType {
return item.SLA
}
}
return "当日闭环"
}
func qualityVehicleLabel(row QualityIssueRow) string {
identity := firstNonEmpty(strings.TrimSpace(row.VIN), strings.TrimSpace(row.Phone), strings.TrimSpace(row.SourceEndpoint), "-")
if plate := strings.TrimSpace(row.Plate); plate != "" {
return plate + " / " + identity
}
return identity
}
func qualityIssueLookup(row QualityIssueRow) string {
return firstNonEmpty(strings.TrimSpace(row.VIN), strings.TrimSpace(row.Plate), strings.TrimSpace(row.Phone), strings.TrimSpace(row.SourceEndpoint))
}
func evidenceDate(value string) string {
value = strings.TrimSpace(value)
if len(value) < len("2006-01-02") {
return ""
}
date := value[:len("2006-01-02")]
if _, err := time.Parse("2006-01-02", date); err != nil {
return ""
}
return date
}
func nextEvidenceDate(value string) string {
if value == "" {
return ""
}
date, err := time.Parse("2006-01-02", value)
if err != nil {
return ""
}
return date.AddDate(0, 0, 1).Format("2006-01-02")
}
func buildHash(page string, keyword string, protocol string, filters map[string]string) string {
values := url.Values{}
if keyword != "" {
values.Set("keyword", keyword)
}
if protocol != "" {
values.Set("protocol", protocol)
}
for key, value := range filters {
if strings.TrimSpace(value) != "" {
values.Set(key, value)
}
}
query := values.Encode()
if query == "" {
return "#/" + page
}
return "#/" + page + "?" + query
}
func qualityNotificationText(row QualityPriorityIssue) string {
return strings.Join([]string{
"【" + row.Priority + " 告警通知】" + qualityIssueTitle(row.IssueType),
"车辆:" + row.VehicleLabel,
"数据来源:" + row.Protocol,
"问题:" + qualityIssueTitle(row.IssueType),
"建议动作:" + row.ActionLabel,
"SLA" + row.SLA,
"最后时间:" + firstNonEmpty(row.LastSeen, "-"),
"详情:" + firstNonEmpty(row.Detail, "-"),
"实时定位:" + row.RealtimeHash,
"轨迹证据:" + row.HistoryHash,
"RAW证据" + row.RawHash,
"车辆服务:" + row.VehicleHash,
}, "\n")
}
func qualityIssueTitle(issueType string) string {
for _, item := range qualityAlertRuleTemplates {
if item.IssueType == issueType {
return item.Title
}
}
if issueType == "CAPACITY_RISK" {
return "容量与存储风险"
}
return issueType
}
func cloneStringMap(values map[string]string) map[string]string {
out := make(map[string]string, len(values))
for key, value := range values {
out[key] = value
}
return out
}
func (s *Service) resolveVehicleQuery(ctx context.Context, query url.Values) (url.Values, error) {
resolved := cloneValues(query)
vin := firstNonEmpty(resolved.Get("vin"), resolved.Get("keyword"))