Merge branch 'master' of github.com:didi/nightingale

.github/ISSUE_TEMPLATE/bug_report.md

@@ -4,16 +4,19 @@ about: Report a bug encountered while operating Nightingale
 labels: kind/bug
 ---
 
-**What happened**:
+**问题现象**:
 
-**What you expected to happen**:
 
-**How to reproduce it (as minimally and precisely as possible)**:
+**复现方法**:
 
-**Anything else we need to know?**:
 
-**Environment**:
-- Nightingale Version (`./n9e-rdb -v`):
-- OS Version (`uname -a`):
-- Logs (stdout/error/warning/http.request/http.response):
-- Others:
+**环境信息**:
+
+- 夜莺服务端版本 (通过`./n9e-server -v`可得知版本):
+- 夜莺客户端版本 (通过`./n9e-agentd -v`可得知版本):
+- 操作系统版本 (通过`uname -a`可得知OS版本):
+
+**日志线索**:
+
+*日志分两部分，一个是logs目录下，重点查看ERROR日志；另一部分是stdout，如果是systemd托管的，可以通过 `journalctl -u <n9e-server|n9e-agentd> -f` 查看*
+

http/router.go

@@ -86,7 +86,6 @@ func configRoutes(r *gin.Engine) {
 		pages.DELETE("/user-group/:id", login(), userGroupDel)
 
 		pages.GET("/classpaths", login(), classpathListGets)
-		pages.GET("/classpaths/tree", login(), classpathListNodeGets)
 		pages.GET("/classpaths/tree-node/:id", login(), classpathListNodeGetsById)
 		pages.POST("/classpaths", login(), classpathAdd)
 		pages.PUT("/classpath/:id", login(), classpathPut)
@@ -159,6 +158,7 @@ func configRoutes(r *gin.Engine) {
 		pages.DELETE("/alert-events", login(), alertEventsDel)
 		pages.GET("/alert-event/:id", login(), alertEventGet)
 		pages.DELETE("/alert-event/:id", login(), alertEventDel)
+		pages.PUT("/alert-event/:id", login(), alertEventNotePut)
 
 		pages.GET("/history-alert-events", login(), historyAlertEventGets)
 		pages.GET("/history-alert-event/:id", login(), historyAlertEventGet)
@@ -205,7 +205,7 @@ func configRoutes(r *gin.Engine) {
 
 		v1.GET("/can-do-op-by-name", login(), canDoOpByName)
 		v1.GET("/can-do-op-by-token", login(), canDoOpByToken)
-
+		v1.GET("/get-user-by-name", login(), getUserByName)
 	}
 
 	push := r.Group("/v1/n9e/series").Use(gzip.Gzip(gzip.DefaultCompression))

http/router_alert_event.go

@@ -54,6 +54,20 @@ func alertEventGet(c *gin.Context) {
 	renderData(c, ae, nil)
 }
 
+type alertEventNoteForm struct {
+	EventNote string `json:"event_note"`
+}
+
+func alertEventNotePut(c *gin.Context) {
+	var f alertEventNoteForm
+	bind(c, &f)
+
+	me := loginUser(c).MustPerm("alert_event_modify")
+	ae := AlertEvent(urlParamInt64(c, "id"))
+
+	renderMessage(c, models.AlertEventUpdateEventNote(ae.Id, ae.HashId, f.EventNote, me.Id))
+}
+
 func alertEventDel(c *gin.Context) {
 	loginUser(c).MustPerm("alert_event_delete")
 	renderMessage(c, AlertEvent(urlParamInt64(c, "id")).Del())

http/router_auth.go

@@ -90,3 +90,8 @@ func canDoOpByToken(c *gin.Context) {
 	can, err := user.CanDo(queryStr(c, "op"))
 	renderData(c, can, err)
 }
+
+func getUserByName(c *gin.Context) {
+	user, err := models.UserGetByUsername(queryStr(c, "name"))
+	renderData(c, user, err)
+}

http/router_classpath.go

@@ -24,6 +24,7 @@ func classpathListGets(c *gin.Context) {
 	}, nil)
 }
 
+//此api暂时不对外开放
 func classpathListNodeGets(c *gin.Context) {
 	query := queryStr(c, "query", "")
 

judge/handler.go

@@ -527,6 +527,12 @@ func sendEventIfNeed(status []bool, event *models.AlertEvent, stra *models.Alert
 func SendEvent(event *models.AlertEvent) {
 	// update last event
 	LastEvents.Set(event)
+	ae, err := models.AlertEventGet("hash_id = ?", event.HashId)
+	if err == nil && ae != nil {
+		logger.Debugf("[event exists do not send again][type:%+v][event:%+v]", event.IsPromePull, event)
+		return
+	}
+
 	ok := EventQueue.PushFront(event)
 	if !ok {
 		logger.Errorf("push event:%v err", event)

models/alert_event.go

@@ -17,6 +17,9 @@ type AlertEvent struct {
 	RuleId             int64             `json:"rule_id"`
 	RuleName           string            `json:"rule_name"`
 	RuleNote           string            `json:"rule_note"`
+	ProcessorUid       int64             `json:"processor_uid"`
+	ProcessorObj       User              `json:"processor_user_obj" xorm:"-"`
+	EventNote          string            `json:"event_note"`
 	HashId             string            `json:"hash_id"`                 // 唯一标识
 	IsPromePull        int               `json:"is_prome_pull"`           // 代表是否是prometheus pull告警，为1时前端使用 ReadableExpression 拉取最近1小时数据
 	LastSend           bool              `json:"last_sent" xorm:"-"`      // true 代表上次发了，false代表还没发:给prometheus做for判断的
@@ -114,6 +117,14 @@ func (ae *AlertEvent) FillObjs() error {
 		ae.NotifyUserObjs = users
 	}
 
+	if ae.ProcessorUid != 0 {
+		processor, err := UserGetById(ae.ProcessorUid)
+		if err != nil {
+			return err
+		}
+		ae.ProcessorObj = *processor
+	}
+
 	return nil
 }
 
@@ -245,6 +256,7 @@ func AlertEventGets(stime, etime int64, query string, status, priority int, limi
 func AlertEventGet(where string, args ...interface{}) (*AlertEvent, error) {
 	var obj AlertEvent
 	has, err := DB.Where(where, args...).Get(&obj)
+
 	if err != nil {
 		logger.Errorf("mysql.error: query alert_event(%s)%+v fail: %s", where, args, err)
 		return nil, internalServerError
@@ -256,3 +268,24 @@ func AlertEventGet(where string, args ...interface{}) (*AlertEvent, error) {
 
 	return &obj, nil
 }
+
+func AlertEventUpdateEventNote(id int64, hashId string, note string, uid int64) error {
+	session := DB.NewSession()
+	defer session.Close()
+
+	if err := session.Begin(); err != nil {
+		return err
+	}
+
+	if _, err := session.Exec("UPDATE alert_event SET event_note = ?, processor_uid = ? WHERE id = ?", note, uid, id); err != nil {
+		logger.Errorf("mysql.error: update alert_event event_note fail: %v", err)
+		return err
+	}
+
+	if _, err := session.Exec("UPDATE history_alert_event SET event_note = ?, processor_uid = ? WHERE hash_id = ? ORDER BY id DESC LIMIT 1", note, uid, hashId); err != nil {
+		logger.Errorf("mysql.error: update history_alert_event event_note fail: %v", err)
+		return err
+	}
+
+	return session.Commit()
+}

models/collect_rule.go

@@ -28,7 +28,7 @@ type CollectRule struct {
 type PortConfig struct {
 	Port     int    `json:"port"`
 	Protocol string `json:"protocol"` // tcp or udp
-	Timeout  int    `json:"timeout "` // second
+	Timeout  int    `json:"timeout"`  // second
 }
 
 type ProcConfig struct {
@@ -41,7 +41,7 @@ type ScriptConfig struct {
 	Params  string            `json:"params"`
 	Stdin   string            `json:"stdin"`
 	Env     map[string]string `json:"env"`
-	Timeout int               `json:"timeout "` // second
+	Timeout int               `json:"timeout"` // second
 }
 
 type LogConfig struct {

models/history_alert_event.go

@@ -15,6 +15,9 @@ type HistoryAlertEvent struct {
 	RuleId             int64           `json:"rule_id"`
 	RuleName           string          `json:"rule_name"`
 	RuleNote           string          `json:"rule_note"`
+	ProcessorUid       int64           `json:"processor_uid"`
+	ProcessorObj       User            `json:"processor_user_objs" xorm:"-"`
+	EventNote          string          `json:"event_note"`
 	HashId             string          `json:"hash_id"`       // 唯一标识
 	IsPromePull        int             `json:"is_prome_pull"` // 代表是否是prometheus pull告警，为1时前端使用 ReadableExpression 拉取最近1小时数据
 	ResClasspaths      string          `json:"res_classpaths"`
@@ -79,6 +82,14 @@ func (hae *HistoryAlertEvent) FillObjs() error {
 		hae.NotifyUserObjs = users
 	}
 
+	if hae.ProcessorUid != 0 {
+		processor, err := UserGetById(hae.ProcessorUid)
+		if err != nil {
+			return err
+		}
+		hae.ProcessorObj = *processor
+	}
+
 	return nil
 }
 

sql/n9e.sql

@@ -99,6 +99,7 @@ insert into `role_operation`(role_name, operation) values('Standard', 'alert_rul
 insert into `role_operation`(role_name, operation) values('Standard', 'alert_rule_modify');
 insert into `role_operation`(role_name, operation) values('Standard', 'alert_rule_delete');
 insert into `role_operation`(role_name, operation) values('Standard', 'alert_event_delete');
+insert into `role_operation`(role_name, operation) values('Standard', 'alert_event_modify');
 insert into `role_operation`(role_name, operation) values('Standard', 'collect_rule_create');
 insert into `role_operation`(role_name, operation) values('Standard', 'collect_rule_modify');
 insert into `role_operation`(role_name, operation) values('Standard', 'collect_rule_delete');
@@ -295,6 +296,8 @@ CREATE TABLE `alert_event` (
     `rule_id` bigint unsigned not null,
     `rule_name` varchar(255) not null,
     `rule_note` varchar(512) not null default 'alert rule note',
+    `processor_uid` bigint NOT NULL default 0,
+    `event_note` varchar(512) not null default 'alert event note',
     `res_classpaths` varchar(1024) not null default '' comment 'belong classpaths',
     `priority` tinyint(1) not null,
     `status` tinyint(1) not null,
@@ -314,24 +317,26 @@ CREATE TABLE `alert_event` (
 ) ENGINE=InnoDB DEFAULT CHARSET=utf8;
 
 CREATE TABLE `history_alert_event` (
-  `id` bigint unsigned NOT NULL AUTO_INCREMENT,
-  `hash_id` varchar(255) NOT NULL COMMENT 'rule_id + point_pk',
-  `rule_id` bigint unsigned NOT NULL,
-  `rule_name` varchar(255) NOT NULL,
-  `rule_note` varchar(512) NOT NULL DEFAULT 'alert rule note',
-  `res_classpaths` varchar(1024) NOT NULL DEFAULT '' COMMENT 'belong classpaths',
-  `priority` tinyint(1) NOT NULL,
-  `status` tinyint(1) NOT NULL,
-  `is_prome_pull` tinyint(1) NOT NULL,
-  `is_recovery` tinyint(1) NOT NULL,
+  `id` bigint unsigned not null AUTO_INCREMENT,
+  `hash_id` varchar(255) not null COMMENT 'rule_id + point_pk',
+  `rule_id` bigint unsigned not null,
+  `rule_name` varchar(255) not null,
+  `rule_note` varchar(512) not null default 'alert rule note',
+  `processor_uid` bigint not null default 0,
+  `event_note` varchar(512) not null default 'alert event note',
+  `res_classpaths` varchar(1024) not null default '' COMMENT 'belong classpaths',
+  `priority` tinyint(1) not null,
+  `status` tinyint(1) not null,
+  `is_prome_pull` tinyint(1) not null,
+  `is_recovery` tinyint(1) not null,
   `history_points` text COMMENT 'metric, history points',
-  `trigger_time` bigint NOT NULL,
-  `notify_channels` varchar(255) NOT NULL DEFAULT '',
-  `notify_groups` varchar(255) NOT NULL DEFAULT '',
-  `notify_users` varchar(255) NOT NULL DEFAULT '',
-  `runbook_url` varchar(255) DEFAULT NULL,
-  `readable_expression` varchar(1024) NOT NULL COMMENT 'e.g. mem.bytes.used.percent(all,60s) > 0',
-  `tags` varchar(1024) NOT NULL DEFAULT 'merge data_tags rule_tags and res_tags',
+  `trigger_time` bigint not null,
+  `notify_channels` varchar(255) not null default '',
+  `notify_groups` varchar(255) not null default '',
+  `notify_users` varchar(255) not null default '',
+  `runbook_url` varchar(255) default NULL,
+  `readable_expression` varchar(1024) not null COMMENT 'e.g. mem.bytes.used.percent(all,60s) > 0',
+  `tags` varchar(1024) not null default 'merge data_tags rule_tags and res_tags',
   PRIMARY KEY (`id`),
   KEY `hash_id` (`hash_id`),
   KEY `rule_id` (`rule_id`),

timer/collect_rule.go

@@ -153,6 +153,11 @@ func changeCollectRule(rule *models.CollectRule) error {
 			return err
 		}
 
+		tags := strings.Fields(rule.AppendTags)
+		for i := 0; i < len(tags); i++ {
+			tags[i] = strings.Replace(tags[i], "=", ":", 1)
+		}
+
 		config := ScriptCollectFormat{
 			Instances: []struct {
 				MinCollectionInterval int               `json:"min_collection_interval,omitempty"`
@@ -162,6 +167,7 @@ func changeCollectRule(rule *models.CollectRule) error {
 				Env                   map[string]string `json:"env"`
 				Stdin                 string            `json:"stdin"`
 				Timeout               int               `json:"timeout"`
+				Tags                  []string          `json:"tags,omitempty"`
 			}{{
 				MinCollectionInterval: rule.Step,
 				FilePath:              conf.Path,
@@ -169,6 +175,7 @@ func changeCollectRule(rule *models.CollectRule) error {
 				Env:                   conf.Env,
 				Stdin:                 conf.Stdin,
 				Timeout:               conf.Timeout,
+				Tags:                  tags,
 			}},
 		}
 
@@ -257,6 +264,7 @@ type ScriptCollectFormat struct {
 		Env                   map[string]string `json:"env"`
 		Stdin                 string            `json:"stdin"`
 		Timeout               int               `json:"timeout"`
+		Tags                  []string          `json:"tags,omitempty"`
 	} `json:"instances"`
 }