Merge pull request #1 from didi/master

update

.dockerignore

@@ -0,0 +1 @@
+web

Dockerfile

@@ -0,0 +1,28 @@
+FROM golang AS builder
+# RUN apk add --no-cache git gcc
+WORKDIR /app
+
+# comment this if using vendor
+# ENV GOPROXY=https://mod.gokit.info
+# COPY go.mod go.sum ./
+# RUN go mod download
+
+COPY . .
+ENV GOPROXY=https://mod.gokit.info
+RUN ./control build docker
+
+FROM buildpack-deps:buster-curl
+LABEL maintainer="llitfkitfk@gmail.com"
+
+WORKDIR /app
+
+COPY --from=builder /app/docker/scripts /app/scripts
+COPY --from=builder /app/etc /app/etc
+# Change default address (hard code) 
+RUN ./scripts/sed.sh
+
+COPY --from=builder /app/bin /usr/local/bin
+
+
+# ENTRYPOINT []
+# CMD []

README.md

@@ -15,7 +15,17 @@ Nightingale user manual: [https://n9e.didiyun.com/](https://n9e.didiyun.com/)
 mkdir -p $GOPATH/src/github.com/didi
 cd $GOPATH/src/github.com/didi
 git clone https://github.com/didi/nightingale.git
-cd nightingale && ./control build
+cd nightingale
+# export env[GOPROXY] if your network is not good
+# export GOPROXY=https://mirrors.aliyun.com/goproxy/
+./control build
+```
+
+## Quick Start (need install docker for [mac](https://docs.docker.com/docker-for-mac/install/)/[win](https://docs.docker.com/docker-for-windows/install/))
+
+```bash
+docker-compose up -d
+# open http://localhost in web browser
 ```
 
 ## Team

README_ZH.md

@@ -15,7 +15,10 @@ Nightingale是一套衍生自Open-Falcon的互联网监控解决方案，融入
 mkdir -p $GOPATH/src/github.com/didi
 cd $GOPATH/src/github.com/didi
 git clone https://github.com/didi/nightingale.git
-cd nightingale && ./control build
+cd nightingale
+# 如果网络环境不好可以尝试aliyun的mirror
+# export GOPROXY=https://mirrors.aliyun.com/goproxy/
+./control build
 ```
 
 ## 团队

control

@@ -139,8 +139,16 @@ build_one()
 	go build -o n9e-${mod} --tags "md5" src/modules/${mod}/${mod}.go
 }
 
+build_docker()
+{
+	mod=$1
+	go build -o bin/n9e-${mod} --tags "md5" src/modules/${mod}/${mod}.go
+}
+
 build()
 {
+  export GO111MODULE=on
+
 	mod=$1
 	if [ "x${mod}" = "x" ]; then
 		build_one monapi
@@ -152,6 +160,16 @@ build()
 		return
 	fi
 	
+	if [ "x${mod}" = "xdocker" ]; then
+		build_docker monapi
+		build_docker transfer
+		build_docker index
+		build_docker judge
+		build_docker collector
+		build_docker tsdb
+		return
+	fi
+
 	build_one $mod 
 }
 

docker-compose.yml

@@ -0,0 +1,88 @@
+version: "3"
+volumes: 
+  mysql-data:
+services:
+  nginx:
+    image: nginx:stable-alpine
+    ports: 
+      - 80:80
+    volumes: 
+      - ./docker/nginx/nginx.conf:/etc/nginx/nginx.conf
+      - ./docker/nginx/conf.d:/etc/nginx/conf.d
+      - ./pub:/home/n9e/pub
+  api:
+    build: .
+    image: api
+
+  monapi:
+    image: api
+    restart: always
+    command: n9e-monapi
+    ports: 
+      - 5800:5800
+      
+  transfer:
+    image: api
+    restart: always
+    command: n9e-transfer
+    ports: 
+      - 5810:5810
+      - 5811:5811
+
+  tsdb:
+    image: api
+    restart: always
+    command: n9e-tsdb
+    ports: 
+      - 5820:5820
+      - 5821:5821
+      
+  index:
+    image: api
+    restart: always
+    command: n9e-index
+    ports: 
+      - 5830:5830
+      - 5831:5831
+
+  judge:
+    image: api
+    restart: always
+    command: n9e-judge
+    ports: 
+      - 5840:5840
+      - 5841:5841
+
+  collector:
+    image: api
+    restart: always
+    command: n9e-collector
+    ports: 
+      - 2058:2058
+
+  # web:
+  #   build:
+  #     context: web
+  #   restart: always
+  #   command: npm run dev
+  #   ports: 
+  #     - 8010:8010
+
+  redis:
+    image: redis
+    restart: always
+    ports: 
+      - 6379:6379
+
+  mysql:
+    image: mysql:5.7
+    restart: always
+    environment:
+      - MYSQL_ROOT_PASSWORD=1234
+    ports: 
+      - 3306:3306
+    volumes:
+      - ./sql:/docker-entrypoint-initdb.d
+      - mysql-data:/var/lib/mysql
+  
+    

docker/nginx/nginx.conf

@@ -0,0 +1,89 @@
+user root;
+
+worker_processes auto;
+worker_cpu_affinity auto;
+
+error_log /var/log/nginx/error.log;
+pid /run/nginx.pid;
+
+include /usr/share/nginx/modules/*.conf;
+
+events {
+    use epoll;
+    worker_connections 204800;
+}
+
+http {
+    log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
+                      '$status $body_bytes_sent "$http_referer" '
+                      '"$http_user_agent" "$http_x_forwarded_for"';
+
+    access_log  /var/log/nginx/access.log  main;
+
+    sendfile            on;
+    tcp_nopush          on;
+    tcp_nodelay         on;
+    keepalive_timeout   65;
+    types_hash_max_size 2048;
+
+    include             /etc/nginx/mime.types;
+    default_type        application/octet-stream;
+
+    include /etc/nginx/conf.d/*.conf;
+
+
+    proxy_connect_timeout   500ms;
+    proxy_send_timeout      1000ms;
+    proxy_read_timeout      3000ms;
+    proxy_buffers           64 8k;
+    proxy_busy_buffers_size    128k;
+    proxy_temp_file_write_size 64k;
+    proxy_redirect off;
+    proxy_next_upstream error invalid_header timeout http_502 http_504;
+
+    proxy_set_header X-Real-IP $remote_addr;
+    proxy_set_header X-Real-Port $remote_port;
+    proxy_set_header Host $http_host;
+    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+
+    upstream n9e.monapi {
+        server monapi:5800;
+        keepalive 10;
+    }
+
+    upstream n9e.index {
+        server index:5830;
+        keepalive 10;
+    }
+
+    upstream n9e.transfer {
+        server transfer:5810;
+        keepalive 10;
+    }
+
+    server {
+        listen       80 default_server;
+        server_name  _;
+        root         /usr/share/nginx/html;
+
+        # Load configuration files for the default server block.
+        include /etc/nginx/conf.d/*.conf;
+
+        location / {
+            root /home/n9e/pub;
+        }
+
+        location /api/portal {
+            proxy_pass http://n9e.monapi;
+        }
+
+        location /api/index {
+            proxy_pass http://n9e.index;
+        }
+
+        location /api/transfer {
+            proxy_pass http://n9e.transfer;
+        }
+    }
+
+}

docker/scripts/sed.sh

@@ -0,0 +1,8 @@
+#!/bin/bash
+
+set -xe
+
+sed -i 's/127.0.0.1:6379/redis:6379/g' /app/etc/judge.yml
+sed -i 's/127.0.0.1:6379/redis:6379/g' /app/etc/monapi.yml
+sed -i 's/127.0.0.1:3306/mysql:3306/g' /app/etc/mysql.yml
+sed -i 's/127.0.0.1:5821/tsdb:5821/g' /app/etc/transfer.yml

etc/collector.yml

@@ -1,11 +1,10 @@
-# use shell if specify is blank
 logger:
   dir: logs/collector
   level: WARNING
   keepHours: 2
 identity:
   specify: ""
-  shell: /usr/sbin/ifconfig `/usr/sbin/route|grep '^default'|awk '{print $NF}'`|grep inet|awk '{print $2}'|head -n 1
+  shell: ifconfig `route|grep '^default'|awk '{print $NF}'`|grep inet|awk '{print $2}'|awk -F ':' '{print $NF}'|head -n 1
 sys:
   # timeout in ms
   # interval in second

etc/index.yml

@@ -4,4 +4,4 @@ logger:
   keepHours: 2
 identity:
   specify: ""
-  shell: /usr/sbin/ifconfig `/usr/sbin/route|grep '^default'|awk '{print $NF}'`|grep inet|awk '{print $2}'|head -n 1
+  shell: ifconfig `route|grep '^default'|awk '{print $NF}'`|grep inet|awk '{print $2}'|awk -F ':' '{print $NF}'|head -n 1

etc/judge.yml

@@ -6,6 +6,7 @@ query:
 redis:
   addrs: 
     - 127.0.0.1:6379
+  db: 0
   pass: ""
   # timeout:
   #   conn: 500
@@ -14,7 +15,7 @@ redis:
 
 identity:
   specify: ""
-  shell: /usr/sbin/ifconfig `/usr/sbin/route|grep '^default'|awk '{print $NF}'`|grep inet|awk '{print $2}'|head -n 1
+  shell: ifconfig `route|grep '^default'|awk '{print $NF}'`|grep inet|awk '{print $2}'|awk -F ':' '{print $NF}'|head -n 1
 
 logger:
   dir: logs/judge

etc/monapi.yml

@@ -14,11 +14,19 @@ ldap:
   host: "ldap.example.org"
   port: 389
   baseDn: "dc=example,dc=org"
+  # AD: manange@example.org
   bindUser: "cn=manager,dc=example,dc=org"
   bindPass: "*******"
   # openldap: (&(uid=%s))
   # AD: (&(sAMAccountName=%s))
   authFilter: "(&(uid=%s))"
+  attributes:
+    dispname: "cn"
+    email: "mail"
+    phone: "mobile"
+    im: ""
+  coverAttributes: false
+  autoRegist: false
   tls: false
   startTLS: false
 
@@ -44,6 +52,7 @@ link:
 # for alarm event and message queue
 redis:
   addr: "127.0.0.1:6379"
+  db: 0
   pass: ""
   # in ms
   # timeout:

etc/nginx.conf

@@ -1,4 +1,4 @@
-user nginx;
+user root;
 
 worker_processes auto;
 worker_cpu_affinity auto;

etc/port/20_2058

@@ -0,0 +1 @@
+n9e-collector

etc/port/20_22

@@ -1 +0,0 @@
-sshd

etc/port/20_5800

@@ -0,0 +1 @@
+n9e-monapi

etc/port/20_5810

@@ -0,0 +1 @@
+n9e-transfer

etc/port/20_5820

@@ -0,0 +1 @@
+n9e-tsdb

etc/port/20_5830

@@ -0,0 +1 @@
+n9e-index

etc/port/20_5840

@@ -0,0 +1 @@
+n9e-judge

pub/index-3eb7c9ca4d88e3ad5c58.js.map

@@ -1 +0,0 @@
-{"version":3,"file":"index-3eb7c9ca4d88e3ad5c58.js","sources":["webpack:///index-3eb7c9ca4d88e3ad5c58.js"],"mappings":"AAAA;;;;;;;AAmsYA","sourceRoot":""}

pub/index-5c8cbf958683e20086f5.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index-5c8cbf958683e20086f5.js","sources":["webpack:///index-5c8cbf958683e20086f5.js"],"mappings":"AAAA;;;;;;;AAmsYA","sourceRoot":""}

pub/index.html

@@ -1 +1 @@
-<!doctype html><html><head><meta charset="UTF-8"><title>Nightingale</title><link rel="shortcut icon" href="/favicon.ico"><link href="/index-3eb7c9ca4d88e3ad5c58.css" rel="stylesheet"></head><body><div id="react-content"></div><script src="/lib-033bee8514de110e36ef.dll.js"></script><script src="/index-3eb7c9ca4d88e3ad5c58.js"></script></body></html>
+<!doctype html><html><head><meta charset="UTF-8"><title>Nightingale</title><link rel="shortcut icon" href="/favicon.ico"><link href="/index-5c8cbf958683e20086f5.css" rel="stylesheet"></head><body><div id="react-content"></div><script src="/lib-033bee8514de110e36ef.dll.js"></script><script src="/index-5c8cbf958683e20086f5.js"></script></body></html>

sql/n9e_mon.sql

@@ -91,8 +91,8 @@ create table `tmp_chart` (
 create table `event_cur` (
   `id` bigint(20) unsigned not null AUTO_INCREMENT comment 'id',
   `sid` bigint(20) unsigned not null default 0 comment 'sid',
-  `sname` varchar(256) not null default '' comment 'name, 报警通知名称',
+  `sname` varchar(255) not null default '' comment 'name, 报警通知名称',
-  `node_path` varchar(256) not null default '' comment 'node path',
+  `node_path` varchar(255) not null default '' comment 'node path',
   `nid` int unsigned not null default '0' comment 'node id',
   `endpoint` varchar(255) not null default '' comment 'endpoint',
   `endpoint_alias` varchar(255) not null default '' comment 'endpoint alias',
@@ -103,7 +103,7 @@ create table `event_cur` (
   `detail` text comment 'counter points pred_points 详情',
   `hashid` varchar(128) not null default '' comment 'sid+counter hash',
   `etime` bigint(20) not null default 0 comment 'event ts',
-  `value` varchar(256) not null default '' comment '当前值',
+  `value` varchar(255) not null default '' comment '当前值',
   `users` varchar(512) not null default '[]' comment 'notify users',
   `groups` varchar(512) not null default '[]' comment 'notify groups',
   `info` varchar(512) not null default '' comment 'strategy info',
@@ -122,8 +122,8 @@ create table `event_cur` (
 create table `event` (
   `id` bigint(20) unsigned not null AUTO_INCREMENT comment 'id',
   `sid` bigint(20) unsigned not null default 0 comment 'sid',
-  `sname` varchar(256) not null default '' comment 'name, 报警通知名称',
+  `sname` varchar(255) not null default '' comment 'name, 报警通知名称',
-  `node_path` varchar(256) not null default '' comment 'node path',
+  `node_path` varchar(255) not null default '' comment 'node path',
   `nid` int unsigned not null default '0' comment 'node id',
   `endpoint` varchar(255) not null default '' comment 'endpoint',
   `endpoint_alias` varchar(255) not null default '' comment 'endpoint alias',
@@ -134,7 +134,7 @@ create table `event` (
   `detail` text comment 'counter points pred_points 详情',
   `hashid` varchar(128) not null default '' comment 'sid+counter hash',
   `etime` bigint(20) not null default 0 comment 'event ts',
-  `value` varchar(256) not null default '' comment '当前值',
+  `value` varchar(255) not null default '' comment '当前值',
   `users` varchar(512) not null default '[]' comment 'notify users',
   `groups` varchar(512) not null default '[]' comment 'notify groups',
   `info` varchar(512) not null default '' comment 'strategy info',
@@ -167,8 +167,8 @@ CREATE TABLE `stra` (
   `converge` varchar(45) NOT NULL DEFAULT '' COMMENT 'n秒最多报m次警',
   `recovery_notify` int(1) NOT NULL DEFAULT 1 COMMENT '1 发送恢复通知 0不发送恢复通知',
   `priority` int(1) NOT NULL DEFAULT 3 COMMENT '告警等级',
-  `notify_group` varchar(256) NOT NULL DEFAULT '' COMMENT '告警通知组',
+  `notify_group` varchar(255) NOT NULL DEFAULT '' COMMENT '告警通知组',
-  `notify_user` varchar(256) NOT NULL DEFAULT '' COMMENT '告警通知人',
+  `notify_user` varchar(255) NOT NULL DEFAULT '' COMMENT '告警通知人',
   `callback` varchar(1024) NOT NULL DEFAULT '' COMMENT 'callback url',
   `creator` varchar(64) NOT NULL COMMENT '创建者',
   `created` timestamp NOT NULL DEFAULT '1971-01-01 00:00:00' COMMENT 'created',

src/dataobj/metric.go

@@ -63,17 +63,14 @@ func (m *MetricValue) CheckValidity() (err error) {
 		return
 	}
 
-	//将保留字替换
+	//检测保留字
-	var illegal bool
+	if HasReservedWords(m.Metric) {
-	m.Metric, illegal = ReplaceReservedWords(m.Metric)
+		err = fmt.Errorf("metric:%s contains reserved words:[\\t] [\\r] [\\n] [,] [ ] [=]", m.Metric)
-	if illegal {
-		err = fmt.Errorf("Metric contains reserved word")
 		return
 	}
 
-	m.Endpoint, illegal = ReplaceReservedWords(m.Endpoint)
+	if HasReservedWords(m.Endpoint) {
-	if illegal {
+		err = fmt.Errorf("endpoint:%s contains reserved words:[\\t] [\\r] [\\n] [,] [ ] [=]", m.Endpoint)
-		err = fmt.Errorf("Endpoint contains reserved word:%s", m.Endpoint)
 		return
 	}
 
@@ -151,7 +148,7 @@ func (m *MetricValue) CheckValidity() (err error) {
 	return
 }
 
-func ReplaceReservedWords(str string) (string, bool) {
+func HasReservedWords(str string) bool {
 	if -1 == strings.IndexFunc(str,
 		func(r rune) bool {
 			return r == '\t' ||
@@ -159,27 +156,13 @@ func ReplaceReservedWords(str string) (string, bool) {
 				r == '\n' ||
 				r == ',' ||
 				r == ' ' ||
-				r == ':' ||
 				r == '='
 		}) {
 
-		return str, false
+		return false
 	}
 
-	return strings.Map(func(r rune) rune {
+	return true
-		if r == '\t' ||
-			r == '\r' ||
-			r == '\n' ||
-			r == ',' ||
-			r == ' ' ||
-			r == ':' ||
-			r == '=' {
-			return '_'
-		}
-		return r
-	}, str), true
-
-	return str, false
 }
 
 func SortedTags(tags map[string]string) string {

src/dataobj/query_item.go

@@ -24,6 +24,17 @@ type QueryDataForUI struct {
 	Comparisons []int64  `json:"comparisons"` //环比多少时间
 }
 
+type QueryDataForUIResp struct {
+	Start      int64      `json:"start"`
+	End        int64      `json:"end"`
+	Endpoint   string     `json:"endpoint"`
+	Counter    string     `json:"counter"`
+	DsType     string     `json:"dstype"`
+	Step       int        `json:"step"`
+	Values     []*RRDData `json:"values"`
+	Comparison int64      `json:"comparison"`
+}
+
 type QueryDataResp struct {
 	Data []*TsdbQueryResponse
 	Msg  string

src/model/ldap.go

@@ -0,0 +1,85 @@
+package model
+
+import (
+	"crypto/tls"
+	"fmt"
+
+	"github.com/didi/nightingale/src/modules/monapi/config"
+	"gopkg.in/ldap.v3"
+)
+
+func genLdapAttributeSearchList() []string {
+	ldapAttributes := []string{}
+	attrs := config.Get().LDAP.Attributes
+	if attrs.Dispname != "" {
+		ldapAttributes = append(ldapAttributes, attrs.Dispname)
+	}
+	if attrs.Email != "" {
+		ldapAttributes = append(ldapAttributes, attrs.Email)
+	}
+	if attrs.Phone != "" {
+		ldapAttributes = append(ldapAttributes, attrs.Phone)
+	}
+	if attrs.Im != "" {
+		ldapAttributes = append(ldapAttributes, attrs.Im)
+	}
+	return ldapAttributes
+}
+
+func ldapReq(user, pass string) (*ldap.SearchResult, error) {
+	var conn *ldap.Conn
+	var err error
+	lc := config.Get().LDAP
+	addr := fmt.Sprintf("%s:%d", lc.Host, lc.Port)
+
+	if lc.TLS {
+		conn, err = ldap.DialTLS("tcp", addr, &tls.Config{InsecureSkipVerify: true})
+	} else {
+		conn, err = ldap.Dial("tcp", addr)
+	}
+
+	if err != nil {
+		return nil, fmt.Errorf("cannot dial ldap: %v", err)
+	}
+
+	defer conn.Close()
+
+	if !lc.TLS && lc.StartTLS {
+		if err := conn.StartTLS(&tls.Config{InsecureSkipVerify: true}); err != nil {
+			return nil, fmt.Errorf("ldap.conn startTLS fail: %v", err)
+		}
+	}
+	//if bindUser is empty, anonymousSearch mode
+	if lc.BindUser != "" {
+		//BindSearch mode
+		if err := conn.Bind(lc.BindUser, lc.BindPass); err != nil {
+			return nil, fmt.Errorf("bind ldap fail: %v, use %s", err, lc.BindUser)
+		}
+	}
+
+	searchRequest := ldap.NewSearchRequest(
+		lc.BaseDn, // The base dn to search
+		ldap.ScopeWholeSubtree, ldap.NeverDerefAliases, 0, 0, false,
+		fmt.Sprintf(lc.AuthFilter, user), // The filter to apply
+		genLdapAttributeSearchList(),     // A list attributes to retrieve
+		nil,
+	)
+
+	sr, err := conn.Search(searchRequest)
+	if err != nil {
+		return nil, fmt.Errorf("ldap search fail: %v", err)
+	}
+
+	if len(sr.Entries) == 0 {
+		return nil, fmt.Errorf("cannot find such user: %v", user)
+	}
+
+	if len(sr.Entries) > 1 {
+		return nil, fmt.Errorf("multi users is search, query user: %v", user)
+	}
+
+	if err := conn.Bind(sr.Entries[0].DN, pass); err != nil {
+		return nil, fmt.Errorf("password error")
+	}
+	return sr, nil
+}

src/model/node.go

@@ -79,7 +79,7 @@ func NodeQueryPath(query string, limit int) (nodes []Node, err error) {
 
 func TreeSearchByPath(query string) (nodes []Node, err error) {
 	session := DB["mon"].NewSession()
-	defer session.Clone()
+	defer session.Close()
 
 	if strings.Contains(query, " ") {
 		arr := strings.Fields(query)

src/model/team.go

@@ -20,7 +20,7 @@ type Team struct {
 
 func (t *Team) Del() error {
 	session := DB["uic"].NewSession()
-	defer session.Clone()
+	defer session.Close()
 
 	if err := session.Begin(); err != nil {
 		return err

src/model/user.go

@@ -1,7 +1,6 @@
 package model
 
 import (
-	"crypto/tls"
 	"fmt"
 	"log"
 	"strings"
@@ -118,6 +117,23 @@ func (u *User) CanModifyTeam(t *Team) (bool, error) {
 	return cnt > 0, err
 }
 
+func (u *User) CopyLdapAttr(sr *ldap.SearchResult) {
+	attrs := config.Get().LDAP.Attributes
+	if attrs.Dispname != "" {
+		u.Dispname = sr.Entries[0].GetAttributeValue(attrs.Dispname)
+	}
+	if attrs.Email != "" {
+		u.Email = sr.Entries[0].GetAttributeValue(attrs.Email)
+	}
+	if attrs.Phone != "" {
+		u.Phone = sr.Entries[0].GetAttributeValue(attrs.Phone)
+	}
+	if attrs.Im != "" {
+		u.Im = sr.Entries[0].GetAttributeValue(attrs.Im)
+	}
+	return
+}
+
 func InitRoot() {
 	var u User
 	has, err := DB["uic"].Where("username=?", "root").Get(&u)
@@ -147,78 +163,31 @@ func InitRoot() {
 }
 
 func LdapLogin(user, pass string) error {
-	var conn *ldap.Conn
+	sr, err := ldapReq(user, pass)
-	var err error
-
-	lc := config.Get().LDAP
-	addr := fmt.Sprintf("%s:%d", lc.Host, lc.Port)
-
-	if lc.TLS {
-		conn, err = ldap.DialTLS("tcp", addr, &tls.Config{InsecureSkipVerify: true})
-	} else {
-		conn, err = ldap.Dial("tcp", addr)
-	}
-
-	if err != nil {
-		return fmt.Errorf("cannot dial ldap: %v", err)
-	}
-
-	defer conn.Close()
-
-	if !lc.TLS && lc.StartTLS {
-		err = conn.StartTLS(&tls.Config{InsecureSkipVerify: true})
-		if err != nil {
-			return fmt.Errorf("ldap.conn startTLS fail: %v", err)
-		}
-	}
-
-	err = conn.Bind(lc.BindUser, lc.BindPass)
-	if err != nil {
-		return fmt.Errorf("bind ldap fail: %v, use %s", err, lc.BindUser)
-	}
-
-	searchRequest := ldap.NewSearchRequest(
-		lc.BaseDn, // The base dn to search
-		ldap.ScopeWholeSubtree, ldap.NeverDerefAliases, 0, 0, false,
-		fmt.Sprintf(lc.AuthFilter, user), // The filter to apply
-		[]string{},                       // A list attributes to retrieve
-		nil,
-	)
-
-	sr, err := conn.Search(searchRequest)
-	if err != nil {
-		return fmt.Errorf("ldap search fail: %v", err)
-	}
-
-	if len(sr.Entries) == 0 {
-		return fmt.Errorf("cannot find such user: %v", user)
-	}
-
-	if len(sr.Entries) > 1 {
-		return fmt.Errorf("multi users is search, query user: %v", user)
-	}
-
-	err = conn.Bind(sr.Entries[0].DN, pass)
-	if err != nil {
-		return fmt.Errorf("password error")
-	}
-
-	cnt, err := DB["uic"].Where("username=?", user).Count(new(User))
 	if err != nil {
 		return err
 	}
 
-	if cnt > 0 {
+	var u User
+	has, err := DB["uic"].Where("username=?", user).Get(&u)
+	if err != nil {
+		return err
+	}
+	u.CopyLdapAttr(sr)
+	if has {
+		if config.Get().LDAP.CoverAttributes {
+			_, err := DB["uic"].Where("id=?", u.Id).Update(u)
+			return err
+		} else {
 			return nil
 		}
-
+	}
-	u := &User{
+	if !config.Get().LDAP.AutoRegist {
-		Username: user,
+		return fmt.Errorf("user has not be created, may be you should enable auto regist: %v", user)
-		Password: "******",
-		Dispname: "",
-		Email:    "",
 	}
 
+	u.Username = user
+	u.Password = "******"
 	_, err = DB["uic"].Insert(u)
 	return err
 }

src/modules/collector/collector.go

@@ -63,6 +63,8 @@ func main() {
 	identity.Init(cfg.Identity)
 	if identity.Identity == "127.0.0.1" {
 		log.Fatalln("endpoint: 127.0.0.1, cannot work")
+	} else {
+		log.Println("endpoint:", identity.Identity)
 	}
 
 	sys.Init(cfg.Sys)
@@ -70,7 +72,6 @@ func main() {
 
 	funcs.BuildMappers()
 	funcs.Collect()
-	stra.GetCollects()
 
 	//插件采集
 	plugins.Detect()
@@ -137,6 +138,5 @@ func ending() {
 func start() {
 	runner.Init()
 	fmt.Println("collector start, use configuration file:", *conf)
-	fmt.Println("runner.Cwd:", runner.Cwd)
+	fmt.Println("runner.cwd:", runner.Cwd)
-	fmt.Println("runner.Endpoint:", runner.Hostname)
 }

src/modules/collector/config/config.go

@@ -62,16 +62,16 @@ func Parse(conf string) error {
 		"enable":   true,
 		"timeout":  1000,
 		"interval": 10, //采集策略更新时间
-		"portPath": "/home/n9e/etc/port",
+		"portPath": "./etc/port",
-		"procPath": "/home/n9e/etc/proc",
+		"procPath": "./etc/proc",
-		"logPath":  "/home/n9e/etc/log",
+		"logPath":  "./etc/log",
 		"api":      "/api/portal/collects/",
 	})
 
 	viper.SetDefault("sys", map[string]interface{}{
 		"timeout":  1000, //请求超时时间
 		"interval": 10,   //基础指标上报周期
-		"plugin":   "/home/n9e/plugin",
+		"plugin":   "./plugin",
 	})
 
 	err = viper.Unmarshal(&Config)

src/modules/collector/log/strategy/strategy.go

@@ -18,9 +18,9 @@ func init() {
 }
 
 func Update() error {
-	strategys := stra.GetLogCollects()
+	strategies := stra.GetLogCollects()
 
-	err := UpdateGlobalStrategy(strategys)
+	err := UpdateGlobalStrategy(strategies)
 	if err != nil {
 		logger.Errorf("Update Strategy cache error ! [msg:%v]", err)
 		return err

src/modules/collector/stra/cron.go

@@ -74,7 +74,7 @@ func getCollects() (CollectResp, error) {
 	url := fmt.Sprintf("http://%s%s%s", addr, StraConfig.Api, identity.Identity)
 	err = httplib.Get(url).SetTimeout(time.Duration(StraConfig.Timeout) * time.Millisecond).ToJSON(&res)
 	if err != nil {
-		err = fmt.Errorf("get collects from remote failed, error:%v", err)
+		err = fmt.Errorf("get collects from remote:%s failed, error:%v", url, err)
 	}
 
 	return res, err

src/modules/collector/stra/log.go

@@ -146,8 +146,8 @@ func DeepCopyStringMap(p map[string]string) map[string]string {
 
 const PATTERN_EXCLUDE_PARTITION = "```EXCLUDE```"
 
-func parsePattern(strategys []*Strategy) {
+func parsePattern(strategies []*Strategy) {
-	for _, st := range strategys {
+	for _, st := range strategies {
 		patList := strings.Split(st.Pattern, PATTERN_EXCLUDE_PARTITION)
 
 		if len(patList) == 1 {
@@ -163,8 +163,8 @@ func parsePattern(strategys []*Strategy) {
 	}
 }
 
-func updateRegs(strategys []*Strategy) {
+func updateRegs(strategies []*Strategy) {
-	for _, st := range strategys {
+	for _, st := range strategies {
 		st.TagRegs = make(map[string]*regexp.Regexp, 0)
 		st.ParseSucc = false
 

src/modules/collector/sys/funcs/push.go

@@ -48,7 +48,9 @@ func Push(items []*dataobj.MetricValue) {
 				logger.Error(err)
 				continue
 			} else {
-				logger.Info("push succ, reply: ", reply)
+				if reply.Msg != "ok" {
+					logger.Error("some item push err", reply)
+				}
 				return
 			}
 		}

src/modules/collector/sys/plugins/reader.go

@@ -32,6 +32,7 @@ func ListPlugins(dir string) map[string]*Plugin {
 		filename := f.Name()
 		arr := strings.Split(filename, "_")
 		if len(arr) < 2 {
+			logger.Warningf("plugin:%s name illegal, should be: $cycle_$xx", filename)
 			continue
 		}
 
@@ -39,10 +40,16 @@ func ListPlugins(dir string) map[string]*Plugin {
 		var cycle int
 		cycle, err = strconv.Atoi(arr[0])
 		if err != nil {
+			logger.Warningf("plugin:%s name illegal, should be: $cycle_$xx %v", filename, err)
+			continue
+		}
+
+		fpath, err := filepath.Abs(filepath.Join(dir, filename))
+		if err != nil {
+			logger.Warningf("plugin:%s absolute path get err:%v", filename, err)
 			continue
 		}
 
-		fpath := filepath.Join(dir, filename)
 		plugin := &Plugin{FilePath: fpath, MTime: f.ModTime().Unix(), Cycle: cycle}
 		ret[fpath] = plugin
 	}

src/modules/collector/sys/plugins/scheduler.go

@@ -50,8 +50,7 @@ func PluginRun(plugin *Plugin) {
 
 	timeout := plugin.Cycle*1000 - 500 //比运行周期少500毫秒
 
-	fpath := filepath.Join(plugin.FilePath)
+	fpath := plugin.FilePath
-
 	if !file.IsExist(fpath) {
 		logger.Error("no such plugin:", fpath)
 		return

src/modules/collector/sys/procs/cron.go

@@ -20,6 +20,6 @@ func loopDetect() {
 
 func detect() {
 	ps := stra.GetProcCollects()
-	DelNoPorcCollect(ps)
+	DelNoProcCollect(ps)
-	AddNewPorcCollect(ps)
+	AddNewProcCollect(ps)
 }

src/modules/collector/sys/procs/proc.go

@@ -9,7 +9,7 @@ var (
 	ProcsWithScheduler = make(map[string]*ProcScheduler)
 )
 
-func DelNoPorcCollect(newCollect map[string]*model.ProcCollect) {
+func DelNoProcCollect(newCollect map[string]*model.ProcCollect) {
 	for currKey, currProc := range Procs {
 		newProc, ok := newCollect[currKey]
 		if !ok || currProc.LastUpdated != newProc.LastUpdated {
@@ -18,7 +18,7 @@ func DelNoPorcCollect(newCollect map[string]*model.ProcCollect) {
 	}
 }
 
-func AddNewPorcCollect(newCollect map[string]*model.ProcCollect) {
+func AddNewProcCollect(newCollect map[string]*model.ProcCollect) {
 	for target, newProc := range newCollect {
 		if _, ok := Procs[target]; ok && newProc.LastUpdated == Procs[target].LastUpdated {
 			continue

src/modules/index/cache/endpoints.go

@@ -6,6 +6,7 @@ import (
 	"time"
 
 	"github.com/didi/nightingale/src/toolkits/address"
+	"github.com/didi/nightingale/src/toolkits/stats"
 
 	"github.com/toolkits/pkg/concurrent/semaphore"
 	"github.com/toolkits/pkg/logger"
@@ -49,11 +50,12 @@ func reportEndpoint(endpoints []interface{}) {
 			err := httplib.Post(url).JSONBodyQuiet(m).SetTimeout(3*time.Second).Header("x-srv-token", "monapi-builtin-token").ToJSON(&body)
 			if err != nil {
 				logger.Warningf("curl %s fail: %v. retry", url, err)
+				stats.Counter.Set("report.endpoint.err", 1)
 				continue
 			}
-
+			if body.Err != "" { //数据库连接出错会出现此情况
-			if body.Err != "" {
 				logger.Warningf("curl %s fail: %s. retry", url, body.Err)
+				stats.Counter.Set("report.endpoint.err", 1)
 				continue
 			}
 

src/modules/index/cache/indexdb.go

@@ -19,6 +19,7 @@ import (
 	"github.com/didi/nightingale/src/toolkits/compress"
 	"github.com/didi/nightingale/src/toolkits/identity"
 	"github.com/didi/nightingale/src/toolkits/report"
+	"github.com/didi/nightingale/src/toolkits/stats"
 )
 
 type CacheSection struct {
@@ -72,8 +73,8 @@ func StartPersist(interval int) {
 		err := Persist("normal")
 		if err != nil {
 			logger.Error("Persist err:", err)
+			stats.Counter.Set("persist.err", 1)
 		}
-		//logger.Infof("clean %+v, took %.2f ms\n", cleanRet, float64(time.Since(start).Nanoseconds())*1e-6)
 	}
 }
 

src/modules/index/cache/tag_map.go

@@ -30,8 +30,9 @@ func (t *TagkvIndex) GetTagkv() []*TagPair {
 	t.RLock()
 	defer t.RUnlock()
 	tagkvs := []*TagPair{}
-	var vs []string
+
 	for k, vm := range t.Tagkv {
+		var vs []string
 		for v, _ := range vm {
 			vs = append(vs, v)
 		}
@@ -50,8 +51,8 @@ func (t *TagkvIndex) GetTagkvMap() map[string][]string {
 	defer t.RUnlock()
 	tagkvs := make(map[string][]string)
 
-	var vs []string
 	for k, vm := range t.Tagkv {
+		var vs []string
 		for v, _ := range vm {
 			vs = append(vs, v)
 		}

src/modules/index/http/routes/index_router.go

@@ -93,9 +93,8 @@ func GetTagPairs(c *gin.Context) {
 
 	resp := []*IndexTagkvResp{}
 
-	tagkvFilter := make(map[string]map[string]struct{})
-
 	for _, metric := range recv.Metrics {
+		tagkvFilter := make(map[string]map[string]struct{})
 		tagkvs := []*cache.TagPair{}
 
 		for _, endpoint := range recv.Endpoints {
@@ -107,6 +106,7 @@ func GetTagPairs(c *gin.Context) {
 			}
 
 			tagkvMap := metricIndex.TagkvMap.GetTagkvMap()
+
 			for tagk, tagvs := range tagkvMap {
 				tagvFilter, exists := tagkvFilter[tagk]
 				if !exists {

src/modules/judge/backend/redi/funcs.go

@@ -42,6 +42,5 @@ func Push(event *dataobj.Event) error {
 		return nil
 	}
 
-	stats.Counter.Set("redis.failed", 1)
 	return fmt.Errorf("redis publish failed finally:%v", err)
 }

src/modules/judge/backend/redi/redis.go

@@ -4,6 +4,7 @@ import (
 	"log"
 	"time"
 
+	"github.com/didi/nightingale/src/toolkits/stats"
 	"github.com/garyburd/redigo/redis"
 	"github.com/toolkits/pkg/logger"
 )
@@ -14,6 +15,7 @@ var Config RedisSection
 type RedisSection struct {
 	Addrs   []string       `yaml:"addrs"`
 	Pass    string         `yaml:"pass"`
+	DB      int            `yaml:"db"`
 	Idle    int            `yaml:"idle"`
 	Timeout TimeoutSection `yaml:"timeout"`
 	Prefix  string         `yaml:"prefix"`
@@ -30,6 +32,7 @@ func Init(cfg RedisSection) {
 
 	addrs := cfg.Addrs
 	pass := cfg.Pass
+	db := cfg.DB
 	maxIdle := cfg.Idle
 	idleTimeout := 240 * time.Second
 
@@ -44,6 +47,7 @@ func Init(cfg RedisSection) {
 				c, err := redis.Dial("tcp", addr, redis.DialConnectTimeout(connTimeout), redis.DialReadTimeout(readTimeout), redis.DialWriteTimeout(writeTimeout))
 				if err != nil {
 					logger.Errorf("conn redis err:%v", err)
+					stats.Counter.Set("redis.conn.failed", 1)
 					return nil, err
 				}
 
@@ -51,6 +55,17 @@ func Init(cfg RedisSection) {
 					if _, err := c.Do("AUTH", pass); err != nil {
 						c.Close()
 						logger.Errorf("ERR: redis auth fail:%v", err)
+						stats.Counter.Set("redis.conn.failed", 1)
+
+						return nil, err
+					}
+				}
+
+				if db != 0 {
+					if _, err := c.Do("SELECT", db); err != nil {
+						c.Close()
+						logger.Error("redis select db fail, db: ", db)
+						stats.Counter.Set("redis.conn.failed", 1)
 						return nil, err
 					}
 				}

src/modules/judge/judge/judge.go

@@ -65,7 +65,7 @@ func Judge(stra *model.Stra, exps []model.Exp, historyData []*dataobj.RRDData, f
 	stats.Counter.Set("running", 1)
 
 	if len(exps) < 1 {
-		stats.Counter.Set("stra.err", 1)
+		stats.Counter.Set("stra.illegal", 1)
 		logger.Warningf("stra:%v exp is null", stra)
 		return
 	}
@@ -105,21 +105,21 @@ func Judge(stra *model.Stra, exps []model.Exp, historyData []*dataobj.RRDData, f
 				Hashid:    getHashId(stra.Id, firstItem),
 			}
 
-			sendEventIfNeed(historyData, isTriggered, now, event)
+			sendEventIfNeed(historyData, isTriggered, event)
 		}
 	}()
 
 	leftValue, isTriggered = judgeItemWithStrategy(stra, historyData, exps[0], firstItem, now)
-	if !isTriggered {
-		return
-	}
-
 	if value == "" {
 		value = fmt.Sprintf("%s: %v", exp.Metric, leftValue)
 	} else {
 		value += fmt.Sprintf("; %s: %v", exp.Metric, leftValue)
 	}
 
+	if !isTriggered {
+		return
+	}
+
 	//与条件情况下执行
 	if len(exps) > 1 {
 		if exps[1].Func == "nodata" { //nodata重新查询索引来进行告警判断
@@ -387,7 +387,7 @@ func GetReqs(stra *model.Stra, metric string, endpoints []string, now int64) ([]
 	return reqs, nil
 }
 
-func sendEventIfNeed(historyData []*dataobj.RRDData, isTriggered bool, now int64, event *dataobj.Event) {
+func sendEventIfNeed(historyData []*dataobj.RRDData, isTriggered bool, event *dataobj.Event) {
 	lastEvent, exists := cache.LastEvents.Get(event.ID)
 	if isTriggered {
 		event.EventType = EVENT_ALERT
@@ -421,6 +421,7 @@ func sendEvent(event *dataobj.Event) {
 
 	err := redi.Push(event)
 	if err != nil {
+		stats.Counter.Set("redis.push.failed", 1)
 		logger.Errorf("push event:%v err:%v", event, err)
 	}
 }

src/modules/judge/stra/stra.go

@@ -56,11 +56,13 @@ func getStrategy(opts StrategySection) {
 
 		if err != nil {
 			logger.Warningf("get strategy from remote failed, error:%v", err)
+			stats.Counter.Set("stra.get.err", 1)
 			continue
 		}
 
 		if resp.Err != "" {
 			logger.Warningf("get strategy from remote failed, error:%v", resp.Err)
+			stats.Counter.Set("stra.get.err", 1)
 			continue
 		}
 

src/modules/monapi/config/yaml.go

@@ -43,6 +43,7 @@ type cleanerSection struct {
 type redisSection struct {
 	Addr    string         `yaml:"addr"`
 	Pass    string         `yaml:"pass"`
+	DB      int            `yaml:"db"`
 	Idle    int            `yaml:"idle"`
 	Timeout timeoutSection `yaml:"timeout"`
 }
@@ -70,10 +71,20 @@ type ldapSection struct {
 	BindUser        string         `yaml:"bindUser"`
 	BindPass        string         `yaml:"bindPass"`
 	AuthFilter      string         `yaml:"authFilter"`
+	Attributes      ldapAttributes `yaml:"attributes"`
+	CoverAttributes bool           `yaml:"coverAttributes"`
+	AutoRegist      bool           `yaml:"autoRegist"`
 	TLS             bool           `yaml:"tls"`
 	StartTLS        bool           `yaml:"startTLS"`
 }
 
+type ldapAttributes struct {
+	Dispname string `yaml:"dispname"`
+	Phone    string `yaml:"phone"`
+	Email    string `yaml:"email"`
+	Im       string `yaml:"im"`
+}
+
 var (
 	yaml *Config
 	lock = new(sync.RWMutex)

src/modules/monapi/cron/checker_judge.go

@@ -11,6 +11,7 @@ import (
 	"github.com/didi/nightingale/src/model"
 	"github.com/didi/nightingale/src/modules/monapi/config"
 	"github.com/didi/nightingale/src/modules/monapi/scache"
+	"github.com/didi/nightingale/src/toolkits/stats"
 )
 
 func CheckJudgeLoop() {
@@ -19,6 +20,7 @@ func CheckJudgeLoop() {
 		time.Sleep(duration)
 		err := CheckJudge()
 		if err != nil {
+			stats.Counter.Set("get.judge.err", 1)
 			logger.Error("check judge fail: ", err)
 		}
 	}

src/modules/monapi/cron/event_consumer.go

@@ -58,7 +58,7 @@ func consume(event *model.Event) {
 			SetEventStatus(event, model.STATUS_UPGRADE)
 
 			if needNotify {
-				if event.EventType == config.ALERT && NeedCallback(event.Sid) {
+				if NeedCallback(event.Sid) {
 					if err := PushCallbackEvent(event); err != nil {
 						logger.Errorf("push event to callback queue failed, callbackEvent: %+v", event)
 					}
@@ -82,7 +82,7 @@ func consume(event *model.Event) {
 		return
 	}
 
-	if event.EventType == config.ALERT && NeedCallback(event.Sid) {
+	if NeedCallback(event.Sid) {
 		if err := PushCallbackEvent(event); err != nil {
 			logger.Errorf("push event to callback queue failed, callbackEvent: %+v", event)
 		}

src/modules/monapi/redisc/redis.go

@@ -7,6 +7,7 @@ import (
 	"github.com/toolkits/pkg/logger"
 
 	"github.com/didi/nightingale/src/modules/monapi/config"
+	"github.com/didi/nightingale/src/toolkits/stats"
 )
 
 var RedisConnPool *redis.Pool
@@ -16,6 +17,7 @@ func InitRedis() {
 
 	addr := cfg.Redis.Addr
 	pass := cfg.Redis.Pass
+	db := cfg.Redis.DB
 	maxIdle := cfg.Redis.Idle
 	idleTimeout := 240 * time.Second
 
@@ -29,6 +31,8 @@ func InitRedis() {
 		Dial: func() (redis.Conn, error) {
 			c, err := redis.Dial("tcp", addr, redis.DialConnectTimeout(connTimeout), redis.DialReadTimeout(readTimeout), redis.DialWriteTimeout(writeTimeout))
 			if err != nil {
+				logger.Errorf("conn redis err:%v", err)
+				stats.Counter.Set("redis.conn.failed", 1)
 				return nil, err
 			}
 
@@ -36,6 +40,16 @@ func InitRedis() {
 				if _, err := c.Do("AUTH", pass); err != nil {
 					c.Close()
 					logger.Error("redis auth fail, pass: ", pass)
+					stats.Counter.Set("redis.conn.failed", 1)
+					return nil, err
+				}
+			}
+
+			if db != 0 {
+				if _, err := c.Do("SELECT", db); err != nil {
+					c.Close()
+					logger.Error("redis select db fail, db: ", db)
+					stats.Counter.Set("redis.select.failed", 1)
 					return nil, err
 				}
 			}

src/modules/transfer/backend/query.go

@@ -12,6 +12,7 @@ import (
 	"github.com/didi/nightingale/src/dataobj"
 	"github.com/didi/nightingale/src/modules/transfer/calc"
 	"github.com/didi/nightingale/src/toolkits/address"
+	"github.com/didi/nightingale/src/toolkits/stats"
 
 	"github.com/toolkits/pkg/logger"
 	"github.com/toolkits/pkg/net/httplib"
@@ -95,15 +96,15 @@ func FetchDataForUI(input dataobj.QueryDataForUI) []*dataobj.TsdbQueryResponse {
 	//进行数据计算
 	aggrDatas := []*dataobj.TsdbQueryResponse{}
 	if input.AggrFunc != "" && len(resp) > 1 {
-		aggrData := &dataobj.TsdbQueryResponse{
-			Start: input.Start,
-			End:   input.End,
-		}
 
 		aggrCounter := make(map[string][]*dataobj.TsdbQueryResponse)
 		if len(input.GroupKey) == 0 || getTags(resp[0].Counter) == "" {
+			aggrData := &dataobj.TsdbQueryResponse{
+				Start:  input.Start,
+				End:    input.End,
+				Values: calc.Compute(input.AggrFunc, resp),
+			}
 			//没有聚合 tag, 或者曲线没有其他 tags, 直接所有曲线进行计算
-			aggrData.Values = calc.Compute(input.AggrFunc, resp)
 			aggrDatas = append(aggrDatas, aggrData)
 		} else {
 			for _, data := range resp {
@@ -132,9 +133,12 @@ func FetchDataForUI(input dataobj.QueryDataForUI) []*dataobj.TsdbQueryResponse {
 			}
 
 			for counter, datas := range aggrCounter {
-				aggrData.Counter = counter
+				aggrData := &dataobj.TsdbQueryResponse{
-				aggrData.Values = calc.Compute(input.AggrFunc, datas)
+					Start:   input.Start,
-
+					End:     input.End,
+					Counter: counter,
+					Values:  calc.Compute(input.AggrFunc, datas),
+				}
 				aggrDatas = append(aggrDatas, aggrData)
 			}
 		}
@@ -161,10 +165,12 @@ func fetchDataSync(start, end int64, consolFun, endpoint, counter string, step i
 	defer func() {
 		<-worker
 	}()
+	stats.Counter.Set("query.tsdb", 1)
 
 	data, err := fetchData(start, end, consolFun, endpoint, counter, step)
 	if err != nil {
 		logger.Warning(err)
+		stats.Counter.Set("query.data.err", 1)
 	}
 	dataChan <- data
 	return

src/modules/transfer/backend/sender.go

@@ -107,6 +107,7 @@ func Send2TsdbTask(Q *list.SafeListLimited, node string, addr string, concurrent
 
 // 将数据 打入 某个Tsdb的发送缓存队列, 具体是哪一个Tsdb 由一致性哈希 决定
 func Push2TsdbSendQueue(items []*dataobj.MetricValue) {
+	errCnt := 0
 	for _, item := range items {
 		tsdbItem := convert2TsdbItem(item)
 		stats.Counter.Set("tsdb.queue.push", 1)
@@ -118,13 +119,13 @@ func Push2TsdbSendQueue(items []*dataobj.MetricValue) {
 		}
 
 		cnode := Config.ClusterList[node]
-		errCnt := 0
 		for _, addr := range cnode.Addrs {
 			Q := TsdbQueues[node+addr]
 			if !Q.PushFront(tsdbItem) {
 				errCnt += 1
 			}
 		}
+	}
 
 	// statistics
 	if errCnt > 0 {
@@ -132,7 +133,6 @@ func Push2TsdbSendQueue(items []*dataobj.MetricValue) {
 		logger.Error("Push2TsdbSendQueue err num: ", errCnt)
 	}
 }
-}
 
 func Send2JudgeTask(Q *list.SafeListLimited, addr string, concurrent int) {
 	batch := Config.Batch
@@ -172,7 +172,7 @@ func Send2JudgeTask(Q *list.SafeListLimited, addr string, concurrent int) {
 
 			if !sendOk {
 				stats.Counter.Set("points.out.judge.err", 1)
-				logger.Errorf("send judge %s fail: %v", addr, err)
+				logger.Errorf("send %v to judge %s fail: %v", judgeItems, addr, err)
 			}
 
 		}(addr, judgeItems, count)
@@ -180,6 +180,7 @@ func Send2JudgeTask(Q *list.SafeListLimited, addr string, concurrent int) {
 }
 
 func Push2JudgeSendQueue(items []*dataobj.MetricValue) {
+	errCnt := 0
 	for _, item := range items {
 		key := str.PK(item.Metric, item.Endpoint)
 		stras := cache.StraMap.GetByKey(key)
@@ -203,12 +204,14 @@ func Push2JudgeSendQueue(items []*dataobj.MetricValue) {
 
 			q, exists := JudgeQueues.Get(stra.JudgeInstance)
 			if exists {
-				q.PushFront(judgeItem)
+				if !q.PushFront(judgeItem) {
+					errCnt += 1
 				}
 			}
-
 		}
 	}
+	stats.Counter.Set("judge.queue.err", errCnt)
+}
 
 // 打到Tsdb的数据,要根据rrdtool的特定 来限制 step、counterType、timestamp
 func convert2TsdbItem(d *dataobj.MetricValue) *dataobj.TsdbItem {

src/modules/transfer/http/routes/query_router.go

@@ -46,18 +46,50 @@ func QueryData(c *gin.Context) {
 func QueryDataForUI(c *gin.Context) {
 	stats.Counter.Set("data.ui.qp10s", 1)
 	var input dataobj.QueryDataForUI
-
+	var respData []*dataobj.QueryDataForUIResp
 	errors.Dangerous(c.ShouldBindJSON(&input))
+	start := input.Start
+	end := input.End
 
 	resp := backend.FetchDataForUI(input)
+	for _, d := range resp {
+		data := &dataobj.QueryDataForUIResp{
+			Start:    d.Start,
+			End:      d.End,
+			Endpoint: d.Endpoint,
+			Counter:  d.Counter,
+			DsType:   d.DsType,
+			Step:     d.Step,
+			Values:   d.Values,
+		}
+		respData = append(respData, data)
+	}
+
 	if len(input.Comparisons) > 1 {
 		for i := 1; i < len(input.Comparisons); i++ {
-			input.Start = input.Start - input.Comparisons[i]
+			comparison := input.Comparisons[i]
-			input.End = input.End - input.Comparisons[i]
+			input.Start = start - comparison
+			input.End = end - comparison
 			res := backend.FetchDataForUI(input)
-			resp = append(resp, res...)
+			for _, d := range res {
+				for j := range d.Values {
+					d.Values[j].Timestamp += comparison
+				}
+
+				data := &dataobj.QueryDataForUIResp{
+					Start:      d.Start,
+					End:        d.End,
+					Endpoint:   d.Endpoint,
+					Counter:    d.Counter,
+					DsType:     d.DsType,
+					Step:       d.Step,
+					Values:     d.Values,
+					Comparison: comparison,
+				}
+				respData = append(respData, data)
+			}
 		}
 	}
 
-	render.Data(c, resp, nil)
+	render.Data(c, respData, nil)
 }

src/modules/transfer/http/routes/routes.go

@@ -18,7 +18,7 @@ func Config(r *gin.Engine) {
 		sys.GET("/alive-judges", judges)
 
 		sys.POST("/push", PushData)
-		sys.POST("/data", QueryDataForJudge)
+		sys.POST("/data", QueryData)
 		sys.POST("/data/ui", QueryDataForUI)
 	}
 

src/modules/transfer/rpc/push.go

@@ -27,9 +27,10 @@ func (t *Transfer) Push(args []*dataobj.MetricValue, reply *dataobj.TransferResp
 		err := v.CheckValidity()
 		if err != nil {
 			stats.Counter.Set("points.in.err", 1)
-			logger.Warningf("item is illegal item:%s err:%v", v, err)
+			msg := fmt.Sprintf("item is illegal item:%s err:%v", v, err)
+			logger.Warningf(msg)
 			reply.Invalid += 1
-			reply.Msg += fmt.Sprintf("%v\n", err)
+			reply.Msg += msg
 			continue
 		}
 

src/modules/tsdb/index/index.go

@@ -47,7 +47,7 @@ func GetIndexLoop() {
 func GetIndex() {
 	instances, err := report.GetAlive("index", Config.HbsMod)
 	if err != nil {
-		stats.Counter.Set("index.get.err", 1)
+		stats.Counter.Set("get.index.err", 1)
 		logger.Warningf("get index list err:%v", err)
 		return
 	}

src/modules/tsdb/rpc/push.go

@@ -58,7 +58,6 @@ func handleItems(items []*dataobj.TsdbItem) {
 		//todo hash冲突问题需要解决
 		if err := cache.Caches.Push(item.Key, item.Timestamp, item.Value); err != nil {
 			stats.Counter.Set("points.in.err", 1)
-
 			logger.Warningf("push obj error, obj: %v, error: %v\n", items[i], err)
 			fail++
 		}

src/modules/tsdb/rrdtool/sync_disk.go

@@ -231,6 +231,7 @@ func FlushRRD(flushChunks map[interface{}][]*cache.Chunk) {
 
 				err := FlushFile(seriesID, items)
 				if err != nil {
+					stats.Counter.Set("flush.rrd.err", 1)
 					logger.Errorf("flush %v data to rrd err:%v", seriesID, err)
 					continue
 				}

src/toolkits/identity/identity.go

@@ -18,11 +18,12 @@ type IdentitySection struct {
 func Init(identity IdentitySection) {
 	if identity.Specify != "" {
 		Identity = identity.Specify
+		return
 	}
 
 	var err error
 	Identity, err = sys.CmdOutTrim("bash", "-c", identity.Shell)
 	if err != nil {
-		log.Fatalln("[F] cannot get hostname")
+		log.Fatalln("[F] cannot get identity")
 	}
 }

src/toolkits/stats/init.go

@@ -3,13 +3,17 @@ package stats
 import (
 	"bytes"
 	"encoding/json"
+	"fmt"
 	"net/http"
+	"path"
 	"time"
 
 	"github.com/didi/nightingale/src/dataobj"
-	"github.com/didi/nightingale/src/toolkits/identity"
+	"github.com/didi/nightingale/src/toolkits/address"
 
+	"github.com/toolkits/pkg/file"
 	"github.com/toolkits/pkg/logger"
+	"github.com/toolkits/pkg/runner"
 )
 
 var (
@@ -17,8 +21,14 @@ var (
 )
 
 func Init(prefix string, addr ...string) {
-	if len(addr) > 0 {
+	if len(addr) > 0 && addr[0] != "" {
+		//如果配置了 addr，使用 addr 参数
 		PushUrl = addr[0]
+
+	} else if file.IsExist(path.Join(runner.Cwd, "etc", "address.yml")) {
+		//address.yml 存在，则使用配置文件的地址
+		port := address.GetHTTPPort("collector")
+		PushUrl = fmt.Sprintf("http://127.0.0.1:%d/api/collector/push", port)
 	}
 
 	Counter = NewCounter(prefix)
@@ -42,7 +52,6 @@ func Push() {
 func NewMetricValue(metric string, value int64) *dataobj.MetricValue {
 	item := &dataobj.MetricValue{
 		Metric:       metric,
-		Endpoint:     identity.Identity,
 		Timestamp:    time.Now().Unix(),
 		ValueUntyped: value,
 		CounterType:  "GAUGE",

web/.dockerignore

@@ -0,0 +1 @@
+node_modules

web/Dockerfile

@@ -0,0 +1,7 @@
+FROM node:lts-alpine AS builder
+
+WORKDIR /app
+
+COPY . .
+
+RUN npm install

web/src/components/EndpointList/index.tsx

@@ -76,7 +76,10 @@ class index extends Component<Props, State> {
   processData = async (endpoints: Endpoint[]) => {
     if (this.state.displayBindNode && endpoints) {
       const idents = _.map(endpoints, item => item.ident);
-      const endpointNodes = await request(`${api.endpoint}s/bindings?idents=${_.join(idents, ',')}`);
+      let endpointNodes: any[] = [];
+      if (idents.length) {
+        endpointNodes = await request(`${api.endpoint}s/bindings?idents=${_.join(idents, ',')}`);
+      }
       const newEndpoints = _.map(endpoints, (item) => {
         const current = _.find(endpointNodes, { ident: item.ident });
         const nodes = _.get(current, 'nodes', []);

web/src/pages/ServiceTree/Node/index.tsx

@@ -184,7 +184,7 @@ class index extends Component<null, State> {
             <div className="mt10 mb10" style={{ wordBreak: 'break-word' }}>
               {_.get(selectedNode, 'path')}
             </div>
-            <Popconfirm title="确定要删除这个节点吗？" onConfirm={this.handleDelNode}>
+            <Popconfirm disabled={isPdlNode} title="确定要删除这个节点吗？" onConfirm={this.handleDelNode}>
               <Button disabled={isPdlNode}>删除</Button>
             </Popconfirm>
             {