第1章 Zabbix簡介及組成

1.1 zabbix簡介

zabbix是一個基於web界面，提供分散式系統監視以及網路監視功能的企業級的開源解決方案。它可以監視各種網路參數，保證伺服器自動的安全運營，並提供靈活的通知機制以讓系統管理員快速定位/解決存在的各種問題

1.1 zabbix組成

zabbix server和zabbix agent，可選組件zabbix proxy

可以通過SNMP、zabbix agent、fping、埠監視等方法對遠程伺服器或網路狀態完成監視/數據收集等功能。支持linux及類unix、windows平臺只能安裝客戶端(被監控)

第2章 Zabbix 3.0.13服務端安裝

zabbix server3.0無法centos6上進行yum安裝，故我們要在centos7上進行yum安裝。如果一定要在centos6上進行安裝，則強烈建議通過源碼方式進行編譯安裝，同時還需要註意PHP的版本

註意：儘管zabbix server3.0在centos6上不能進行yum安裝，但zabbix agent3.0在centos6上是可進行yum安裝的

第3章 Web頁面報錯總結

3.1 問題一Zabbix alerter processes more than 75% busy

問題原因：

zabbix伺服器郵件進程繁忙導致的，一般是因為設置動作的間隔太短。特殊情況下會產生大量告警，如伺服器發幾萬封郵件過程中，郵件進程發掛了

解決方案：

01.刪除資料庫解決(風險較大，不建議)

02.修改郵件腳本，將郵件的動作改為列印時間，等待郵件完全釋放再改回來，如下

 1 [root@m01 ~]# cat /usr/lib/zabbix/alertscripts/sms 2 3 #!/bin/bash 4 5 echo `date` >>/tmp/sms.txt 

3.2 問題二Zabbix discoverer processes more than 75% busy

問題原因：

01.配置了discovery自動發現任務，配置的每個discovery任務在一定時間內占用1個進程，而zabbix_server.conf中預設配置只有1個discovery(被註釋，預設生效)

02.為了快速驗證自動發現效果，將discovery任務的"Delay"由預設3600s設置成60s

解決方案：

01.修改配置文件中的StartDiscoverers進程數量，取消其之前的#號並將數值修改為5，最後重啟服務

(註：根據系統硬體配置，可以設置成更高的數值，但其範圍為0~250)

1 [root@m01 ~]# grep 'StartDiscoverers' /etc/zabbix/zabbix_server.conf
2 
3 ### Option: StartDiscoverers
4 
5 StartDiscoverers=5
6 
7 [root@m01 ~]# systemctl restart zabbix-server.service

02.編寫定時任務腳本重啟zabbix_server來降低負載

1 [root@m01 ~]# crontab -e
2 
3 @daily service zabbix-server restart > /dev/null 2>&1
4 
5 #計劃會每天自動重啟Zabbix服務以結束僵屍進程並清理記憶體等

3.3 問題三Zabbix poller processes more than 75% busy

問題原因：

01.通過Zabbix agent採集數據的設備死機或其他原因導致zabbix agent死掉server獲取不到數據

02. server向agent獲取數據時時間過長，超過了server設置的timeout時間

解決方案：

01.增加Zabbix Server啟動時初始化的進程數量

 1 ### Option: StartPollers 2 3 StartPollers=10 #改成多少取決於伺服器的性能和監控的數量，如果記憶體足夠的話可以設置更高 

02.修改模板自動發現規則中的保留失去的資源期間為0

3.4 問題四Zabbix housekeeper processes more than 75% busy

問題原因：

為了防止資料庫持續增大，zabbix有自動刪除歷史數據的機制即housekeeper，而mysql刪除數據時性能會降低，就會報錯

解決方案：

調整HousekeepingFrequency參數

 1 HousekeepingFrequency=12 #間隔時間 2 3 MaxHousekeeperDelete=1000000 #最大刪除量 

3.5 問題五Zabbix server記憶體溢出，無法啟動

問題原因：

zabbix使用一段時間後，再次加入一批交換機監控，zabbix-server將無法啟動，查看日誌顯示如下(提示記憶體溢出，需調整zabbix伺服器配置zabbix_server.conf)

1 2816:20170725:174352.675 [file:dbconfig.c,line:652] zbx_mem_realloc(): out of memory (requested 162664 bytes)
2 
3 2816:20170725:174352.675 [file:dbconfig.c,line:652] zbx_mem_realloc(): please increase CacheSize configuration parameter

解決方案：

 1 vim zabbix_server.conf 2 3 CacheSize=1024M #預設為8M 

3.6 PHP Fatal error: Allowed memory size of 134217728 bytes exhausted (tried to allocate 11 bytes)

問題原因：

zabbix某些頁面無法打開，查看php日誌發現，當訪問這個頁面時報錯記憶體不足

解決方案：

不清楚是否記憶體泄露，最簡單的方法是調大php進程的可用記憶體

 1 [root@zabbix-master ~]# grep 'memory_limit' /etc/httpd/conf.d/zabbix.conf 2 3 php_value memory_limit 512M #預設128M 

3.7總結

此處筆者提供一份配置文件，其中的參數數值可供大家修改

  1 # This is a configuration file for Zabbix server daemon
  2 # To get more information about Zabbix, visit http://www.zabbix.com
  3 ############ GENERAL PARAMETERS #################
  4 ### Option: ListenPort
  5 #    Listen port for trapper.
  6 #
  7 # Mandatory: no
  8 # Range: 1024-32767
  9 # Default:
 10 #說明:服務端監聽埠用於接收二級代理或直連AGENT的採集數據
 11 # ListenPort=10051
 12 ### Option: SourceIP
 13 #    Source IP address for outgoing connections.
 14 #
 15 # Mandatory: no
 16 # Default:
 17 #說明：服務端監聽IP，建議指定
 18 # SourceIP=
 19 ### Option: LogType
 20 #    Specifies where log messages are written to:
 21 #        system  - syslog
 22 #        file    - file specified with LogFile parameter
 23 #        console - standard output
 24 #
 25 # Mandatory: no
 26 # Default:
 27 # LogType=file
 28 ### Option: LogFile
 29 #    Log file name for LogType 'file' parameter.
 30 #
 31 # Mandatory: no
 32 # Default:
 33 # LogFile=
 34 #說明：zabbix服務端日誌路徑，視具體情況指定
 35 LogFile=/tmp/zabbix_server.log
 36 ### Option: LogFileSize
 37 #    Maximum size of log file in MB.
 38 #    0 - disable automatic log rotation.
 39 #
 40 # Mandatory: no
 41 # Range: 0-1024
 42 # Default:
 43 #說明：日誌達到多少M里就輪轉；若此參數值為0時，則不輪轉，日誌將不斷變大，建議設置成輪轉
 44 LogFileSize=50
 45 ### Option: DebugLevel
 46 #   Specifies debug level:
 47 #   0 - basic information about starting and stopping of Zabbix processes
 48 #   1 - critical information  災難日誌，日誌量較少
 49 #   2 - error information  錯誤級別，日誌量大於CRITICAL級別
 50 #   3 - warnings    告警級別，日誌量大於ERROR級別
 51 #   4 - for debugging (produces lots of information)調試級別，日誌量大於WARNING
 52 #   5 - extended debugging (produces even more information)
 53 #
 54 # Mandatory: no
 55 # Range: 0-5
 56 # Default:
 57 #說明：日誌級別0~4，單位時間內生成日誌的量不斷增大
 58 DebugLevel=3
 59 ### Option: PidFile
 60 #    Name of PID file.
 61 #
 62 # Mandatory: no
 63 # Default:
 64 #說明：zabbix服務端程式PID路徑
 65 PidFile=/tmp/zabbix_server.pid
 66 ### Option: DBHost
 67 #    Database host name.
 68 #    If set to localhost, socket is used for MySQL.
 69 #    If set to empty string, socket is used for PostgreSQL.
 70 #
 71 # Mandatory: no
 72 # Default:
 73 #說明：指定資料庫信息，對於mysql，若設置為localhost則mysql用SOCKET來連接(需配合參數 DBSocket 使用)，否則用IP連接；若DHHOST值為空，則預設連接PostgreSQL
 74 # DBHost=localhost
 75 DBHost=
 76 ### Option: DBName
 77 #    Database name.
 78 #    For SQLite3 path to database file must be provided. DBUser and DBPassword are ignored.
 79 #
 80 # Mandatory: yes
 81 # Default:
 82 # DBName=
 83 #說明：服務端連接資料庫的庫名
 84 DBName=
 85 ### Option: DBSchema
 86 #    Schema name. Used for IBM DB2 and PostgreSQL.
 87 #
 88 # Mandatory: no
 89 # Default:
 90 #說明：專門用於 IBM DB2資料庫的連接信息
 91 # DBSchema=
 92 ### Option: DBUser
 93 #    Database user. Ignored for SQLite.
 94 #
 95 # Mandatory: no
 96 # Default:
 97 #說明：連接資料庫的用戶
 98 # DBUser=
 99 DBUser=
100 ### Option: DBPassword
101 #    Database password. Ignored for SQLite.
102 #    Comment this line if no password is used.
103 #
104 # Mandatory: no
105 # Default:
106 #說明：連接資料庫的密碼
107 DBPassword=
108 ### Option: DBSocket
109 #    Path to MySQL socket.
110 #
111 # Mandatory: no
112 # Default:
113 #說明：指定MYSQL的SOCK連接路徑
114 DBSocket=/tmp/mysql.sock
115 ### Option: DBPort
116 #    Database port when not using local socket. Ignored for SQLite.
117 #
118 # Mandatory: no
119 # Range: 1024-65535
120 # Default (for MySQL):
121 #說明：指定連接資料庫的埠，預設3306
122 DBPort=3306
123 ############ ADVANCED PARAMETERS ################
124 #高級參數
125 ### Option: StartPollers
126 #    Number of pre-forked instances of pollers.
127 #
128 # Mandatory: no
129 # Range: 0-1000
130 # Default:
131 #說明;初始化時，啟動子進程數量，數量越多，則服務端吞吐能力越強，對系統資源消耗越大
132 StartPollers=300
133 ### Option: StartIPMIPollers
134 #    Number of pre-forked instances of IPMI pollers.
135 #
136 # Mandatory: no
137 # Range: 0-1000
138 # Default:
139 #
140 #說明:主要用於IPmi技術用於獲取硬體狀態場景。若無相關監控項，建議設置為0
141 # StartIPMIPollers=0
142 ### Option: StartPollersUnreachable
143 #    Number of pre-forked instances of pollers for unreachable hosts (including IPMI and Java).
144 #    At least one poller for unreachable hosts must be running if regular, IPMI or Java pollers
145 #    are started.
146 #
147 # Mandatory: no
148 # Range: 0-1000
149 # Default:
150 #說明：預設情況下，ZABBIX會啟用指定進程用於探測某些不可達主機的（含IPMI場景）；若使用場景中含有代理端，建議保持預設；若直接agent較多，可視具體情況調整
151 StartPollersUnreachable=50
152 ### Option: StartTrappers
153 #    Number of pre-forked instances of trappers.
154 #    Trappers accept incoming connections from Zabbix sender, active agents and active proxies.
155 #    At least one trapper process must be running to display server availability and view queue
156 #    in the frontend.
157 #
158 # Mandatory: no
159 # Range: 0-1000
160 # Default:
161 #說明：用於設置諸如SNMP STRAPPER場景提交來的數據的接收進程數，若客戶機SNMP TRAPPER技術較多，建議加大此參數值
162 StartTrappers=50
163 ### Option: StartPingers
164 #    Number of pre-forked instances of ICMP pingers.
165 #
166 # Mandatory: no
167 # Range: 0-1000
168 # Default:
169 #說明：用於設置啟用icmp協議PING主機方式啟動線程數量，若單台代理所管理機器超過500台，建議加大此數值
170 # StartPingers=10
171 ### Option: StartDiscoverers
172 #    Number of pre-forked instances of discoverers.
173 #
174 # Mandatory: no
175 # Range: 0-250
176 # Default:
177 #說明：用於設置自動發現主機的線程數量，若單台代理所管理機器超過500台，可以考慮加大此數值（僅適用於直接AGENT場景）
178 StartDiscoverers=15
179 ### Option: StartHTTPPollers
180 #    Number of pre-forked instances of HTTP pollers.
181 #
182 # Mandatory: no
183 # Range: 0-1000
184 # Default:
185 #說明：用於設置WEB撥測監控線程數，可視具體情況增加或減少此數值。
186 # StartHTTPPollers=1
187 ### Option: StartTimers
188 #    Number of pre-forked instances of timers.
189 #    Timers process time-based trigger functions and maintenance periods.
190 #    Only the first timer process handles the maintenance periods.
191 #
192 # Mandatory: no
193 # Range: 1-1000
194 # Default:
195 #說明：各實例計時器數量，主要用於觸發器，標有維護標識的主機，但只第一個計時器用於計算維護標識主機。
196 # StartTimers=1
197 ### Option: StartEscalators
198 #    Number of pre-forked instances of escalators.
199 #
200 # Mandatory: no
201 # Range: 0-100
202 # Default:
203 #說明：用於處理動作中的步驟的進程，zabbix動作較多時建議調大。
204 StartEscalators=30
205 ### Option: JavaGateway
206 #    IP address (or hostname) of Zabbix Java gateway.
207 #    Only required if Java pollers are started.
208 #
209 # Mandatory: no
210 # Default:
211 #說明：JAVAGATEWAY 場景下使用
212 JavaGateway=10.238.0.180
213 ### Option: JavaGatewayPort
214 #    Port that Zabbix Java gateway listens on.
215 #
216 # Mandatory: no
217 # Range: 1024-32767
218 # Default:
219 #說明：JAVAGATEWAY 場景下使用
220 JavaGatewayPort=10052
221 ### Option: StartJavaPollers
222 #    Number of pre-forked instances of Java pollers.
223 #
224 # Mandatory: no
225 # Range: 0-1000
226 # Default:
227 #說明：JAVAGATEWAY 場景下使用
228 StartJavaPollers=30
229 ### Option: StartVMwareCollectors
230 #    Number of pre-forked vmware collector instances.
231 #
232 # Mandatory: no
233 # Range: 0-250
234 # Default:
235 #說明：用於設置監控VMWARE Esxi主機實例時使用，若為0則不啟用，若要監控ESXI主機，此值最少為1 ；視監控ESXI數量設置對應數值
236 # StartVMwareCollectors=0
237 ### Option: VMwareFrequency
238 #    How often Zabbix will connect to VMware service to obtain a new data.
239 #
240 # Mandatory: no
241 # Range: 10-86400
242 # Default:
243 #說明：代理端訪問 VMWARE service的頻率，單位:秒
244 # VMwareFrequency=60
245 ### Option: VMwarePerfFrequency
246 #    How often Zabbix will connect to VMware service to obtain performance data.
247 #
248 # Mandatory: no
249 # Range: 10-86400
250 # Default:
251 # VMwarePerfFrequency=60
252 ### Option: VMwareCacheSize
253 #    Size of VMware cache, in bytes.
254 #    Shared memory size for storing VMware data.
255 #    Only used if VMware collectors are started.
256 #
257 # Mandatory: no
258 # Range: 256K-2G
259 # Default:
260 #說明：划出多少共用記憶體用於存儲VMWARE數據
261 VMwareCacheSize=256M
262 ### Option: VMwareTimeout
263 #    Specifies how many seconds vmware collector waits for response from VMware service.
264 #
265 # Mandatory: no
266 # Range: 1-300
267 # Default:
268 #說明：等待VMWare返回數據的最長時間
269 VMwareTimeout=10
270 ### Option: SNMPTrapperFile
271 #    Temporary file used for passing data from SNMP trap daemon to the server.
272 #    Must be the same as in zabbix_trap_receiver.pl or SNMPTT configuration file.
273 #
274 # Mandatory: no
275 # Default:
276 #說明：指定SNMP TRAPPER 時的臨時文件，用於代理端啟用SNMP TRAPPER功能時使用
277 # SNMPTrapperFile=/tmp/zabbix_traps.tmp
278 ### Option: StartSNMPTrapper
279 #    If 1, SNMP trapper process is started.
280 #
281 # Mandatory: no
282 # Range: 0-1
283 # Default:
284 #說明：是否啟用 snmptrapper功能 ，預設不啟用=0，啟用=1（配合參數SNMPTrapperFile使用）
285 # StartSNMPTrapper=0
286 ### Option: ListenIP
287 #    List of comma delimited IP addresses that the trapper should listen on.
288 #    Trapper will listen on all network interfaces if this parameter is missing.
289 #
290 # Mandatory: no
291 # Default:
292 #說明：啟用SNMPTRAPPER里 ，接收端監聽的IP，此參數與StartSNMPTrapper，SNMPTrapperFile 聯合使用
293 # ListenIP=0.0.0.0
294 ListenIP=10.238.0.180
295 ### Option: HousekeepingFrequency
296 #    How often Zabbix will perform housekeeping procedure (in hours).
297 #    Housekeeping is removing outdated information from the database.
298 #    To prevent Housekeeper from being overloaded, no more than 4 times HousekeepingFrequency
299 #    hours of outdated information are deleted in one housekeeping cycle, for each item.
300 #    To lower load on server startup housekeeping is postponed for 30 minutes after server start.
301 #    With HousekeepingFrequency=0 the housekeeper can be only executed using the runtime control option.
302 #    In this case the period of outdated information deleted in one housekeeping cycle is 4 times the
303 #    period since the last housekeeping cycle, but not less than 4 hours and not greater than 4 days.
304 #
305 # Mandatory: no
306 # Range: 0-24
307 # Default:
308 #說明：多少小時清理一次代理端資料庫的 history, alert, and alarms，以保持代理端資料庫輕便，建議保持預設
309 HousekeepingFrequency=24
310 ### Option: MaxHousekeeperDelete
311 #    The table "housekeeper" contains "tasks" for housekeeping procedure in the format:
312 #    [housekeeperid], [tablename], [field], [value].
313 #    No more than 'MaxHousekeeperDelete' rows (corresponding to [tablename], [field], [value])
314 #    will be deleted per one task in one housekeeping cycle.
315 #    SQLite3 does not use this parameter, deletes all corresponding rows without a limit.
316 #    If set to 0 then no limit is used at all. In this case you must know what you are doing!
317 #
318 # Mandatory: no
319 # Range: 0-1000000
320 # Default:
321 #說明:每個HouseKeeper任務刪除的最大記錄數，1.8.2開始支持
322 MaxHousekeeperDelete=1000000
323 ### Option: SenderFrequency
324 #    How often Zabbix will try to send unsent alerts (in seconds).
325 #
326 # Mandatory: no
327 # Range: 5-3600
328 # Default:
329 #說明:多少秒後重試發送失敗的報警信息
330 SenderFrequency=30
331 ### Option: CacheSize
332 #    Size of configuration cache, in bytes.
333 #    Shared memory size for storing host, item and trigger data.
334 #
335 # Mandatory: no
336 # Range: 128K-8G
337 # Default:
338 #說明;zabbix初始化時占用多少系統共用記憶體用於存儲配置信息，HOST,ITEM,TRIGGER數據，視監控主機數量和監控項調整，建議調整到32M或者更大
339 CacheSize=8G
340 ### Option: CacheUpdateFrequency
341 #    How often Zabbix will perform update of configuration cache, in seconds.
342 #
343 # Mandatory: no
344 # Range: 1-3600
345 # Default:
346 #說明：zabbix更新操作系統CACHE頻率，若管理頁面操作不頻繁，可以考慮加大參數值
347 CacheUpdateFrequency=300
348 ### Option: StartDBSyncers
349 #    Number of pre-forked instances of DB Syncers.
350 #
351 # Mandatory: no
352 # Range: 1-100
353 # Default:
354 #說明:將採集數據從CACHE同步到資料庫線程數量，視資料庫伺服器I/O繁忙情況，和資料庫寫能力調整。數值越大，寫能力越強。對資料庫伺服器I/O壓力越大。
355 StartDBSyncers=20
356 ### Option: HistoryCacheSize
357 #    Size of history cache, in bytes.
358 #    Shared memory size for storing history data.
359 #
360 # Mandatory: no
361 # Range: 128K-2G
362 # Default:
363 #說明：用於設置劃分多少系統共用記憶體用於存儲採集的歷史數據，此數值越大，資料庫讀壓力越小
364 HistoryCacheSize=2048M
365 ### Option: HistoryIndexCacheSize
366 #    Size of history index cache, in bytes.
367 #    Shared memory size for indexing history cache.
368 #
369 # Mandatory: no
370 # Range: 128K-2G
371 # Default:
372 #說明：3.0.0開始支持，歷史索引大小，一個監控項需要100bytes來存儲
373 HistoryIndexCacheSize=2048M
374 ### Option: TrendCacheSize
375 #    Size of trend cache, in bytes.
376 #    Shared memory size for storing trends data.
377 #
378 # Mandatory: no
379 # Range: 128K-2G
380 # Default:
381 #說明：用於設置劃分多少系統共用記憶體用於存儲計算出來的趨勢數據，此參數值從一定程度上可影響資料庫讀壓力
382 TrendCacheSize=512M
383 ### Option: ValueCacheSize
384 #    Size of history value cache, in bytes.
385 #    Shared memory size for caching item history data requests.
386 #    Setting to 0 disables value cache.
387 #
388 # Mandatory: no
389 # Range: 0,128K-64G
390 # Default:
391 #說明：划出系統多少共用記憶體用於已請求的存儲監控項信息，若監控項較多，建議加大此數值
392 ValueCacheSize=16G
393 ### Option: Timeout
394 #    Specifies how long we wait for agent, SNMP device or external check (in seconds).
395 #
396 # Mandatory: no
397 # Range: 1-30
398 # Default:
399 # Timeout=3
400 #說明：與AGNET\SNMP設備和其它外部設備通信超時設置，單位為秒；若採集數據不完整或網路繁忙，或從管理頁面發現客戶端狀態變化頻繁，可以考慮加大此數值。註意若此數值加大，應該考慮參數 StartPollers 是否有相應加大的必要。
401 Timeout=10
402 ### Option: TrapperTimeout
403 #    Specifies how many seconds trapper may spend processing new data.
404 #
405 # Mandatory: no
406 # Range: 1-300
407 # Default:
408 #說明：啟用 trapper功能，用於進程等待超時設置。根據需要調整
409 TrapperTimeout=50
410 ### Option: UnreachablePeriod
411 #    After how many seconds of unreachability treat a host as unavailable.
412 #
413 # Mandatory: no
414 # Range: 1-3600
415 # Default:
416 #說明：當AGNET端處於不可用狀態下，間隔多少秒後，嘗試重新連接。建議根據具體情況設置。註意，若此數值過小，右agent端業務系統繁忙時，有可能造成報警信息誤報
417 # UnreachablePeriod=45
418 ### Option: UnavailableDelay
419 #    How often host is checked for availability during the unavailability period, in seconds.
420 #
421 # Mandatory: no
422 # Range: 1-3600
423 # Default:
424 #說明:當AGENT端處於可用狀態下，間隔多少秒後，進行狀態檢查。若出現可正常採集數據，但管理頁面AGENT狀態不正常；若在網路，埠等均通暢情況下，AGENT狀態仍不正常，可以考慮加大此數值
425 # UnavailableDelay=60
426 ### Option: UnreachableDelay
427 #    How often host is checked for availability during the unreachability period, in seconds.
428 #
429 # Mandatory: no
430 # Range: 1-3600
431 # Default:
432 #說明：當agent端處於不可達狀態下，延遲多少秒後，進行重新嘗試，建議保持預設，在AGENT接入調試階段，可考慮減少此數值
433 # UnreachableDelay=15
434 ### Option: AlertScriptsPath
435 #    Full path to location of custom alert scripts.
436 #    Default depends on compilation options.
437 #
438 # Mandatory: no
439 # Default:
440 #說明：監控報警腳本路徑，非研發人員不建議修改此參數值
441 # AlertScriptsPath=${datadir}/zabbix/alertscripts
442 AlertScriptsPath=/home/zabbix/bin
443 ### Option: ExternalScripts
444 #    Full path to location of external scripts.
445 #    Default depends on compilation options.
446 #
447 # Mandatory: no
448 # Default:
449 #說明：自定義腳本存儲路徑，非研發人員不建議修改此參數值
450 # ExternalScripts=${datadir}/zabbix/