Ad Widget

Collapse

ZABBIX just check once after reboot, than stop working

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • lippoliv
    Junior Member
    • Oct 2011
    • 15

    #1

    ZABBIX just check once after reboot, than stop working

    Hey there,
    my plan is to use zabbix to monitor 61 URLs of mine, that number will grow and grow and grow

    So now I got an Problem: Zabbix checks all WEB-Szenarios once, but then doesn't check them anymore!?


    So an URL should be checked each 150 seconds. It is checked directly after reboot, and then never more.


    Any Ideas. I will give you everything you need to analyse my problem.

    Thanks, Oli
  • lippoliv
    Junior Member
    • Oct 2011
    • 15

    #2
    I did check syslog, I did check ZABBIX-Log.... but didn't find something interesting. Its ZABBIX 2.0.9 ^^

    Comment

    • lippoliv
      Junior Member
      • Oct 2011
      • 15

      #3
      So guys I did the following:
      1. Updated to ZABBIX 2.2
      2. DISABLED all 64 WEB-Scenarios
      3. ENABLE 5 WEB Scenarios
      4. wait till they where processed (worked)
      5. ENABLE another 5 Scenarios
      6. repeat last 2 steps till all will be activated (still in process)


      But: That can't be the truth? I also managed a little in the zabbix-conf, but that doesn't changed something, that was befor step 1.
      I set to debug output, so May if the problem occures twice, I can have a look in the debug-log.


      It would be nice if someone watch this over

      Comment

      • lippoliv
        Junior Member
        • Oct 2011
        • 15

        #4
        Another Point:

        In the Server-Stats I see, that the whole night the trafic was like this:

        0 5.45.98.96 166,75 MB 101,43 MB 268,18 MB
        1 5.45.98.96 166,93 MB 101,54 MB 268,47 MB
        2 5.45.98.96 166,82 MB 101,51 MB 268,33 MB
        3 5.45.98.96 167,30 MB 101,77 MB 269,07 MB
        4 5.45.98.96 167,47 MB 101,91 MB 269,38 MB
        5 5.45.98.96 167,47 MB 102,11 MB 269,57 MB
        6 5.45.98.96 167,09 MB 101,66 MB 268,75 MB
        7 5.45.98.96 166,71 MB 101,41 MB 268,12 MB
        8 5.45.98.96 166,20 MB 101,10 MB 267,29 MB
        9 5.45.98.96 149,74 MB 137,92 MB 287,67 MB

        since 9 oclock I'am working on the machine. This Server does nothing else than ZABBIX, so if the traffic is that high, can it be, that the monitoring worked, but the writing to the DB doesn't?


        EDIT

        When I updated some Scenarios of the 5th node, it starts "hanging". In the log i find

        3040:20131118:103317.948 In DCmass_add_history()
        3040:20131118:103317.948 query [txnlev:1] [insert into history (itemid,clock,ns,value) values (23600,1384767194,30995035,53661.000000),(23601,13 84767194,30995035,0.079591),(23597,1384767194,3138 6814,53661.000000);
        insert into history_uint (itemid,clock,ns,value) values (23602,1384767194,30995035,200),(23598,1384767194, 31386814,0);
        ]
        3041:20131118:103317.949 history syncer #2 [synced 0 items in 0.000034 sec, syncing history]
        3041:20131118:103317.949 In DCsync_history() history_first:761 history_num:0
        3041:20131118:103317.949 history syncer #2 [synced 0 items in 0.000038 sec, idle 5 sec]
        3049:20131118:103317.949 history syncer #4 [synced 0 items in 0.000036 sec, syncing history]
        3049:20131118:103317.949 In DCsync_history() history_first:761 history_num:0
        3049:20131118:103317.949 history syncer #4 [synced 0 items in 0.000035 sec, idle 5 sec]
        3040:20131118:103317.949 In zbx_vc_add_value() itemid:23600 value_type:0 timestamp:1384767194.30995035
        3040:20131118:103317.949 End of zbx_vc_add_value():FAIL
        3040:20131118:103317.949 In zbx_vc_add_value() itemid:23601 value_type:0 timestamp:1384767194.30995035
        3040:20131118:103317.949 End of zbx_vc_add_value():FAIL
        3040:20131118:103317.949 In zbx_vc_add_value() itemid:23602 value_type:3 timestamp:1384767194.30995035
        3040:20131118:103317.950 End of zbx_vc_add_value():FAIL
        3040:20131118:103317.950 In zbx_vc_add_value() itemid:23597 value_type:0 timestamp:1384767194.31386814
        3040:20131118:103317.950 End of zbx_vc_add_value():FAIL
        3040:20131118:103317.950 In zbx_vc_add_value() itemid:23598 value_type:3 timestamp:1384767194.31386814
        3040:20131118:103317.950 End of zbx_vc_add_value():SUCCEED
        3040:20131118:103317.950 End of DCmass_add_history()
        Looks like it is similar to https://svn.zabbix.com/browse/ZBX-7363


        RESOLVED

        Don't know how stable it is, because I don't know if its the true reason but:

        I could activate all Web-Scenarios expect that one of the 5th node. So after all was running I "step into" the 5th node and activeted Scenario by scenario. Everytime watching if ZABBIX hangs or not.

        After some Scenarios I wasnt sure if I did spell an URL correct and visit this URL manually. "Endless redirect" was the result. That was an DNS Redirect to its own URL. After fixing this up, the redirect wasnt endless and so I could activate ALL my WEB-Scenarios again.


        I did modifyed the Log-Level back (so its not debug anymore) and reboot the VPS. So now I did watch the web-scenario-list if its still working.


        May someone did the same mistake as I.


        For later discuss: Why did ZABBIX stop working on endless redirect? I had to reboot for getting ZABBIX working again? Why isn't there an seperated thread to process this "mistaken" scenario?
        Last edited by lippoliv; 18-11-2013, 12:29.

        Comment

        Working...