Ad Widget

Collapse

Problems monitoring Windows machine (alerts not generated)

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • david
    Junior Member
    • Sep 2007
    • 3

    #1

    Problems monitoring Windows machine (alerts not generated)

    Hi all,

    I'm just testing out Zabbix and I seem to have hit some problems. Could someone confirm I'm doing things correctly?

    Set-up:
    Server - Zabbix 1.4.2 on Ubuntu with apache/MySQL
    Target - Windows XP SP2 with agent installed

    Using the standard templates I've got the monitor running and gathering data which I can see in the graphs.

    I'm trying to set up a test alert that monitors the disk space of the C: drive of the target and sends a warning when it drops below 11Gb. I'm using the standard Items "Template_Windows:Free Disk Space on C:" (vfs.fs.size[c:,free]) and I'm running it every 5 seconds.

    I can see the correct values appear in the "Monitoring - Latest Data" tab so I think this is working ok.

    I then set up a trigger based on "Template_Windows:Low free disk space on Davids PC volume c:" which reads "{Davids PC:vfs.fs.size[c:,free].last(0)}<11000".

    Finally, I've set up a number of actions (basically one from each of the possible trigger values. e.g. Host, Severity level, description, Value, Period) which should each send an email to my account when the trigger happens.

    The problem is that although the graphs clearly show the free space rise and fall there is no mail generated. I did manage to get it to mail me once, so the triggers and actions do look to be properly set, but I've had nothing after that despite not making any changes and I'm just manually copying and deleting a file to get me below/above the trigger value.

    What I really want to happen is to get a mail every time the free space drop, a follow up at given intervals while below the specified level (which I don't see any way of specifying although the manual seems to imply it exists) and a final mail when the space goes back above the trigger point.

    So can anyone tell me what I'm doing wrong?
  • surlyjake
    Junior Member
    • Jul 2007
    • 28

    #2
    so is the trigger itself working? (check this in Monitoring -> events)
    if so, does anything show up under Monitoring -> actions?

    Comment

    • david
      Junior Member
      • Sep 2007
      • 3

      #3
      Hi, thanks for the reply.

      Yes, I can see items in both the events and actions sections but only for the mails that were sent as mentioned above.

      I would have expected to see around 15 changes (that's how many times I went under/over the trigger point) and that should have generated about 6 emails for each on and off event (due to the number of actions set that would have reacted to the same trigger condition).

      Further to my explanation above I rebooted my pc on Friday (after I had posted) and Zabbix caught this and mailed me about the loss of connection, the connection coming back and the fact that the PC had restarted (my actions still had a 'host=Davids PC' condition set to trigger on any monitored change plus some others which triggered correctly as far as I can tell).

      Events log:

      2007.Sep.07 16:39:27 Davids PC has just been restarted OFF
      2007.Sep.07 16:29:27 Davids PC has just been restarted ON
      2007.Sep.07 16:28:55 Server Davids PC is unreachable OFF
      2007.Sep.07 16:26:49 Server Davids PC is unreachable ON
      2007.Sep.07 14:14:12 < 11Gb free disk space on Davids PC volume c: OFF
      2007.Sep.07 13:46:47 < 11Gb free disk space on Davids PC volume c: ON
      2007.Sep.07 13:43:52 < 11Gb free disk space on Davids PC volume c: OFF
      2007.Sep.07 13:42:52 < 11Gb free disk space on Davids PC volume c: ON
      2007.Sep.07 13:36:44 < 11Gb free disk space on Davids PC volume c: OFF
      2007.Sep.07 13:34:01 < 11Gb free disk space on Davids PC volume c: ON


      Actions: I won't post the log but there are mails sent for the above events.


      So to recap, I received these emails just as the events and actions logs show and I would have expected but that only accounts for 3 out of 15 changes that should have activated the trigger. None of the disk space changes made between 2.15pm and 4.25pm were recorded but the later loss of connectivity was.

      Comment

      • surlyjake
        Junior Member
        • Jul 2007
        • 28

        #4
        Only thing i can think of is A spam filter somewhere along the way. If it isn't that.... I dont have any ideas

        Comment

        • david
          Junior Member
          • Sep 2007
          • 3

          #5
          Thanks for the idea Jake but I've just had a quick look at the mailserver spam folder and that doesn't seem to be the case. I don't think the missing alerts were ever generated.

          Not sure where I can go from here though...

          Comment

          • Henrik Jakobsen
            Junior Member
            • Aug 2007
            • 2

            #6
            Outgoing mail logs

            Try checking your outgoing mail log to see if any alert mails are sent

            Comment

            Working...