none
Logical Disk Free Space alert issue

    السؤال

  • Hi,

    My scom server doesn't generate Logical Disk Free Space alert on warning percent threshold(which is default).Below is the threshold setting on my scom server:

    The default threshold values for the Logical Disk Free Space monitoring routine include:

    System Drive Free Space Thresholds (Defaults)

    Parameter

    Default Value

    System Drive Error Mbytes Threshold

    100

    System Drive Error Percent Threshold

    5

    System Drive Warning Mbytes Threshold

    200

    System Drive Warning Percent Threshold

    10

    Non-System Drive Free Space Thresholds (Defaults)

    Parameter

    Default Value

    Non-System Drive Error Mbytes Threshold

    1000

    Non-System Drive Error Percent Threshold

    5

    Non-System Drive Warning Mbytes Threshold

    2000

    Non-System Drive Warning Percent Threshold

    10

         

    And let me know one more thing if Error percent threshold is 5% then it should be generate on this threshold but most of time error alert generated like this:

    The disk F: on computer server-1.abc.com is running out of disk space. The values that exceeded the threshold are 0% free space and 301 free Mbytes. 

    while the total space of drive F is 95GB.Idealy Error alert should be generated when the free disk space will remain 4.75GB and warning on 9.5GB.

    Please suggest why this happen & whats the resolution.

    thanx

     


    Syed
    22/جمادى الأولى/1432 12:44 م

الإجابات

  • Hi, actually it depends on what your environment is normally like.

    If all data disks are about the same in size, or that you always want to be notified at around 5GB of free space as a warning you could even set the percentage to a higher level than the MB threshold. That way you know that the alert is just waiting for the MB threshold to pass and you will have it.

    In your example. All non-system disks being 50 GB. If you want to be warning at 5 Gb and critical at 2,5 GB you can use 10% and 5% and 5120 and 2560 MB. This will cause you to pass both percent and MB threshold at the same time. Thats OK. Or you could set one of them higher so they are for sure over threshold (so percentages at 12 and 7% for instance or much higher). In that case you are basically monitoring for the amount of MB left on the system as your threshold.

    Syed, in your example the warning levels are at 5Gb and critical at 2,5GB. You have to pass both yellow thresholds to get yellow and both red thresholds to get to a red state.

    The other question you ask one of us is about the timing of the monitor. It does run once every hour. Which is generally enough for disks. However in your current case with the low threshold of only 1 GB for the data disk and if you have big chunks of data you are likely to go from green state to red state (or full disk) in a very short time. Possibly shorter than that one hour. We assume you already size your disks in a way that they dont fill up in minutes, but you should plan your alerts (thresholds) accordingly. SO for the big disks (50 and 100 Gb) it is normal to set the MB thresholds to 10 and 5 GB for instance. Just an example.

    About the alerting: This monitor does change state at yellow and red. But it alerts as alert only at critical state.  You can also set an override for that on the same monitor (alert on state - set it to the one that lists both warning and error).


    Bob Cornelissen - BICTT (My BICTT Blog)
    • تم الاقتراح كإجابة بواسطة Bob CornelissenMVP, Moderator 23/جمادى الأولى/1432 05:51 ص
    • تم وضع علامة كإجابة بواسطة Syed-DC 23/جمادى الأولى/1432 09:21 ص
    22/جمادى الأولى/1432 05:35 م

جميع الردود

  • it's an AND statement. So for a non-OS disk you get a critical at less than 5% AND less than 1GB free.

    To change this behaviour you got to override the thresholds (the mb threshold set to 5 resp. 10GB would do the trick for that disk).


    Rob Korving
    http://jama00.wordpress.com/
    22/جمادى الأولى/1432 12:48 م
  • Hi Syed,

    That is because this is a combined monitor.

    You actually have to go over BOTH thresholds for MB and percent. SO when you have a big disk like you have you will go over the percent thresholds first for warning and error and in the end you will pass the MB thresholds of 2 and 1 GB for non-system disks. Once it passes both 10% AND 2000 MB thresholds it turns to Warning level and alert.

    SO advisable to check what is best for your environment. You could put the MB value to a higher value so you reach it faster and so that it comes near the percentage threshold. Just first think what thresholds you would use for instance (in percent OR in MB) and caculate back what the other threshold would be that you want to set.


    Bob Cornelissen - BICTT (My BICTT Blog)
    22/جمادى الأولى/1432 12:49 م
  • and to add to what Rob and Bob already has said.

    By default (can be changed with an override), this monitor runs once an hour.

    That explains why you get the 0% and 301 MBytes (given that the disk has filled up the last hour).

    /Roger


    This posting is provided "AS IS" with no warranties, and confers no rights.
    22/جمادى الأولى/1432 01:44 م
  • Thanks to Rob and Bob for kind support. 

    And Thanks a lot dear,to explain these things:

    By default (can be changed with an override), this monitor runs once an hour.

    That explains why you get the 0% and 301 MBytes (given that the disk has filled up the last hour).

     

    Some thing I want to confirm that as Graham Davies has written in blog:

    Note that by default you don't get alerts for warning thresholds - just critical thresholds.

    then how to get warning alerts(If My org need is first warning and then critical alert on same server)

     

    And one more thing I want to confirm that :

    If the disk space size is 50GB.

    And If I set 5% (2.5GB) with 5GB and 10% (5GB) with 10GB in logical disk threshold.

    Then the critical alerts will generate at 2.5GB or 5GB?

    Kindly confirm.......

    Thanks 


    Syed
    22/جمادى الأولى/1432 02:37 م
  • Thanx Bob,

    one more thing I want to confirm that :

    If the disk space size is 50GB.

    And If I set 5% (2.5GB) with 5GB and 10% (5GB) with 10GB in logical disk threshold.

    Then the critical alerts will generate at 2.5GB or 5GB?

    Kindly confirm.......

    Thanks 


    Syed
    22/جمادى الأولى/1432 02:38 م
  • Thanx Rob,

    one more thing I want to confirm that :

    If the disk space size is 50GB.

    And If I set 5% (2.5GB) with 5GB and 10% (5GB) with 10GB in logical disk threshold.

    Then the critical alerts will generate at 2.5GB or 5GB?

    Kindly confirm.......

    Thanks 


    Syed
    22/جمادى الأولى/1432 02:39 م
  • Well, I think 50 GB amounts to 51200 MB.

    5% of that would be 2560 MB, 10% 5120 MB

    I would set the % tresholds to 6% and 11% resp. since the monitors do not change state until the treshold is passed (<6%, <10%).

    This is my assumption anyhow, if anyone has seen a clarification on how rounding (for %) works and whether the thresholds can take fractions, please elaborate.

    Regards

    Roger

     


    This posting is provided "AS IS" with no warranties, and confers no rights.
    22/جمادى الأولى/1432 03:43 م
  • Hi, actually it depends on what your environment is normally like.

    If all data disks are about the same in size, or that you always want to be notified at around 5GB of free space as a warning you could even set the percentage to a higher level than the MB threshold. That way you know that the alert is just waiting for the MB threshold to pass and you will have it.

    In your example. All non-system disks being 50 GB. If you want to be warning at 5 Gb and critical at 2,5 GB you can use 10% and 5% and 5120 and 2560 MB. This will cause you to pass both percent and MB threshold at the same time. Thats OK. Or you could set one of them higher so they are for sure over threshold (so percentages at 12 and 7% for instance or much higher). In that case you are basically monitoring for the amount of MB left on the system as your threshold.

    Syed, in your example the warning levels are at 5Gb and critical at 2,5GB. You have to pass both yellow thresholds to get yellow and both red thresholds to get to a red state.

    The other question you ask one of us is about the timing of the monitor. It does run once every hour. Which is generally enough for disks. However in your current case with the low threshold of only 1 GB for the data disk and if you have big chunks of data you are likely to go from green state to red state (or full disk) in a very short time. Possibly shorter than that one hour. We assume you already size your disks in a way that they dont fill up in minutes, but you should plan your alerts (thresholds) accordingly. SO for the big disks (50 and 100 Gb) it is normal to set the MB thresholds to 10 and 5 GB for instance. Just an example.

    About the alerting: This monitor does change state at yellow and red. But it alerts as alert only at critical state.  You can also set an override for that on the same monitor (alert on state - set it to the one that lists both warning and error).


    Bob Cornelissen - BICTT (My BICTT Blog)
    • تم الاقتراح كإجابة بواسطة Bob CornelissenMVP, Moderator 23/جمادى الأولى/1432 05:51 ص
    • تم وضع علامة كإجابة بواسطة Syed-DC 23/جمادى الأولى/1432 09:21 ص
    22/جمادى الأولى/1432 05:35 م
  • This monitor doesn't generate an alert on Warning, by default - only critical.
    HTH, Jonathan Almquist - MSFT
    23/جمادى الأولى/1432 02:41 ص
  • Thanx a lot dear,

    now i can set this monitor for both as per my Org. needs.

     

    Thank you once again.

     


    Syed
    23/جمادى الأولى/1432 09:25 ص