System monitoring is an important part of daily administrative tasks. Following our recent webinars, showing the ways we can monitor our Open-E DSS V7, we decided to prepare a how-to instruction for you.

This article is outdated as the functionality has been improved. In order to get the updated information, go to the article: Monitoring Open-E DSS V7 and Open-E JovianDSS with Check_MK

We will show you how to configure the necessary tools, update your Open-E DSS V7 and finally, access and use the monitoring interface.

What is Nagios, OMD and check_mk?

Before we jump into a configuration process, you may need a quick introduction to the tools we will mention in this post.

Nagios is one of the most popular open-source monitoring software applications. It offers monitoring and alerting for servers, switches, applications and services.

OMD (Open Monitoring Distribution) bundles Nagios together with many important add-ons and can easily be installed on every major Linux distribution. Using it, we can avoid compiling and integrating Nagios add-ons manually.

Check_mk is a status GUI written in Python which supports user definable views and is able to display the status of several sites in one combined view. A very intuitive design and an easy operation are one of the strongest points of this tool.

The following instruction was made in Ubuntu. However, a similar configuration can be made in other Linux distributions.

Installing and configuring OMD

We start with the installation and configuration of OMD packages. Please follow the steps below:

    1. From root level*, update and upgrade your distribution’s package index using the following commands: 
      apt-get update
      apt-get upgrade

      In addition you can upgrade your distribution’s kernel:

      apt-get dist-upgrade

      *You may need to use “sudo su –” command, as Ubuntu do not login root by default.

    2. In order to install the OMD package, go to OMD website and navigate to Downloads section.
    3. Go to package repositories and choose the relevant version of the repository (in our case it will be “Ubuntu Precise 12.04”).
    4. Install the relevant GPG key in Ubuntu.
gpg --keyserver keys.gnupg.net --recv-keys F8C1CA08A57B9ED7
gpg --armor --export F8C1CA08A57B9ED7 | apt-key add -
    1. Next, we enable the stable release repository (in our case, it is the one dedicated to Ubuntu Precise 12.04):
echo 'deb http://labs.consol.de/OMD/ubuntu precise main' 
>> /etc/apt/sources.list
  1. Following the installation, we run apt-get update to refresh our distribution repository: 
    apt-get update
  2. We then install the OMD:
  3. apt-get install omd-0.56

Following the installation of OMD repository, we go to our Open-E DSS V7 server to perform an additional configuration and install a small update.

Applying the small update to Open-E DSS V7

Click here, to find out how to obtain small updates. Please note, it is always best to confirm with our technical support, before installing any updates to your system.

To install the necessary update, we need to log in to our Open-E DSS V7 GUI and do the following:

    1. From the Maintenance tab, select System update and navigate to System software update.
    2. Click Choose File to pick the small update upd_70139-DSS-V7.upd, then click on upload and apply.system software updateYou will then need to manually restart the system.
    3. Go to Maintenance tab, Shutdown, find System Restart and click on Restart button in order to reboot the server.

After the installation, the small update will be visible in the menu (it can be removed by clicking on the trash bin).

small update

Enabling API in Open-E DSS V7

  1. Go to Setup tab and Administrator settings, then navigate to CLI/API Configuration.
  2. Tick the box to Use CLI/API, then specify port – 22223 and password.
    cli configuration
  3. In order to use the CLI/API functionality without password, you need to generate ssh key. You can do it by expanding show advanced menu and clicking on the generate and download button.cli configuration
  4. Save the file in the relevant location (in this example, we will call it “dss60.key”).
    To be able to use the ssh key in our Linux distribution, we may create a NAS share and then mount it in Ubuntu.

    Go to manual, to find out how to create a NAS volume and a NAS share in Open-E DSS V7.
  5. After creating a NAS share, enable SMB with “Guest” access and NFS access (with default settings) for that share, so you will be able to use it under Ubuntu.
  6. Connect to the created share (via IP address of your storage server e.g. Run -> \\192.168.232.60) and copy the ssh key to the “Test” share.
  7. In Ubuntu, create directory /mnt/test: 
    mkdir /mnt/test
  8. Mount our Open-E DSS V7 share to that location: 
    mount –t nfs 192.168.232.60:/test /mnt/test
  9. Now, you can copy the file to the OMD folder (/opt/omd) – created during installation. You can use midnight commander to copy the file.
  10. The next step is the creation of a new site (ours will be called “dssmonitor”) 
    omd create dssmonitor

    What you get is:

    • • a site directory with preconfigured configuration files
    • • a new user “dssmonitor” and a new group “dssmonitor” (identical with the name of your site). The new user is also a member of the group omd, which is created during installation.
  11. Then, we start the omd on our newly created site: 
    omd start

    We are now ready to perform an additional configuration of OMD.

  12. We edit the file “/opt/omd/sites/dssmonitor/etc/check_mk/main.mk” (we can use mcedit command for that purpose): 
    mcedit /opt/omd/sites/dssmonitor/etc/check_mk/main.mk
  13. In the relevant space, we insert the host’s IP and type of the protocol used: 
    all_hosts = [‘192.168.232.60|ssh’]
  14. Next, we append the following entries, to define commands that will be used to connect to API (see convention below): 
    datasource_programs = [ (“ssh –p default port number – i  
ssh key's location path –l username for API  + host IP + command 
that will be run over API to DSS, [‘protocol used’] , 
in which section our monitoring statuses will be shown), ]

    In our case, it will be:

    datasource_programs = [ (“ssh –p 22223 –i /omd/dss60.key –l 
api 192.168.232.60 check_mk_agent”, [‘ssh’], ALL_HOSTS ), ]
  15. Save changes to file.
  16. The important part is to ensure the correct ownership (our OMD user) and access permission (read and execute for owner only) of our ssh key.
  17. To change the owner of our ssh key file, we use the following command: 
    chown dssmonitor /omd/dss60.key
  18. To change access permission, so only the owner has read and execute rights, we use: 
    chmod 500 /cmd/dss.key
  19. Then, we log to OMD as dssmonitor using “su” command: 
    su dssmonitor
  20. To test the configuration, we can run “api” command: 
    ssh –p 22223 –i /omd/dss60.key –l api 192.168.232.60 check_mk_agent

    When prompted to continue connecting type yes.
    The output will be returned in the terminal, which will later be interpreted in check_mk tool.

  21. We need to run two more commands:
    First will do the following: retrieve data from our Open-E DSS V7, find all available services, and list them:

    check_mk -I

    The second, will reload the OMD configuration and validate that it is working properly:

    check_mk -O

Monitoring the Open-E DSS V7

  1. In order to check the status of our Open-E DSS V7 in OMD, go to your internet browser and navigate to the following address: your Ubuntu IP*/name of your site – in our case it will be: 
    192.168.232.58/dssmonitor

    *you can check the IP of your Ubuntu using the ifconfig command

  2. When prompted for a username and password – the default credentials are: 
    Username: omdadmin
    Password: omd
  3. We will then be able to choose one of the available GUIs. In our example we will go for “Check_MK Multisite”
  4. When in GUI, from available views, choose Hosts -> All hosts, and select the relevant host (see below).host view

You will then see the statuses of all available services (that are being monitored by the tool):

monitor view

You will also have access to detailed statistics (including graphs) about each service:

detailed view

The services that you can monitor with the tool include:

  • CPU load
  • CPU utilization
  • Disk IO Summary
  • NI information
  • Memory utilization
  • TCP connections
  • Uptime
  • NAS Volume space taken
  • and many more…

Check out the videos below and watch our engineers performing the same configuration (English and German version).

The names of actual companies and products mentioned herein may be the trademarks of their respective owners.

Dominik Niewiadomy

15 Comments

  • Denny /

    20, 03 2013 01:03:59

    hi,

    I have already a fully install Icinga server and installed the small updated (check_mk_agent) plugin for DSS7, but the first test fails:

    icinga@icinga:~/.ssh$ ssh -p 22223 -i id_dsa_iscihead-s -l api 192.168.1.100 check_mk_agent
    CLI/API: method not found

    So, what could be the problem?

    cu denny

    Reply

    • Denny /

      20, 03 2013 01:36:34

      hi,

      fixed it by myself. I missed the reboot 🙂

      Reply

    • Kasia Kolodziej /

      20, 03 2013 01:50:05

      Hey Denny! Glad you found a solution! A lot of people forget to reboot the server after applying the small update.

      Reply

  • rajmundo /

    10, 04 2013 08:34:19

    The link to SU is broken!!! Please correct it…

    regards!

    Reply

    • Dominik Niewiadomy /

      12, 04 2013 12:40:35

      Hi rajmundo! Many thanks for your info, the link is now updated. In case of any issues, please contact our technical support.

      Reply

  • martin /

    03, 07 2013 09:05:53

    Hi.

    Thank you so much for adding nagios support – finally!!!

    However there are two problems here, (installing small update and configuring nagios or icinga in this case is no problem), setup is a active-active cluster (of course both servers are identical), both problems on both servers.
    1) check_mk reports a critical state for drbd:
    “DRBD drbd8 status”
    “CRIT – Connection State: WFConnection, Roles: Secondary/Unknown, Disk States: UpToDate/Outdated”
    But in open-e dashboards all resources are “OK” (volumes, replication and so on, nor error logs).
    So How do I see which ope-e volume is DRDB8? and how can I fix it? why does check_mk report a problem that is not visible in open-e webfrontend?

    2) check_mk reports a “unknown” status for something:
    “RAID Unit 0 Optimal”
    “UNKNOWN – invalid output from agent, invalid check parameters or error in implementation of check dss_raid. Please set debug_log to a filename in main.mk for enabling exception logging.”
    Logfile:
    Invalid output from plugin or error in check:
    Check_MK Version: 1.2.2p2
    Date: 2013-03-07 10:49:37
    Host: 10.0.0.12
    Service: RAID Unit 0 Optimal
    Check type: dss_raid
    Item: ‘0 Optimal’
    Parameters: None
    Traceback (most recent call last):
    File “/var/lib/check_mk/precompiled/10.0.0.12”, line 702, in do_all_checks_on_host
    ^A^N^A^R^B^F^A^F^A^F^A^C^A^U^A1^B^R^A^Q^B^L^B^R^A^Q^B^L^B^P^A^R^A^F^A^Q^A^R^A^F^A^Q^B^F^A^O^A
    File “/var/lib/check_mk/precompiled/10.0.0.12”, line 1484, in check_dss_raid
    IndexError: list index out of range

    Agent info: [[‘LSI;0’, ‘0;Optimal’, ‘Drives#;2’],
    [‘0;Optimal’, ‘Drives#;2’],
    [‘0;Optimal’, ‘Drives#;15’]]

    So what is going on here? Which RAID unit is this? The checks of the Logical Volumes of the RAID Controller ae somewhere else…

    Thank you so far
    Regards
    Martin

    Reply

    • Kasia Kolodziej /

      04, 07 2013 08:17:55

      Hello Martin! Nagios lists all volumes that have drbd enabled. Any chances one of them is not configured in failover or not connected at the moment?
      Open-E DSS failover manager monitors only volumes that have been configured in your cluster.

      If you are still having issues, please open a support ticket with up-to-date logs so our team can investigate if everything is set up correctly.

      Reply

      • martin /

        04, 07 2013 12:45:36

        Hello Kasia.

        Yes, quite sure that all volumes are (or should be?) configured in failover setup.
        Who can I see the assignments DRBDopen-e volume names? With only the DRBD name, and no hint which open-e volume is meant, the information is quite useless…

        Any ideas for the second problem?

        Another question: I recognized that you are using a very, very old megacli ” Ver 1.01.09 May 25, 2006″. Is there any plan to upgrade to a current version?

        Thank you for your help
        Regards
        Martin

        Reply

        • Kasia Kolodziej /

          09, 07 2013 02:13:41

          Martin,
          In regards to you first issue, the check is an original check of check_mk and was not modified by our team. Name is drbd minor and can be found in logs in /drbd directory log files. We can improve this check to be more user friendly if requested in a support ticket.
          As for the second problem, please make sure you are using the same version of Small Update as check files for nagios. Link to the most recent one can be found in the post.
          Checks are still being developed and constantly improved.
          It is also possible that your RAID is returning some strange state that was not recognized during the implementation. Please open a support ticket with logs from your system to confirm.
          Finally, yes there is new MegaCli available. Please open a support ticket to receive this small update.

  • Matthias Raithel /

    13, 08 2013 10:44:57

    Ok.. am I right in thinking that there’s no way aound check_mk here if i want my open-E monitored by nagios?
    I’m also very interested in knowing about the low level state of things (raid controller)

    Any chances there?

    Reply

    • Kasia Kolodziej /

      20, 08 2013 07:26:50

      Hey Matthias. There is a plugin available for check_mk and for RAID support. Please open a support ticket to receive it.
      You can also use basic SNMP MIBs.

      Reply

Leave a Comment