Vmware-Monitor for Xymon (VMX) (was: Xymon monitoring of ESX servers)

Thomas Eckert thomas.eckert at IT-Eckert.de
Sun Dec 27 15:10:12 CET 2009


Hi Bruce, list,

I'm in the last stages of building a vmware monitoring extension for xymon
that does not require any changes to the ESX-hosts monitored. It uses the vMA
(more precisely the vCLI installed on the vMA) to fetch the information from
the vmware hosts (see below for a list of features).

My apologies to the list for announcing both a not-yet-public-available
and a "light-commercial" extension here.

This thing is named "Vmware-Monitor for Xymon (VMX)" and requires the
following environment:
- vMA with hobbit- or xymon-client installed (hobbit-4.2.0 w/o all-in-one
   patch works fine)
- xymon-server with working user-channel and a working >=gawk-3.1.3
- in particular _no_ vCenter is needed (as a nice side-effect monitoring of
   vmware-server-2.0.x-hosts is possible too)
The names of the VMs should be identical to their hostnames.

Currently the following features are implemented (short version):
- central configuration of VMware-monitoring on the XYmon server 
- current CPU- and memory-usage, both of the ensemble of all hosts and for
   each individual VMware host and running VM (and graphical trending of these
   and more of course)
- per VM status-page with info about VMDKs, vCPUs, memory, ... and
   vmware-tools (along with additional infos if tools are running)
- overview and tracking of virtual machines on each VMware host, no matter
   if the VM was moved by vmotion, DRS, manually added/removed, ...
   So it's possible to easily determine when a VM was started or stopped on a
   particular host!
- warning if a host is in maintainance-mode or has the "reboot-needed" flag
   set
- vMEM/pMEM-ratio, i.e. how much of the physically available RAM is assigned
   to the running VMs
- one-click overview/summary for the following metrics:
 	o number of monitored hosts (along with software version and patch-level)
 	o number of running VMs
 	o total CPU (MHz) available
 	o CPU (MHz) used
 	o total vCPU-count running
 	o total physical RAM available
 	o virtual RAM used (both "hostmemoryusage" and "guestmemoryusage")
- ressource friendly: low monitoring overhead due to non-forking design on
   the server-side and minimized disk-access to avoid I/O problems
- all status-pages are hyperlinked to each other so navigation from the
   overview-page down to a particular VM is quick and easy

The roadmap for possible additional features currently contains (no particular
order):
- disk-IO
- network-IO
- duplicate VM detection
- configuration of alarm-triggers (e.g. for high cpu-/mem-usage)

Currently a ESX-cluster with 7 ESX 3.5 hosts is monitored without problems.
ESX 4 / vSphere has been verified to work too as.


Documentation is not complete yet but installation instructions, explanation
of the status columns generated and graphing-details and -options are already
written.

This extension will be available at a very affordable price. Of course a
portion of the revenues will go to Henrik for providing XYmon at all -- the
"best monitoring solution"(tm).

If you are interested (either in getting an e-mail with the official
announcement, in beta-testing or if you have questions) please let me know
off-list.


A Happy New Year to you all,

 	Thomas


On Mon, 21 Dec 2009, White, Bruce wrote:

> Hi all,
>
>
>
> I saw a couple of threads on this topic from January and was wondering
> if there is any newer insight into this question.   At that time, it
> appeared that many were just installing a RedHat based version of the
> Xymon (hobbit) client (with modifications to  the df [vdf] and top
> [esxtop] commands) on their ESX servers despite the warning that it may
> void your support.   I have started down the SNMP monitoring path and
> have found that most of the fields listed in VM's MIBs are not
> populated.  So, are people still running modified versions of the hobbit
> client without issue?   Has anyone made any real progress on monitoring
> ESX servers via SNMP?
>
>
>
> Thanks in advance,
>
> Bruce
>
>
>
>
>
>
>
>
>
> Bruce White
> Senior Enterprise Systems Engineer | Phone: 630-671-5169 | Fax: 630-893-1648 | bewhite at fellowes.com | http://www.fellowes.com/
>
>
>
> Disclaimer: The information contained in this message may be privileged and confidential and protected from disclosure. If the reader of this message is not the intended recipient or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by replying to the message and deleting it from your computer. Thank you. Fellowes, Inc.
>
>

-- ----------------------------------------------------------------------
| IT-Beratung Eckert       | Hartkirchweg 54        | tel: 0761/ 594 9898
| Thomas Eckert            | 79111 Freiburg i.Br.   | fax: 0761/ 594 9899
| http://www.it-eckert.de/



More information about the Xymon mailing list