Driving the Open Source CMDB(2005 total words in this text) (1565 Reads)  This article has only recently been added to the public area of this site. It has exisited for sometime as an article included in both the members only area of this site as well as the content area of the ECDB Tools VM Guest
The CDMB.info build details a configuration of tools that provide much of what you would expect from a mature product offering. The key is understanding the operability factors you want to measure and knowing what these tools are capable of reporting.
This is just a brief overview to get started. A review of the sections below should provide enough information to get the system up and running in your environment. Hopefully many of you are reading this to assist with setting up the VM Image of the ECDB Toolset. Some of the suggestions below relate specifically to the Guest configuration, so if its not making sense, download the Guest and review the functions of that configuration.
TIP: To open any of the applications in a new window, just do a right click on the menu link and select "Open in New Window" from the available options.
If you would like to make a contribution to the documentation available in the "Driving the Open Source CMDB", just forward an email to feedback@cmdb.info
This tool is one of the industry standards. It has been complemented over the years by a variety of third party front ends. One of the best is the Oreon front end configuration tool. Oreon itself has some additional features for graphing and reporting which you should also explore. The key use for Oreon is to write the configuration files for Nagios. There are a variety of ways to operate Nagios but this how we get started with this implementation of Nagios. Go to the Oreon section for the first steps in defining your Nagios configuration files.
Now that you have some hosts defined in the config files and your having a look around, the key information is more in the trending and availability reports rather than watching for alerts, but whatever suits your requirements its up to you. There is extensive information that can be monitored by the use of additional plugins and probes. The critical part of having a well developed system is in detailed definition of the services and service groups. When running a report for the availability of a specific application, it is crucial that all end to end elements be monitored within the single service grouping.
Nagios is a very worthy tool for monitoring of many conditions and it is strongly recommended that you have a detailed look at its capabilities. Link to the homepage below.
Setting up the config files for Nagios has never been easier. For anyone who has had experience of setting up the raw text conf files, they would surely agree. Log into Oreon and select the "Configuration" menu item. Select Hosts, and you will find some sample hosts already defined with services. If starting Oreon with a clean slate you need to define services as well as hosts before you can get a clean set of configuration files. Once you have edited the IP addresses in the examples and saved them, go to the Nagios tab, Select both Yes options for the Move export files and click the export button. You will see the nagios restart report below and it shouldn’t be stating errors with an IP address change. This has restarted Nagios with your host devices defined. Check out Nagios as it schedules the contact with your devices.
Always use templates, when defining new services and host definitions you will find a few example templates to get you started. If you commit the save of form data and it doesn’t save, it just appears back on the screen, it means you have not entered data correctly in a mandatory field. this has even been the case in some instance where it should have picked it up from the template. Anyway experiment a bit with defining service and host templates.
The graphing from the RRD Tool works for both Nagios and Oreon. Oreon has some other good features so have a look around the program
This tool is very effective for the asset discovery phase of setting up your data store of information. Exactly what do have under your control, what else maybe out there. This software needs no agents to discover devices on the network and report back to your database what they most likely are, and what they are doing. Much of this information is gleamed from the port stats which state what ports were found to be open. Most port mappings can be traced to an identified service. Not 100% accurate but close.
To guess to Operating system of the devices your scanning, the system needs to run as root, Apache Web Server really wont allow you to do this so go to the admin page and set up a CRON job which will kick off the job under Root privilege. Other admin function allows for refresh (flush) of the database.
Data presentation many exportable formats for easy distribution in your environment.
This is also a fantastic tool. The first action besides setting up some other login ID's is to deploy the agent to some machines in your environment. Basically the only information you really need to input during the agent install is the IP Address or Host Name (Please test the DNS resolution of your Host Name) of your host guest. ie: From your desktop environment : > ping cmdbtools02 to validate that you are resolving the host name.
You can find both the M$ and Linux agents on the SAMBA share of this guest. read the SAMBA section for instructions on connecting to the SAMBA share. Please copy the agent to a more suitable location in your environment. If you hit the SAMBA share with multiple clients wanting to load the agent you will most likely stress the VM Guest a bit. There are multiple ways for the agent to execute to the devices in your environment, so please do a bit of reading regarding the load of the agent and its daily execution. It can be included in login scripts if required.
Define which devices you would like to provide detailed asset reports and deploy agents to these devices to enable them to report back to OCS Inventory. The best approach for this is to run some scans from Nmap2Mysql as root under the CRON job scanning and make a determination on which devices you want to be reporting back to the OCS server.
Login into the OCS front end and have a look at the sample machines that were reporting back to the server in the CMDB.info test environment.
The OWL Doc Centre is great for sharing documented information with non-trusted users within your site. Although in many cases this information is confidential it helps where Business Group A does not trust Business Group B with free access to all of its data. Just put their systems design and other IT documentation up to the Doc Centre for collaborative sharing of information.
OWL Doc Centre is also capable of sending alerts when review and or update conditions are placed on documents, however the email functionality has not been factored into this guest but is a reasonably simple upgrade task if you want to enable these features.
For this VM Guest all documents stored in the OWL Doc Centre are stored as binary files in the MYSQL database, they can also be stored as native file types direct to disk if you choose to reconfigure the system. Please note that this guest has been built as a 4Gb image so there is some limitation on disk space.
Webmin is an industry standard admin interface for Linux and Unix systems. It is awesome. Take some time to check the variety of features, many of them are not configured as their supporting components have not been installed on this guest. The key areas you may wish to use would be, File Manager, the Command Line interface, and the Mysql Server interface. Although all of the required functionality has already been built into the VM Guest, we understand you may want to have a real play around with this server. Plenty of features available for experimenting.
- Post Nuke Content Management System
Post Nuke is software that creates an impressive, dynamic web site and provides the webmaster with a site they can administer with a minimal amount of HTML knowledge through a web browser.
But, Post Nuke’s functionality can be increased by installing modules, blocks and themes. For example, you can add a forum, a gallery and contact form. You can also change how your entire site looks by changing themes. All of this can be done with just a few clicks in the administration panel saving you hours of time, both in the initial creation of the website and in its day to day maintenance.
Post Nuke can do anything from traditional blog websites, to a community members' only website with hundreds of users. Post Nuke can scale to thousands of uses but is just as easy to create a small website for friends and family.
So as you can see it is ideal for the purpose of building a front end portal for a range of Open Source tools and bring them together under a single collaborative banner.
The Post Nuke Calendar is one of the handiest add ons for Post Nuke that I have found., but there were a few versions that didn’t work or didn’t work properly. Essentially this add on has been included for the sole purpose of publishing planned changes and system events in your environment.
Some default groupings have been established to reflect what are the prime groups within the average computing environment. Each group can raise an appointment to reflect the planned change and publishes that as a planned change that is yet to be authorised. Review groups then have an opportunity to reject the change based on the information supplied. Define your own process for authorisation. Most organisations will need to model some processes around this tool and possibly the doc centre can play a part in hosting supporting documentation for some changes.
- Post Nuke BB Chat / Tech Talk
this component has been included for several reasons but the primary one is that typically, all organisations suffer internal communication issues. Open forums are a great way to foster a collaborative approach and retain visibility of historical information.
Another great benefit is that it is fully integrated with the Post Nuke authentication and operates fully within the architecture of the content management system.
Just login with the admin account, go to the forum index and follow the link at the top of the page to the BB Chat admin page. You will soon see how to create forums, perform key actions and manage the overall system.
There is way too much to be said about the MYSQL DB and its setup for this system. It is recommended that anyone wanting to know more about MYSQL should reference the MYSQL User Guide.pdf from the MYSQL Web site. For simple administration tasks try the MYSQL interface under WEBMIN.
To connect to your server's SAMBA SHARE just connect using your M$ "Map Network Drive" \\your ip address\dbuser\code and give the Password and User ID as stated in the PASSWORD Page. You will be presented with the explorer shell view of the SAMBA SHARE. If your operating directly on the Linux Guest under a command line, to get to the same file structure go to /home/dbuser/code
SAMBA is simplistic yet can also be very complex, to learn more about SAMBA simply reference it on a web search engine and you will find a great deal of data available.
Apologies for the delay in getting this info up on the site. I'll be updating this section again around July 2008 |
