Setting up MySQL Enterprise Monitor version 3.0 with your private clouds is easy.
There are two functionalities that you should know: phone home and process discovery adviser.
If you have a fresh install of the MySQL Service Manager, you will want to go to shell and run an unattended install of the agent. You can write a script that would specify the unattended installation mode, as well as all the options that are related to your instance, local host and port, and other options related to the Service Manager and the default group where the new instance is to be added in MySQL Enterprise Monitor.
The good news is that if you are using Oracle VMServer, Open Stack or any other private cloud, you can just incorporate the agent in the base virtual machine easily. When you check back on Service Manager, you will see that the instance is already being monitored. If your specified production group did not previously exist, that group will be automatically created for you.
You can also check how many whole machines are being monitored, the number of MySQL instance that are being monitored and how many instances are not being monitored.
Those instances that are not monitored are automatically identified, which leads us to the process discovery adviser in the Service Manager.
Process Discovery Adviser
You can find the advisers in the monitoring and support services group in your Service Manager. Edit the global configuration for the adviser and choose default authentication information. After this, you will find that the Service Manager will automatically begin monitoring all those previously known unmonitored instances. This means that you get to save time from having to manually configure everything.
So using these two features (phone home and process discovery adviser), you can easily incorporate MySQL Enterprise Monitor into your private cloud or DevOps environments.
Here’s the video demo:
Welcome! In this video I’ll demonstrate how easy it is to incorporate MEM 3.0 into your private clouds in your automated DevOp environments using two distinct features. The first is the Phone Home functionality of the agent and the second is the Process Discovery Adviser that’s built into the Service Manager.
So let’s first look at the Phone Home feature of the agent. Here I have a fresh install of the service manager. The only thing that I’m currently monitoring is the bundled instance of MySQL at the Service Manager uses for its own repository. I’ll now run a simple shell script that first installs three test instances of the MySQLd. Followed by an unattended installation of the agent. And as you can see, I’m specifying the unattended installation mode. All of the options related to the instance of MySQL will automatically begin monitoring, which it would often simply be local host and port 3306, followed by all the options related to the Service Manager. And then finally I’m specifying the default group that I want this instance to be added to in MEM.
In the group is a key aspect here as this instance will then inherit all the policies configuration that I’ve set for that particular group. So if you’re using a private cloud like Oracle VM Server, OpenStack, Cloud Stack, Eucalyptus and so on, you can easily incorporate the agent into the base VM image used within your private cloud or you can add the installer to your post initialization scripts. So if you’re using DevOp’s automation tools like Chef Puppet in CF engine, then you can of course simply add the installer to your setup scripts. So I’ll go ahead and run that and I skip ahead in time, but as you can see it only took about 30 seconds on this machine.
And now when I come back to the service manager, I can see that the production group was created because it didn’t previously exist. And I can see that this instance is now being monitored. I can see that I have one host or machine being monitored, two instances of MySQL being monitored and two unmonitored MySQL instances. And those unmonitored MySQL instances were automatically detected by the agent. And that leads us to the second feature that I wanted to demonstrate – and that’s the Process Discovery Adviser in the Service Manager. So let’s go to the adviser’s tab, and we can find that adviser in the monitoring and support services group. We can then edit the global configuration up for that adviser and specify default authentication information.
And when we’re done with this, the service manager will then automatically attempt to contact to begin monitoring any known unmonitored instances. Let’s specify all the account info, which is identical for all of my test instances and that we’ll save the adviser configuration. Now when we come back to the instance management screen, I can see that my other two instances are now being monitored. So using the other of these two methods, I can have a set up so that any new machine is instantiated is automatically monitored by MEM without the need for any manual intervention.
So as you can see, it’s very easy to incorporate MEM 3.0 into your private cloud base setup or your automated DevOps environments.
And that’s all for this video. Thank you for watching and please be sure to look for our other video demos on MEM 3.0.