Cherwell Monitoring in Production
I have been working on a couple of monitoring ideas for Cherwell. I didn’t see anything with a quick online search, and I enjoy authoring MPs to monitor applications, it is the closest I’ll get to 007.
I’ve hit a major hurdle and I need to ask for a hand from the community. We have a lab environment that’s worked great while developing the Cherwell integration for Connection Center, however, it is not a good simulation for an actual deployment. While many of the components are in play, the usage patterns and requirements are far from a day-to-day used system.
Below are details on what has been drafted out so far, if you have any feedback or would like a copy of the MP please let me know at nathan.foreman@cookdown.com.
Barebones MP
To get off the ground I’ve started with a DA that is largely centred around our single server instance, and should get a new top-level object. My first thoughts were to base a DA off the URL used to access the platform, however, this is one of many items I’d like to hear how the Cherwell community views things.
In the skeleton today, properties are pulled of the Cherwell application server and a discovery fills in further information on connections defined for that server. For performance the initial discovery is registry based and targeted at the Windows OS, another item that I need help to confirm is applicable.
Growing further
IIS and SQL play a large role in the Cherwell ITSM platform, and both have Microsoft authored MPs that we can leverage. I have been drafting out models that use the base IIS and SQL classes to bring this data into the rudimentary DA above. Deployments that are in a server farm would benefit even more from bringing this all together.
If you have seen common pain points when using Cherwell that you think monitoring will help with, I would greatly appreciate your feedback and suggestions. Please send me an e-mail at nathan.foreman@cookdown.com or reach out on twitter @Nate_Foreman.