CMDB best practice

[PaulSylo]

Hi Experts,

I have a couple of questions in terms of Best practices for CMDB data.

1. We are using the network and server data for RU billing purposes every month. One challenge we are facing here is how to validate to RU. As you see, the auto-discovery is in place and the agreement is to find all the devices that are active and scanned by discovery in the last 30 days and validate and bill them. But, I see there is a lot of mismatch in the data when it is validated by the client and data available in CMDB. fixes are done reactively by the respective team which results in temporary fixes and again data is mismatched when it is validated by clients. 

After my investigation, in a lot of differences in scheduled jobs, and no best practices are followed and running in a siloed manner, which loses the trust in the data , which results in multiple references of the same data, which defeats the purpose of CMDB. How to fix these data and how to make these validate same across both parties. Do I need to bring in some dates in the agreement, that the client needs to check the data in between dates ( since CMDb data will be changing)? 

2. How to handle these CMDB data mismatches as best practices and avoid the ongoing issues

Kindly share your thoughts and expertise pls.

Regards,

Paul

[Deepak Shaerma]

Hi @PaulSylo 
Addressing CMDB (Configuration Management Database) data accuracy issues, especially for billing based on resource usage (RU), requires a multi-faceted approach focusing on data quality, process standardization, and continuous improvement. Here’s how you can address these challenges:

- Establish CMDB Governance: Create a governance body or committee that includes representatives from all stakeholders, including those from client-side. This body will define, enforce, and regularly review CMDB standards, policies, and procedures.
- Define and Document Best Practices: Clearly document all processes related to CMDB management, including data entry, modification, validation, and deletion. Ensure these practices align with ITIL guidelines for CMDB management.
- Scheduled Jobs and Discovery Tools Configuration: Standardize the configuration of scheduled jobs and auto-discovery tools. Regularly review their performance and accuracy to ensure they are capturing the required data correctly.
Automated Validation Tools: Implement tools that can automate the process of validating CMDB data against actual resources in the environment. This includes ensuring that only active devices are billed.
- Regular Reconciliation: Establish a routine reconciliation process where CMDB data is compared with independent data sources (like network scans, asset lists from other systems, etc.) to identify and correct mismatches.
- Data Quality Metrics: Define key metrics to measure the quality of CMDB data. These could include accuracy, completeness, timeliness, and consistency. Regularly review these metrics to identify areas for improvement.
Change Management Integration: Integrate CMDB management with your IT service management (ITSM) change management processes. This ensures that any change in the IT environment is reflected in the CMDB promptly and accurately.
- Comprehensive Documentation: Ensure comprehensive documentation of the CMDB, including the data model, relationships among CI (Configuration Item), and the rationale behind the organization of this information. This will aid in resolving discrepancies and improving understanding among stakeholders.
Note: Please Mark this Helpful and Accepted Solution. If this Helps you to understand. This will help both the community and me..
- Keep Learning ‌‌
Thanks & Regards 
Deepak Sharma

[Ashok Sasidhara]

I have explained some of the best practices to consider for CMDB data management in the following article:

https://medium.com/@ashoksasidharan/best-practices-for-cmdb-data-management-in-servicenow-9cb6ae18eb...

Analyze your current data, process, configuration etc. for each area mentioned in the article. Based on the gaps identified from your analysis, list down action items to improve the configuration management process and CMDB data quality. Start by reviewing the 'CMDB and CSDM data foundations dashboard' and 'CMDB health dashboard' so that you can identify some quick wins which can be delivered in a short timeframe.

[Dr Atul G- LNG]

Hi @PaulSylo 

Connect with @AJ-TechTrek 

[Sohail Khilji]

Hi @PaulSylo ,

Hope your doing well !

Managing CMDB is pain ! I can help you with some inputs...

Lets decode your questions:

1. We are using the network and server data for RU billing purposes every month. One challenge we are facing here is how to validate to RU. As you see, the auto-discovery is in place and the agreement is to find all the devices that are active and scanned by discovery in the last 30 days and validate and bill them. But, I see there is a lot of mismatch in the data when it is validated by the client and data available in CMDB. fixes are done reactively by the respective team which results in temporary fixes and again data is mismatched when it is validated by clients. 

Sohail : Looking at your scenario its sound very natural to me. CMDB does require daily operation activites to maintain it. You will have to work a lot on CMDB IRE RULES & SOME AUTOMATIONS for you to streamline this daily actions. When multiple sources update cmdb it becomes very hard to manage which source updates which table and which field resulting in confussion which your facing. To maintain the data for validation you can reduce the frequency of data update on cmdb. so when your client checks the data it may be the data data which you validated. (Make it weekly update if possible). Try to brainstrom on which data is being updated frequenty and action on it my automating the cmdb and fixing it via IRE rules as much as possible. By doing so, Over the period of time CMDB becomes much indipendent. 

Wanted to know which of your data is being updated ? and how your expecting it to be...

After my investigation, in a lot of differences in scheduled jobs, and no best practices are followed and running in a siloed manner, which loses the trust in the data , which results in multiple references of the same data, which defeats the purpose of CMDB. How to fix these data and how to make these validate same across both parties. Do I need to bring in some dates in the agreement, that the client needs to check the data in between dates ( since CMDB data will be changing)? 

Sohail : This is hard job to do, go for crawl, walk, jump,leap approch. List down the top problamatic cmdb issues and try to include them in your agile stories to fix the known issues slow. For client statisfaction and to maintain the version of data which you validated and the client is going to validate i recommend you to reduce the frequency of data update so that you both are on same page.

2. How to handle these CMDB data mismatches as best practices and avoid the ongoing issues:

a. Focus to build reliable IRE rules for your cmdb (they are backbone).

b. Identify active problems try to remidate with workflow/flow and other remidation actions.

c. find the root cause and bring fixes in place.

d. Build lot of reports...

I hope my inputs helps...