Addmi 11-intro to-patterns
- 1. Introduction to Patterns How the Intelligence Works
- 2. Outline Pattern Basics What is a pattern? Where do I get new patterns? Pattern Controls Demystifying Patterns Pattern components Patterns and the Model Patterns and Provenance Simple Pattern Configuration
- 4. What Is a Pattern? A way to customise Atrium Discovery so it can infer things in the datastore based upon data collected Patterns are event driven Written in The Pattern Language (TPL) Common software types can recognised by built-in TKU (updated monthly) Some organisations need their own custom written patterns Extend discovery information collected Uncommon or custom software components Business Applications which are specific to the organization
- 5. What Can Patterns Do? Triggered during the discovery process A certain OS version is found A certain process is running Can take the initial information collected and use it to collect new data Pattern matching details within the process arguments May go back to the host and run other commands Collect configuration files Collect inventory information from databases Use very specific commands to discover version Pattern management under the discovery tab Discovery -> Pattern Management
- 6. Where Do I Get New Patterns? The Knowledge Update (TKU) service Regular releases of a TKU A TKU can contain New Patterns Updates to existing Patterns Updates to End of Life Data Updates to Hardware Reference Data Updated every month, with new/updated patterns TKU-CORE-2009-06-1.zip Core patterns for detecting Software Instances TKU-DBDETAILS-2009-06-1.zip Deep database discovery TKU-SUPPORTDETAILS-2009-06-1.zip Last dates for software support
- 7. Upload and Activate Patterns Upload a single pattern TPL files (.tpl) Forms a single Pattern Package Upload batches of patterns In zip files The zip file is the Pattern Package
- 8. Controlling Patterns Patterns are grouped into packages, which are grouped into modules Activate or deactivate patterns to change the behavior of discovery
- 9. Deactivate and Delete If a package or pattern is deactivated, it does not execute during Discovery Can reactivate to include in the next discovery run Deleting a pattern or package of patterns: all data which was inferred based upon patterns in that package is also deleted
- 11. Example Pattern Each pattern must have a trigger If matched successfully this become the Primary Inference for the SI or BAI Triggers are matched at the point of discovery Triggered on DDD node Finds a host node in the datastore Infers a Software Instance node
- 12. Triggers Every pattern has a trigger Contained within the “triggers” declaration When a pattern “fires” the trigger declaration has been met Example: the host has a certain process running Example: An SI has been created or modified
- 13. Trigger Example
- 14. Data Model Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Host Host Host Software Instance Software Instance Software Instance Software Instance Software Instance Business Application Business Application Pattern
- 15. Data Model – Provenance Provenance is the source of all inferred information and is automatically tracked Stored with Inference relationships Referred to by the role of the evidence node The three core roles: Primary Contributor Associate Superseded relationships – marked destroyed
- 16. Data Model – Maintainer and Request Maintainer and Request inference relationships are a specific type of provenance Every node created or update by a pattern is linked back to the pattern that is responsible for it’s maintenance: Every pattern based discovery request, successful or not, is linked back to the pattern that requested it Business Application Pattern Maintainer Discovery Result Pattern Request
- 17. Data Model – Current Live Provenance Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Host Host Host Software Instance Software Instance Software Instance Software Instance Software Instance Business Application Business Application Pattern
- 18. Data Model – Old Superseded Provenance Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Host Host Host Software Instance Software Instance Software Instance Software Instance Software Instance Business Application Business Application Pattern
- 19. Data Model – Maintainer Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Discovery Result Discovery Run Discovery Access Discovery Result Discovery Result Discovery Result Discovery Access Discovery Result Discovery Result Host Host Host Software Instance Software Instance Software Instance Software Instance Software Instance Business Application Business Application Pattern
- 21. Host Provenance Example Use the “Show Provenance” button to show Provenance is defined as: The records or documents authenticating such an object or the history of its ownership.
- 22. SI Version Provenance How we know the full version of java? Click through the provenance link to the DiscoveredCommandResult node
- 23. SI Version Provenance Results Answer: The java version was determined as the result of running the java command with the “-fullversion” flag
- 26. Why Have Configurations? Allows changes to behaviour using the UI Without pattern editing, or reactivation Typical uses Turn pattern options on/off Update paths to find files or commands Any other parameters
- 27. Where Does it Appear? (1) If configuration blocks are defined in the pattern
- 28. Where Does It Appear? (2) On the “All pattern module configurations” page Discovery > Pattern Management > View Configuration of all Patterns
- 29. How Changes Are Applied User must have correct permission to edit: reasoning/pattern/config Otherwise can view readonly Changes take place immediately Be careful if discovery is running Warning shown if this is the case
- 30. Online Documentation: http://www.tideway.com/confluence/display/81/Pattern+Management http://www.tideway.com/confluence/display/Configipedia/Schedule+and+Roadmap Further Resources
Editor's Notes
- #8: There must be no structure in the ZIP file!