Is there a high level description/tutorial for utilizing an external application for pulling data from RapidInsight 3? I get asked periodically by customers/potential customers how they can get their fault management information programmaticaly instead of through the GUI.
What needs to be done on the RI3 side to make it happen and is there a default XML schema?
Thanks,
Karl
Along these lines, probably
Along these lines, probably the best example would be... what would it take to have the fault management information available as an RSS formatted feed. User/group segregation would still apply - E.g. - when authenticated as user XYZ, only devices/notifications tagged with XYZ would be returned in the XML.
If performance data is required in the interface that comes from SMARTS for example, what would the practical limit be in terms of the number of devices?
Karl
RapidInsight natively
RapidInsight natively provides functionality to serve data in RapidInsight repository as XML. The schema is the RapidInsight model that can be seen at grails-app/conf/*ModelConfiguration.xml files or through the repository browser (/RapidSuite/browser.gsp).
There are number of different approaches to provide data as XML over HTTP, and the best approach depends on your customers' objectives.
1. RapidInsight standard UI consumes XML, hence if you customer wants to build alternative UIs, the best approach is to look at the XML data served to the UI by the server. RapidInsight UI is designed to handle large data sets, hence this has inherent capabilities to break the data in chunks, etc. The search method provides a lot of flexibility in retrieving the relevant data. For example the following url would return first 100 RsEvent objects in RapidInsight repository as xml.
http://localhost:12222/RapidSuite/search?offset=0&sort=id&order=asc&max=...
sort: which object property should be used for sorting
max: how many objects should be returned
offset: starting object count
searchIn: which class to perform the search
query: the search string
2. RapidInsight uses Groovy scripting language which provides a powerful "builder" to construct XML easily, hence you are not restricted to a particular schema and construct any XML structure you'd like. The data can come from RapidInsight or any data source RapidInsight has access (databases, files, etc.)
For example, to provide an RSS feed, you can retrieve the event objects in a groovy script and build the RSS XML using the groovy XML builder. Segmentation of the data would be handled by RapidInsight in the background automatically, so users can only access the objects they are allowed to.
I need more information on the "practical limits" question. The techniques I mentioned above make it possible to work with large data sets by dividing it in chunks. What do you mean by performance data?