Hi Lucky,
1. Is IRE and Identification and Reconciliation rules different? I am still confused who will create a new CI record?
No they are not different, IRE stands for Identification and Reconciliation Engine. These rules are part of the IRE. IRE will do both create and update operations based on whether or not the CI in the payload matches an existing CI in the CMDB. IRE will create and/or updated even if there are no reconciliation rules. These are optional and are case by case basis, depending on the user's configurations.
2. If no match is found from the identifier, then how come the one who creates a new CI record will come to know that the record has to be created in this class only?
If no match is found in the CMDB CI and all the criterion attributes needed to create a CI are present in the payload, IRE will create a new CI record in the class that was specified in the payload.
Here is an example payload:
{
"items": [
{
"className": "cmdb_ci_linux_server",
"lookup": [],
"values": {
"short_description": "Community question sample",
"name": "Sample Linux CI"
}
}
],
"relations": []
}
In this example, we are passing in a Linux Server CI. OOB, Linux Servers can be identified based on below 5 identifiers which can be viewed in the CI Class Manager. This can look different based on what plugins you have installed.
As you can see, in the payload we passed 'name' and there is a matching identifier entry which says we can uniquely identify this Linux Server based on name only with priority 300.
Behind the scenes, discovery or SGC will send a payload to IRE and should send criterion attributes of at least one of the rules in that payload. That's how IRE knows what class to query and when to create/update.
Let me know if you have additional questions about IRE beyond this!
