What is an IOR file, what does it do, and how does it work?

Question

I believe it is related to CORBA in some way (I'm not sure). I'm curious as to its function and how it works. Google isn't helping me when I search for "IOR file", and I'm not sure what else I could search for. Thanks to anyone who can at least point me in the right direction with available resources.

Epithalamium

Answer 1

An IOR file is a file which contains an Interoperable Object Reference which is a kind of a locator string. The IOR file itself contains the IOR. The IOR is an CDR encoded string which, depended on the CORBA version, contains various information regarding the servant who created this string. But basically it works as a locator string.

Inside the IOR normally an IP, portnumber and object reference of the servant could be found.

In a simple hello world example the servant (server) will create this file. The client reads this file and the client ORB (Object Request Broker) will delegate the call from the client to the servant transparently.

All about CORBA: OMG CORBA Website or just visit Wikipedia

Rearward

How does the client read this file? He doesn't know the address of the server. And if he did know the address of the server already, he wouldn't need an IOR, no? – Leadership

Good question. I think because 1) The IOR file contains information in addition to server & port (see links and other answers) and 2) the IOR file's location is different to the server's location. A client needs to only know the location of the IOR file, and not the server itself. – Contort

Sharing the IOR of the custom CORBA service with clients is an alternative for using a CORBA Name Service to resolve static names to service references (i.e. IORs). In case you have (A) a common filesystem or (B) a database available for both the client and the server, and (C) the number of services to locate is limited (i.e. 1 or 2), it was considered easier to store the server IOR in the database or a static file on the shared filesystem. – Sarcophagus

Answer 2

IOR stands for Interoperable Object Reference and is related to Corba

You can check out "Corba in 5 minutes" here : http://www.pvv.ntnu.no/~ljosa/doc/encycmuclopedia/devenv/corba-index.html

There's a section explaining what is an IOR

Mellissamellitz

OK. That's pretty good for beginners, but I still have questions. Like "why do I need to point both the sender and the receiver of CORBA messages at the same IOR file instead of giving each an exact copy of the IOR file?" – Epithalamium

It's not true. Normally the servant (server) creates the IOR file (or its content) and the client reads it. The IOR in an encoded locator string which tells the client where to find the servant object. – Rearward

Answer 3

IOR is a CORBA or RMI-IIOP reference that uniquely identifies an object on a remote CORBA server.

IOR can be transmitted in binary over TCP/IP via GIOP-IIOP (encoding may be big-endian or little endian), or serialized into a string of hexadecimal digits (prefixed by string IOR:) to facilitate transport by non-CORBA mechanism such as HTTP, FTP, and email.

To locate a server object at run-time, the client application requires a reference to it. This reference is called an Interoperable Object Reference (IOR). An IOR is a text string encoded in a specific way, such that a client ORB can decode the IOR to locate the remote server object. It contains enough information to allow:

• A request to be directed to the correct server (host, port number)
• An object to be located or created (classname, instance data)

Turves