FWIW the questions asked in this thread remind me of this "slide" I created internally to help disambiguate our communication.
All too often someone uses the term "autotest" to mean one of the things below when it would be much easier to communicate if they were more specific and qualified it using one of the terms below. Autotest has mutliple meanings to different people: 1. The server: The Autotest server doing job scheduling and host management. 2. Database of test runs, their pass/fail and performance result data. 3. The client: set of Python APIs to use to write a single machine test, or profiler/wrapper/antagonist to run against any client test. 4. Server side tests: Multi machine tests using an API similar to “the client” but should not be confused with “the server.” 5. autoserv: A command line interface to launch a test on a machine directly from a users workstation without using the server or database. 6. AFE: A web GUI frontend used to launch jobs containing one or more tests on hosts. 7. atest: A command line interface to the AFE. 8. TKO: A web GUI frontend used to display test results from the Database. 9. The open source project. There are more that could be listed, I didn't mention the json rpc interfaces that the AFE and TKO web server sides provide that both the GUIs and the CLI use for instance but the only people who tend to know about those are a few developers. -gps
_______________________________________________ Autotest mailing list [email protected] http://test.kernel.org/cgi-bin/mailman/listinfo/autotest
