Lucas, On Tue, Mar 2, 2010 at 12:01 AM, K.D. Lucas <[email protected]> wrote: > You control individual test cases by the statements in the control file > and/or python script for that particular test. You can also use the front > end to start/stop and monitor the test case progress (it will grab various > logs for you to look at during and after the test runs). The amount of > control you have over each individual test case will depend on how you write > and manage them. Whatever your top level control file is, you will have > control to manage that process. > Pass and Fail is indicated on the dashboard if you are using the front end > (Green color signifies pass), or from the command line it will log the > results into the run logs. All of your stderr/stdout will be logged, as well > as any messages you want to pass to the logs (there are various functions > you can use to log messages). > There may be some better examples of looking at the results then > test.kernel.org, as that seems to be limited to a subset of tests. There are > also graphing capabilities in AutoTest that work well with benchmarks, etc. > If you have a spare Linux box at home, you may just want to install it and > play around with it.
I would try to use it first. Thanks a lot for your help. > kdl > > On Mon, Mar 1, 2010 at 6:00 AM, shiraz hashim > <[email protected]> wrote: >> >> Hello Lucas >> >> On 3/1/2010 12:13 AM, K.D. Lucas wrote: >> > In AutoTest a control file can execute either an individual test case or >> > a suite of test cases. You can also create a control file that will >> > execute multiple test cases. The control file will usually point to a >> > python script that will execute your test case. Each test case is >> > usually separated into it's own directory with it's own control file. >> > It's very flexible as you can define dependencies. >> >> OK. And how will I control individual test case. So lets say a control >> file executes many test cases, will I be able to select/deselect, pause >> or abort individual test case during execution. >> Secondly, how is the pass/fail of an individual test case is communicated >> to serve and how is the log generated by test cases maintained by the >> server. >> >> > I was looking at STAF long ago (it may have changed in the last 4 years) >> > and at that time I concluded it doesn't have near the power or >> > flexibility of AutoTest. >> >> You may be right. But now at first look it seems that STAF/STAX and >> autotest >> have overlapping goals. STAF is basically communication framework with >> number of >> possible services. STAX is one of these service to execute and >> maintain processes >> on local or destination machines (using STAF framework). STAX provides an >> XML/python based interface to do this. It also provides a gui job monitor, >> has structured logging and message mechanism. >> >> > Are you just looking for a way to execute your tests, or do you want >> > additional system statistics, profiling, and reporting? >> >> The basic requirement is testing but any thing above it, that is supported >> by the framework would be great. >> I think related to this, autotest has some good features ( and therfore, I >> am confused ;) ) >> I did go through test.kernel.org to see report generated by autotest. >> I could find the results of the tests for different kernel version. >> But I couldnot search for details (in readable form) on individual test >> cases their logs etc. >> >> > Kelly >> >> regards >> Shiraz >> >> > >> > On Sun, Feb 28, 2010 at 12:18 AM, shiraz hashim >> > <[email protected]> wrote: >> >> >> >> Hello, >> >> >> >> I am looking for an automated test framework for Linux based embedded >> >> environment. The objective is to have an automated, less bulky, >> >> efficient framework, which can be used often during development and >> >> also >> >> after for comprehensive/stressed testing. >> >> >> >> I came across autotest and STAF/STAX which seems to have similar goals >> >> and are based on python. >> >> >> >> While test cases are clearly separable, definable and controllable in >> >> STAX, I could not figure out how this is done in autotest. >> >> What I understand is that in autotest, one can execute a set of test >> >> cases >> >> through control file but how are individual test cases controlled and >> >> separated is not known to me. >> >> >> >> Can you please help me on this. >> >> Sorry, If I am asking a very naive question. Actually I am bit confused >> >> and not able to figure out what would be good for me. >> >> >> >> thanks >> >> >> >> -- >> >> regards >> >> Shiraz Hashim >> >> _______________________________________________ >> >> Autotest mailing list >> >> [email protected] >> >> http://test.kernel.org/cgi-bin/mailman/listinfo/autotest >> > >> > >> > >> > -- >> > K.D. Lucas >> > [email protected] >> > >> >> >> >> -- >> regards >> Shiraz Hashim > > > > -- > K.D. Lucas > [email protected] > -- regards Shiraz Hashim _______________________________________________ Autotest mailing list [email protected] http://test.kernel.org/cgi-bin/mailman/listinfo/autotest
