Pascal,

 you've done exactly what we were planning to develop ! So you can imagine that I'm impatient to test it !

When will it be released ?

Is it possible to use the environment variables to name the results files ? With this possibility, there will be no need to use the "Archive the artifacts" option.

What do you think of this idea ?

Hello!

 First of all, I wish to thank you very much for a usefull plugin!

 I have a following question. The plugins provides a possibility to specify a URL to the suiteFile (e.g. in my case http://localhost/testslib/html/Login.html). As I understand a local copy of the suite file is created and executed. Unfortunately, I never succeeded to run a test providing an URL to the suiteFile because the test files (linked from the suite file) seem not to be downloaded. All I see the following error in the test's iframe of the TestRunner:

HTTP ERROR: 404

Not foundRequestURI=/selenium-server/tests/tests/login.html

Powered by Jetty://

 Help would be greatly appreciated.

Thank you in advance,

Daniel Tkatch

First off, thank you. This is a very help piece of coding.

I have a usage question. We are doing something like this but not so nicely integrated into Hudson. We run the selenium tests with ant and then parse the html file into Junit style reports.

What I would like to do is to simply use what you have developed and run my TestSuite.html from the Hudson dashboard rather then the build script, it will make the process much more flexible and easy to use. What I don't understand is how to pass some extra parameters to the "SeleniumHQ htmlSuite Run" build step.

How would I do that? Here is a listing of the parameters:
haltonfailure="true"
multiWindow="true"
timeoutInSeconds="3000"
firefoxProfileTemplate="/firefox-https-profile"

http://release.seleniumhq.org/selenium-remote-control/0.9.2/doc/server/org/openqa/selenium/server/ant/SeleneseAntTask.html

Hi Pascal Martin,

 First of all I appreciate the great work you did!

 Currently, I have tried following stuffs:

    Hudson 1.310

    Selenium HQ plugin 0.2

    Browser : *firfox, *ietha,etc..,

 But Selenium IDE testscripts are NOT running at all

Could you please help out?

-------------------------------------------------------------------------------

 Previously I used

     Hudson 1.306

    Selenium HQ plugin 0.1

    Browser : *firfox, *ietha,etc..,

That time all worked fine 

Thanks,

Raghuraman.Shanmugam

Is it possible to run a set of tests ?

e.g.
<selenese
files="c:\absolute\path\to\my\test*.html"
I don't want to maintain a test suite manually.

Thank you

Did anybody experience resource conflicts - when multiple SeleniumHQ tasks attempt to use the same browser / Selenium server JAR?
My example: two builds complete closely in time, both start (different) SeleniumHQ smoketests using *iexplore, with a 90% probability neither of the test won't complete to success. After this:
==
12:16:30.712 INFO - Java: Sun Microsystems Inc. 14.1-b02
12:16:30.712 INFO - OS: Windows XP 5.1 x86
12:16:30.728 INFO - v1.0.1 [2696], with Core v (VERSION) [@REVISION@]
12:16:30.931 INFO - Version Jetty/5.1.x
12:16:30.931 INFO - Started HttpContext[/selenium-server/driver,/selenium-server/driver]
12:16:30.931 INFO - Started HttpContext[/selenium-server,/selenium-server]
12:16:30.931 INFO - Started HttpContext[/,/]
12:16:31.181 INFO - Started SocketListener on 0.0.0.0:4332
12:16:31.181 INFO - Started org.mortbay.jetty.Server@dbe178
12:16:33.853 INFO - Launching Embedded Internet Explorer...
12:16:34.884 INFO - Launching Internet Explorer HTA...
==
nothing happens. A fantom iexplore.exe process is reported by the task manager; all further attempts to run SeleniumHQ steps with *iexplore also fail until this process is killed.

Workarounds:
1. Spread the SeleniumHQ runs in time. Ensure no overlapping.
2. Use the Build Timeout plugin to abort the stuck job. I'm not sure whether this will correctly kill the stuck browser process though.

—
Question closed (follow the answer of Selenium Grid team)

Selenium Grid does not officially support running multiple Internet Explorer on a single Windows machine. This is mostly because: People who know IE better than I do (Dan Fabulich) tell me that if you run 2 browsers as the same user in HTA mode they end up sharing a singleton instance in memory, which could cause problems. The *iexplore mode is changing the registry settings at each session start/end to have IE use a specific Remote Control as HTTP proxy. If you run multiple Remote Controls at the same time you can see the problems coming! Currently, the only robust solution for running multiple IE instances on a single machine with Selenium Grid is to use virtualization (multiple VMs, a single IE instance per VM).

How to use slave machine to run test suite ?

I've got a lot of problems making it working

is there a tutorial ?

How do configure selenium-server.jar path in order to allow the slave to access it ? (copy it on slave manually ?)
How do create a selenium test suite that will make the slave be able to locate each tests (currently path are relative in TestSuite file) ?

thanks in advance