I am having a very strange issue trying to run automated tests. Basically I have a cronjob that runs the following command to start our testing script-
Now, when I run this script normally, as in, opening up eggplant from the dock and running the script, I have no problems at all.
However, when I run it from the terminal, whether via cronjob or or entering the above command manually, the script hangs in the same place, and the screengrab of the crash indicates that it’s not seeing what it actually there. It is in fact seeing a portion of the window it’s supposed to be clicking on, but it’s cutoff as if it were outside the screen in the upper right corner, which makes no sense since it’s in the middle. I’ve attached the screengrab from Eggplant’s logs to give you an idea of what I mean.
This only occurs when the script is run via terminal, anyone have any ideas?
Edit- Well I tried to add the screenshot, it doesn’t seem to want to upload.
That is a bit perplexing. A couple of questions/notes:[list][]What version of Eggplant are you using?[]Make sure you are running the command (or cron job) as the same user who you normally run from. Eggplant preferences are stored by user account.[]Your command indicates that you run the Schedule for your suite and then run a single script from that same suite, I just want to make sure that’s what you intended.[]I think your upload might be failing based on image type, I’m sure you can upload GIF and JPG but I’ll try to add a few more types so try uploading again.[/list:u]
The command is being run from the only user on the system.
I want to run a single script from a suite that basically runs some other suites. That much seems to be working just fine, i just get the strange failure i spoke of.
I changed the shot to jpeg and you ought o be able to see what I’m talking about now. The strange piece of a screen that you see is part of larger window which has the buttons that eggplant is looking for. but sfor some reason that it what it sees at the failure point when the script is run from the terminal. I don’t get this issue when running from eggplant, and can in fact see on the SUT that the screen is normal no matter how I run it.
Try forcing a the VNC server and Eggplant to sync up with a forced refresh. If this doesn’t solve your problem, there is something else going on maybe at a network level that is borking the dirty region of the screen to be transmitted or updated.
That does seem to help. The problem is it keeps having the issue in different places, and short of throwing a refresh screen in after nearly every command I can’t think of a way around it.
First I would recommend upgrading to 4.1 (which is a free upgrade) which did address some small issues when running CLI. I’m not sure if your error is specifically addressed but we it might be fixed.
If you must you can force screens to always fresh using one of the Run Options under Eggplant Preferences but that really shouldn’t be necessary. Let us know if you have to use this workaround and we’ll investigate why.
This brings up an issue of general exception handling. You may want to looking into the SenseTalk try … catch … end try. This will bring into light the ability to handle an exception, do something like a refresh, and then try again. You can also look into the Omega13 feature which may be of some use to you as well. Try typing Omega13 into your run script window for fun
