[UI.Vision RPA] No tab for dashboard yet

Hi,

I was doing some tests with RPA command line and suddendIy a pop-up appeared in my Firefox browser when Firefox and RPA launched. And the Macro didn’t start :
image

I have browsed the forum and I found that some users have encountred the problem but with no solution.

I don’t know if there is something to do with memory (that’s what the other threads suggested). When it happened I had 50% memory left, but indeed firefox and RPA were slow to open at this time. After closing the browser everything came fine.

I manage the problem this way (but it didn’t happen again so far). My first lines in my macros are : I “csvSave” a file (ok.csv) an I do a “LocalStorageExport”. My command line is called by a program which test in a loop if the ok.csv exists. If not after let’s say 30 seconds, it kills Firefox and start again the macro, and there is a loop as well with a tempo. With this code, I assume that it is a temporary problem.

But do you have more information about this problem or a better solution ?

Thanks.

Eric

1 Like

This error shows up if UIVision inside the browser is not ready in time for the command line. So it has nothing to do with the macro, it happens before UIVision IDE even opens. As you observed, it happens mostly if the browser is really slow - but actually it should not happen at all. It clearly is a bug.

Question: For the command line, are you using the new method with the ui.vision.html file?

Hi,

Thanks for your answer.

Indeed I use the “autostart HTML page” ui.vision.html as it is said in https://ui.vision/rpa/docs#cmd.

I am brand new on ui.vision so this is the only way I know (I didn’t try to analyze or create my own ui.vision.html file).

Clearly, I had the feeling it was a “tempo” problem". It seems that the IDE didn’t start early enough and it stays in standby.

That why I scripted a test in the macro to save a file to be sure from the point of view of the program that the macro started.

Thanks.

Eric.

Thanks for the info. It is good that you have a workaround.

Also, I added this issue to our todo list for some more testing.