Process crashed - Can I get the data?


I’ve had an attended bot process running for a few days and this morning it has inexplicably failed. I don’t want to have to do all the work again, and was wondering if the data table is stored anywhere as a temp file that I can use to at least get out what had been completed?


Im afraid not, DataTable are in memory only… if you are running a very long process like that, then you may consider changing your project using a Try Catch and in the catch you verify the current data and save it to an excel file and that to the disk…

Funnily enough, did have the try catch for exactly this reason, but when I was debugging it meant I couldn’t see why it was failing. I then ran it working and didn’t have a stop in the studio so kept it going. After all, it was working. I was hoping to get to the end of the process

I’ve now put the working sequence back into the try/catch, and broken the input down into smaller batches. Unfortunately I’m having to redo the work!

Hi @SoqedHozi

I would like to suggest something here…

if the datatable that you are looping through contain critical data of the process, and you want to track the status of each item while the process is going on, or when it come across a situation like this, I suggest you to add the data in the datatable to an excel file before looping it. Once done, add a status column as well.

next, you can loop through the datatable using a for each row activity, and at the end of each iteration, make sure to update the status column for each row with the status success or failure.

This way, you will be able to track in case of a failure…

Also, add logging activities so it would be easier for you to track the errors too :slight_smile:

1 Like

Yep, this is all stuff I’d normally do but i didn’t in this case and hoped I had a fallback!

On issue logging, I’ve only ever been able to write stuff with a log message - is there a better way to get the error reason when inside a try catch? Generally I just use “exception” as the reason to. Finally, is there a good way to “retry”, without using retry scope? I don’t particularly like the maintenance issue of dropping the try into the catch with copy/paste.