when to believe error message

[rudbeckia]

I thought I would start a new post for this.

I posted a question earlier about an error message, that after running the tool repeatedly corrected itself without my changing anything.

This has happened across multiple tools, the tool will stop, give me an error, I just run it again, without changing anything and it works fine.

Why does this happen? Obviously the error is not real or it would not run correctly without changing anything. How do you know if you should pay attention to the error, or just keep running the tool over and over until it works?

Thanks

[DanPatterson]

more questions

• are these tools you are running being directly from arctoolbox and not through modelbuilder or arcpy?
• is the "failure" preceeded by running a tool that may require some time to complete? (eg large datasets)
• is the source data for the tool located on your local pc?  

It could be that the input for the tool isn't completely updated/ready for use even though one would assume that the tool in arctoolbox wouldn't allow for this.

A specific list of tools would be useful since others have complained of similar and perhaps related issues with errors in various scenarios.

I haven't experienced this, but I work with small locally stored datasets using tools directly from arctoolbox or via python scripts in my own custom toolboxes.

[rudbeckia]

• are these tools you are running being directly from arctoolbox and not through modelbuilder or arcpy? DIRECTLY FROM TOOLBOX
• is the "failure" preceeded by running a tool that may require some time to complete? (eg large datasets) TOOL COMPLETES IN AT MOST 5MIN
• is the source data for the tool located on your local pc?  NO A NETWORK

The two tools I am using are Distance Accumulation and Least Cost Corridor, although I have had gis do this across other tools

[DanPatterson]

To try:

move data off the network onto your local machine

when specifying output, keep the path simple (c:/test  or something clever)

specify the output to that folder and... specify a *.tif as the output extension so that you get a *.tif in a folder which would obviously have not spaces.

contact tech support, and document your attempts with file names inputs and output types and names.

working local on your own machine may not be suitable/possible for many, but it should be your first goto when trying to remove issues from the equation.

good luck... 

[RTPL_AU]

@rudbeckia 
They say that the mark of insanity is to do the same thing over & over hoping to see a different outcome. 

After using Pro for 10 years I can honestly say we will NEVER go insane....

What you see has been my experience with Pro from day 1.  I feel you cannot trust the software at all and even if you don't get an error you have to double check the output every time.  

When it works it can be a fantastic tool but when it is out on a bender it drives me nuts. 

Edit:
To add, some weird behaviour can be traced back to network latency, other background processes interfering with Pro at the point where it tried to read/write something on the network, etc  but, for me, of all the apps I run on my machines, Pro is the most unpredictable.

It is still worth doing troubleshooting - if only to make sure nothing is going on at a hardware or server level that can compromise your valuable data. 

[rudbeckia]

But even if its say something else running, the errors change, then the tool finally works

Does it just make up an error?

[rudbeckia]

For example, I got a "No Data" error, a "Path has Spaces" error and a "tool could not completed" type error, only to run the tool again and it worked 

[RTPL_AU]

"Does it just make up an error?"  That would imply that it is sentient and malicious or mischievous .......  🤣

"Path has Spaces"  is something new 3.5 
It probably occurs after you've saved a project and then try loading data from within a map to a geoprocessing task. Adding the data using the folder browser sees the task complete fine. 

You're not using data saved in OneDrive or similar by any chance?

[rudbeckia]

I definitely believe gis is being mean to me sometimes 🙂

In the example "Path has Spaces"...... i got this error 3 times, just kept running the tool, and eventually it worked fine. Without my changing anything.

Do you see what I am saying? Clearly that error wasn't real or it would not have worked eventually. Right?

[AlfredBaldenweck]

In related news, I spent like 4 hours off and on today replicating a bug, finding some wrinkles to it, tracking down those very specific wrinkles and writing down the steps to reproduce (so I did each one like 20 times) and then when it time to record the wrinkles, I could only reproduce 3 of the 5. This was the same project, session, and data. The only difference was a couple hours. Super frustrating.

I did send them anyway, with the note that I wasn't able to reproduce them when I tried to record.

The lesson here is to record before you do anything else.