Skip to main content

Hi All,

 

I have a workspace published to FME server with a few MI tab readers, nothing special. (read several MI tab files-> manipulate attributes -> extract coordinates -> output to CSV). For whatever reason when the job runs per its scheduled time, It fails with the following error in the log:

MITAB Reader: Failed to open and parse file 'E:\\ProgramData\\Safe Software\\FME Server\\repositories\\Workspaces\\SSALocator\\RUR_ADS_PROPNUM.TAB'. Make sure that the file exists and is valid

 

The tab file is located on a network share so i'm not sure why its referencing the location above? Theres no files here other than a copy of the workspace.

 

It doesn't have a long path name (>260) If i manually click resubmit job, it works fine so i don't believe its file access(I used the full UNC path for the readers). Any suggestions are appreciated.

Could it be that that path is hardcoded in your workspace?

If it is, and that network share is not mapped as the E: drive in FME Server it can't find the file. Try with a UNC path instead (or make a parameter out of it so you can set it while setting up the workspace on Server)


Hi @kieranmg,

This is the location that files are stored in if you choose to publish them along with the workspace up to FME Server. Is the source dataset referenced by a published parameter? In the Schedule set up can you check that the value of the Source dataset is the correct filepath? I suspect it will reference a macros $(FME_MF_DIR), changing this value and saving the schedule should resolve the issue.

Note the schedule will retain the values specified for any published parameters when it was first created, so it's possible that when this workspace was first published/schedule created the data was uploaded and stored in FME Server Share along with the workspace, however if you were to ever republish this workspace and not check the box to upload data files they would no longer exist there and the schedule would fail.


Thanks for the suggestions @hollyatsafe,@redgeographics

Yep i think it was something to do with the Schedule resources. I recreated it and noticed the $(FME_MF_DIR) listed twice within the top resources was replaced by the full paths. I'm not sure why this is here though, those 2 resources are already listed below on separate source lines with the full path. Screenshot shows before, the only change from recreating the schedule is the top line has had those macros replaced to show full paths. Is there a reason those 2 resources are referenced twice? I havent uploaded the data with the workspace and its not doing anything fancy. Either way its working so thank you.


Thanks for the suggestions @hollyatsafe,@redgeographics

Yep i think it was something to do with the Schedule resources. I recreated it and noticed the $(FME_MF_DIR) listed twice within the top resources was replaced by the full paths. I'm not sure why this is here though, those 2 resources are already listed below on separate source lines with the full path. Screenshot shows before, the only change from recreating the schedule is the top line has had those macros replaced to show full paths. Is there a reason those 2 resources are referenced twice? I havent uploaded the data with the workspace and its not doing anything fancy. Either way its working so thank you.

Hi @kieranmg,

I'm glad you were able to resolve the issue. Unfortunately if you did not upload the data with the workspace I can't explain why it populated with that FME Server macros. The only other time I have seen this is if on your local machine the data is stored in exactly the same directory as your .fmw.

 

 

From the screen shot provided I believe the first Source MITAB Published Parameter is the one that is created by default in FME Workbench when you add any new reader. Since you created and are using other published parameters to reference these files you can probably delete this published parameter altogether to avoid future confusion.

 


Reply