I got this error last night while running a data step on a large set of data:
ERROR: Write to WORK.ALL.DATA failed. File is full and may be damaged.
The log says that 19million observations were read but my data set has about twice that.
I changed the default work folder in SASV9.CFG to an external 2000G drive a while back so there should be enough space. Any idea what is going on? Is there a setting elsewhere that limits the size of these temp work files? Will I have any better luck if I use permanent user files? I'm using EG 4.1.
I'm providing an update and possible solution in case anyone else encounters this problem ...
I think the issue is the limitation on file size for hard drives formatted using FAT32 which has a maximum file size of 4GB. Most new external drives are formatted using FAT32 by default. I think the fix is to reformat using NTFS.