"gsnell@datasavantconsulting.com" wrote: > > Just a quick heads-up to those of you with error-scanning > programs/functions that I have just run across the following missed error. > This was a V8 batch job creating SAS datasets from Oracle tables using PROC > SQL. It was a simple pull without any joins or conditions. Here is the > only indication of failure I received: > > --------------- > Segmentation Violation In Task ( SQL ] > Fault Occurred at [0x103ebc54(?)] > > Task Traceback > > sqxcrta -- bridge stub > sqlloop -- bridge stub > NOTE: The SAS System stopped processing this step because of errors. > Stack overflow detected In Task ( SQL ] > Fault Occurred at [0x103ebc54(?)] > > 272 quit; > --------------- > > But check-out the notes at the end of my job: > > --------------- > NOTE: SAS Institute Inc., SAS Campus Drive, Cary, NC USA 27513-2414 > NOTE: The SAS System used: > real time 28:01.32 > cpu time 12:47.91 > --------------- > > Except for the simple note about "stopped processing" there is no other > "typical" indication that an error occurred. Looks like I need to add > "stopped processing" when scanning for "errors". > > Regards, > > Gregg P. Snell > Data Savant Consulting > > -------------------------------------------------------------------- > mail2web - Check your email from the web at > http://mail2web.com/ .

Another one that is not common, but I think needs review is "multiple datasets with repeats of by values". I have added this to my scanlog routines. This one indicates there a duplicate by values on both datasets and indicates the merge would not be reliable in most cases.