The actual text of the error message was:
DBCC CHECKDB (<Database name>) WITH all_errormsgs, no_infomsgs, physical_only, tableresults executed by SQLLaptop\SQLSoldier terminated abnormally due to error state 6. Elapsed time: 0 hours 0 minutes 30 seconds.
To make a long story short, I was testing a pre-beta version of a utility to make running DBCC CHECKDB easy for DBAs who are new to the process (more on this later after I help them work out the bugs). It seemed to work really well for small databases but would crash for large database. I noticed a pattern as the time for the error message was always 30 seconds into running CHECKDB. It seemed obvious to that the utility was using the default query timeout value of 30 seconds and killing CHECKDB at the 30 second point.
So the cause of my abnormal termination due to error code 6 was that the process was cancelled by the end user. I was able to confirm this by manually killing a running CHECKDB both by just clicking cancel in the query window and by running the KILL command in another window. It generated the same error though with a much smaller time frame than 30 seconds.
I can’t say that this is the only scenario that would generate this particular error, but this is definitely one cause that consistently generates the error.