I did redirect stdout and stderr to /dev/null (I refer you to my post #23 for the screen shot of the setup)--but that redirection only applies to the last command in the chain. Thus, if I do as you suggest:
If you want to do backups AND config checks, you can change the cronjob to instead be something like: /mnt/tank/chkconfig.sh && /mnt/tank/bkpconfig.sh
...the redirection applies only to bkpconfig.sh, not to chkconfig.sh. But that said, that's a good point that I hadn't fully processed. It still seems like a fourth script whose only purpose is to call the first three is needless added complexity, but it may be the best way to go.
In any event, apparently I wasn't very clear about my reason for posting in this thread. It wasn't "this is broken, fix it for me." It was "it didn't work for me this way, but when I made this small change, it did." The point was to help others who might encounter the same problem I did--and there's nothing at all in the GUI that will tell you why chaining another script will cause the second one to fail. The subsequent discussion has been helpful, I think, and points out pros and cons of some possible ways to address the problem.
People will have questions about guides. Sometimes the questions will be because they didn't read thoroughly (or at all), sometimes they will read carefully and just not understand something, sometimes they'll need something not quite covered by the guide, and sometimes the guide is just wrong--and there are probably lots of other reasons I didn't even think of. If users can't ask questions, what is the purpose of a support forum?