RE: install_initd tests: new issues

>Yes, agreed, we need to balance any calls to install_initd
>with remove_initd. That is in the per testcase cleanup.
>I'm guessing that any extra remove_initd's
>would be harmless

I was going to add this in but then realized I'm not sure
how test result codes get set.  There are these individual
lines "tpresult" - they don't get things from the exit
status of the last command, do they?  In other words, I
don't want to mess up something related to reporting
by adding extra calls.