helping impaired inittab lines respawn _almost_ too rapidly

[Frank Winans]

When working with multiuser applications, you often arrive
at a state where the server cannot run several lines of
/etc/inittab right now, but will be able to run them ok in the
near future. It is annoying that you must either endure the
'respawning too rapidly' messages on the console, or else
go edit inittab to disable those impacted lines.

It seems there is a need for a buffer script that runs the
line of inittab after a 14 second delay, if the last time it ran
was in the same hh:mm time as 'now.' So an inittab line

pk33:345:choke_it pk33 d3 -33 -t ttyaa33

could stutter along quietly until the subsystem demon was
launched to service this client line, and the tty in question was
available for the client line to open.

Bash script 'choke_it' will
a) create /var/lock/initguys/ if it doesn't exist yet
b) if file /var/lock/initguys/pk33 both exists and has
same timestamp hh:mm as 'now' then sleep 14 seconds
c) touch /var/lock/initguys/pk33
d) exec pk33 d3 -33 -t ttyaa13
So the line still respawns every 14 seconds if problems exist,
but once the problem is resolved it runs normally. The trip point
for init is 10 respawns in 120 seconds, so init stays quiet.