No, the whitespace that gets stripped is at the *start* of the record.
This is what I put in, which works fine:
administrator@hinzelmann:~$ cat /etc/bind/pri.strangenoises.org
@ IN SOA ns0.strangenoises.org. admin.strangenoises.org. (
... etc, etc ...
;;;; MAKE MANUAL ENTRIES BELOW THIS LINE! ;;;;
NS ns2.strangenoises.org. ; because ispconfig only lets us enter two
(Note also newlines *after* the record.)
The bug is that the spaces on the NS line *before* the NS get stripped out when ISPConfig regenerates the file, and then the zone file is invalid.
So every time I change this zone in ISPConfig I then have to manually edit the file and re-insert those spaces and restart bind before the zone will work again.
You're presumably spooling the contents of the file below the MAKE MANUAL ENTRIES line from the old version of the file to the end of the new one, line by line, and - this is the mistake - you're trimming the not-so-superfluous spaces from the start of each line in the process. You're probably trimming from the end too, but there they really would be superfluous.
NS at the very start of a line in a zone file means the whole zone fails to load when bind is reloaded. But ISPConfig doesn't report the error, it sits silently until any and all backup nameservers have expired and then the domain stops resolving for the world.