'Flush' launchd without rebooting server?

Question

I created a daily 6am disk usage report, then moved it to 8am... but then it began running at 6am and 8am... so I unloaded the job and found the 8am job kept running, even though there's no mention of it in the 'launchd list' results!
I've used the following one-liner to print out the contents of all the loaded launchd jobs, but none of them have been overwritten by my plist, or in any other way call the script that my launchd job was created to run... very weird.
for i in `launchctl list | awk '{ print $NF }'`; do launchctl list $i ; done
I've unloaded, reloaded, and anything else I can think of at the commandline or in Lingon (launchd interface), but can't stop this job!
Is there any way to restart launchd, or flush its cache/database of jobs, without rebooting the server??

Answer 1

Thanks for the thought BDAqua, but cron keeps its jobs in /etc/crontab while launchd uses XML plists and each job is a seperate file.
I haven't touched cron, and cant see a way that launchctl or Lingon (both solely launchd utils) would have injected the job into cron, but I checked anyway and found it wasn't the case. cron has been deprecated in favour of launchd.
launchd is the root of all processes, so i understand if there's no mechanism for restarting it (because it would kill everything else running) but I'm hoping there's a way to get it to rescan all the plist locations like it does on boot, so it will hopefully drop the spurious job and reingest the updated plist.