This is a doctored up description from one of the developers...
" In the past, commit would cause a reread of the entire config and
init/cleanup for all the components.
The commit process was later optimized so that only the relevant
parts of the config will be processed. In the extreme case, if you do commit
on a config that didn't change, no work should happen.
"Commit full" does today what "commit" used to do in the past.
Sometimes you want to "shake up" the box, and cause the config to be read
again. Then you use commit full, which is a hidden option."
Disclaimer: I only recommend that hidden commands be used under the guidance
of JTAC....
HTHs
> -----Original Message-----
> From: juniper-nsp-bounces@puck.nether.net
> [mailto:juniper-nsp-bounces@puck.nether.net] On Behalf Of Eric Van Tol
> Sent: Thursday, December 18, 2003 1:22 PM
> To: juniper-nsp@puck.nether.net
> Subject: [j-nsp] commit full
>
>
> Hi all,
> Does anyone know what the 'commit full' command does? How is
> this hidden command different from the regular 'commit'
> command? I had a problem with an M5 today spitting out some
> SNMPD errors after one of my commits and the 'commit full'
> command seemed to fix it.
>
> TIA,
> eric
>
> _______________________________________________
> juniper-nsp mailing list juniper-nsp@puck.nether.net
> http://puck.nether.net/mailman/listinfo/junipe> r-nsp
>