>>>>> "pbh" == Paul B Henson <[email protected]> writes:
pbh> I've got a cron job running every hour on the backend servers
pbh> crawling around and fixing permissions on new directories :(.
To my view, if there's a problem it's first with the build system,
second with NFS. You can fix Solaris to do what you want, right---it
is like a 1-line change, and you've already found the line to change?
Why can't you just fix all your systems in a local hg repository? Is
that more painful than the cron job?
When I had a lot of NetBSD systems, I had plenty of local patches, and
with cvs and build.sh it wasn't a nightmare at all. Is the Solaris
build framework really so much less convenient, or is it just a
barrier to allowing the first local patch into your site? or you are
running a closed-source release where you have to roll over and beg
zfs-discuss mailing list