[3882] in testers
many small mkserv bugfixes
daemon@ATHENA.MIT.EDU (Jonathon Weiss)
Tue Aug 18 19:06:18 1998
From: Jonathon Weiss <jweiss@MIT.EDU>
To: mkserv@MIT.EDU
Cc: testers@MIT.EDU, ghudson@MIT.EDU, rbasch@MIT.EDU, cat@MIT.EDU,
jweiss@MIT.EDU, dkk@MIT.EDU, ops@MIT.EDU
Date: Tue, 18 Aug 1998 19:06:04 EDT
A number of small bugs have been fixed in mkserv, I'm being lazy and
only sending one piece of mail, so you may not be interested in all of
these.
mkserv scripts should now have /usr/bsd in their path. Admittedly
this will be more useful on SGIs. :-) problem reported by rbasch and cat.
mkserv now invokes synctree with the -nosrcrules option (both for
copying things off the packs and for updating the root symlinks.)
This solves the problem of my .el files not getting updated when the
packs were completely rebuilt, since the old packs somehow ended up
with newer mod times on these files, and the .rconf file on the packs
appearantly overrode the rules mkserv passed to synctree.
mkserv now assumes /usr/afsws is never on /os. This seems to be the
case, and I cant see it changing. This allows mkserv to optomize and
not blow away /usr/afsws during the /os synctree only to re-create it
during the /srvd synctree. Never formally reported, but it's been
bothering me since shortly after /usr/afsws came into existance.
In order to maintain my sanity, I'm not worrying about backward
compatibility for 8.1 or earlier platforms when building the mkserv
binary. That means that the above changes apply only to 8.2 machines.
8.1 and earlier machines will continue to work but won't have these
fixes. If anyone really thinks that is a problem come talk to me and
we'll figure out what to do about it.
mkserv ops now brings iostat local (this patch was applied for both
8.1 and 8.2). Problem reported by dkk.
Jonathon
