var/db/sysmerge/etcsum is created after the cksum calls and etcsum is
part of the etc set).
"sure" deraadt@

This way, the first time sysmerge is run on a new box, it will already
have reference sums needed for automatic processing.
Original idea from and discussed with deraadt@
"commit it" deraadt@, tweaks and ok millert

Update the BSD.x11.dist version to the one in xenocara.

crontab(1) does.
ok miod@

in preparation for improvements in /etc/daily and /etc/weekly
using feedback and suggestions from jmc@ and sthen@
ok jmc@, and sthen@ agreed with the general direction

ridiculous; xdm is totally broken since it never starts anything which resembles
a "login shell".  As a result, no configuration is brought into a process
context to give to future xterms or *sh shells and thus cause them to run their
.profile or .kshrc or such a thing, to get futher configuration.  Therefore
people are left with a totally bland unconfigured Unix environment in their
xterms, and don't know how to change this since .profile is ignored.  This
problem shows hundreds of thousands of google hits.  xdm is fundamentally
broken, but we must solve this also for the startx methods, too, and for people
running csh.  It is clear that .xsession is not a solution to this problem at
all (that is, assuming the people who suggest such a thing really mean a
.xsession file with the execute bit set).  This now becomes the recommended
way for new users to get out of this stupid situation; if someone does not
like it they can change it or delete it.  Few will.  Just watch.
ok kettenis guenther millert

since we perhaps cannot assume that is what kernels will be in the
pathetically stupid world that SGI created

build the sets before we build the media.  While there we can get rid
of DESTDIR/snapshot too, and simply install straight into RELEASEDIR.
(This also ends up saving an astounding amount of traffic/latency in a nfs
environment)

some of the sets)

"please enable it" deraadt@

get it in, deraadt@

ok deraadt@ beck@ reyk@ phessler@

ok jsing@

ok deraadt@

pubkey authentication users with the correct file permissions a bit easier.
ok djm krw henning miod and many others

approved by deraadt@, ok thib@

we've found a better solution for chrooted applications.

replace IPv4 and IPv6 loopback zones with BIND's autogenerated empty zones.
move root.hint to /var/named/etc
remove empty directory /var/named/standard

what the daemon does - it is a relayer that pays attention to the status
of pools of hosts; not a status checkers that happens to do some relaying

At this time, there is no sensitive information in that file.
ok beck@, millert@, jdixon@, deraadt@

ok dlg@

is anything but useful. Nevertheless this was a very good platform to use as
an OpenBSD/*arm* starting point, but nowadays this platform is not necessary
anymore.
Somehow these CATS motherboard are just like AUI hubs, with the difference that,
when my AUI hub caught fire, 10base5 support was not removed from the networking
code.

(this decreases build time a bit too)
ok todd (and i assume legacy-architecture-miod would not object either)

use of a GZIP environment variable for extra command flags.
ok deraadt@, millert@

ok miod@, henning@

ok drahn@

ok henning@ deraadt@

ok claudio@

releases...