the RCD_SCRIPTS rc.d script(s) to the PLIST.
This GENERATE_PLIST idea is part of Greg A. Woods'
PR #22954.
This helps when the RC_SCRIPTS are installed to
a different ${RCD_SCRIPTS_EXAMPLEDIR}. (Later,
the default RCD_SCRIPTS_EXAMPLEDIR will be changed
to be more clear that they are the examples.)
These patches also remove the etc/rc.d/ scripts from PLISTs
(of packages that use RCD_SCRIPTS). (This also removes
now unused references from openssh* makefiles. Note that
qmail package has not been changed yet.)
I have been doing automatic PLIST registration for RC_SCRIPTS
for over a year. Not all of these packages have been tested,
but many have been tested and used.
Somethings maybe to do:
- a few packages still manually install the rc.d scripts to
hard-coded etc/rc.d. These need to be fixed.
- maybe remove from mk/${OPSYS}.pkg.dist mtree specifications too.
* Fix checkrad.pl::cisco_snmp and usrhiper_snmp so that communities
other than "public" can be used too.
* Error out on superflous command line args (optind >= argc)
* Encrypt CHAP-Password in radclient so that CHAP can be tested too
* Add "wildcard" option to realms - if you set this option, you can
match on the entire username using shell wildcards in the realms file.
* If the nastype in /etc/raddb/naslist is set to 'none' for a nas,
logins on that nas will have no simultaneous use restrictions imposed
and those logins will not count towards the total amount of logins.
* Removed 'raduse' and manpage.
* Can disable radutmp with "-u none" (likewise radwtmp with "-W none",
though we already had "-w" for that).
* Call checkrad with an extra argument, the framed IP address.
* Fix bug where $INCLUDEing a file without any records would
cause the rest of the original file to be ignored
* Support for 64-bit wide integers (integer8 type). If compiled with
gcc, users file can contain both hex and decimal 64-bit values,
and 64-bit values are printed in decimal. With other compilers
(no "long long" support) only hex 64-bit values are supported.
* Change "_" to "-" in dictionary.redback and change the 64 bits
values to integer8.
* Moved the dictionary files to /usr/local/share/radius. Now
only /etc/raddb/dictionary remains that $INCLUDES all the others.
For new installs only; existing installations won't be changed
* Make integer in the format string of sprintf(buf, "%03d:%.20s", ..) in
make_wtmp() unsigned (%u), otherwise the sprintf could in some cases
(negative NAS-Port, very long NAS-name) overrun the buffer by one
byte, overwriting the least significant byte of the return address
on the stack with a \0. Not sure if this is exploitable or not,
but it could be a security problem.
* Make sure ut.login (struct radutmp member) is treated everywhere
as a NON-zero terminated string.
* Make sure unsigned integers are used in all places using lvalues
(32 bits radius values), especially nas_port.
* radrelay: update id of packet when retransmitting.
* Print an error and free the request struct if we receive an unknown
packet type.
* rad_check_multi: if username/NAS/port match, don't count as dup.
- Honour PKG_SYSCONFDIR.
- Use OWN_DIRS_PERMS to handle the /var/log/radacct directory.
- Use RCD_SCRIPTS to install a NetBSD rc.d compatible script.
Bump PKGREVISION to 1.
* Bug fixes
* Clean up BSD password file support again
* Fixed bug in auth.c wrt decoding of short password attributes.
* Server now replies to Status-Server inquiries.
* Add "-v" switch to radzap (suggested by Dan Harkless).
* Add "-Z" switch to radiusd, prevents logging of passwords for -y
(suggested by Dan Harkless).
* Change "duplicate detected" message to actually show _what_ authentication
packet causes the duplicate
* Wrote manualpages for radlast,radzap, radwho, radclient, radtest,
radwatch and raduse. Yes, documentation!
* Added -P pidfile option to radiusd and radrelay
See the doc/ChangeLog file for a comple list of changes!
