For now, BROKERNL_SSL* is designed for port that defines GNU_CONFIGURE
and USES=pkgconfig to get libssl's include and library paths by means
of pkg-config --cflags/--libs or just use CFLAGS etc. including ${LOCALBASE}
net/mpd5 does not use pkgconfig nor GNU configure and since last update
it forces usage of system libssl despite of presence of other *SSL packages,
so BROKEN_SSL* is unneeded and does only harm here.
Import upstream revisions r2258-r2260: do not use -I${LOCALBASE}/include
by default when option NG_IPACCT is off. If NG_IPACCT is on,
use -nostdinc to force needed header inclusion order to force
usage of base system libssl.
Bump PORTREVISION as package can change if built with default options
when net-mgmt/ng_ipacct is installed.
PR: 233471
openssl-1.1.1 is in FreeBSD12 base system and that was already
resolved in r482627.
While here, explicitly mark the port incompatible with alternative SSL
libraries such as LibreSSL due to dependency of base system libfetch
that links with base system libssl.
PR: 233471
Now as FreeBSD 12 got openssl-1.1.1 in base system and
/usr/include/openssl/ssl.h and /usr/local/include/openssl/ssl.h
are indentical, it is safe to allow building net/mpd5
with DEFAULT_VERSIONS having ssl=openssl111 and no reasons to break
the build.
However, resulting binary still links with base system /usr/lib/libssl.so
because of libfetch(3) needs it and mpd5 links with libfetch too.
mpd5 daemon can be used as replacement for stock ppp(8) daemon
to provide global connectivity (PPPoE, mobile networks etc.)
Follow ppp(8) behavior and run startup script early before
NETWORKING (and netwait), so other services that require
working networking to be operational before starting won't fail.
Bump PORTREVISION.
mpd5 successfully compiles with openssl-devel or openssl111
but cannot be linked because it links with base system libfetch(3)
that currently links with /usr/lib/libssl.so.8 (openssl 1.0.2p-freebsd).
So, one should not attempt to build mpd5 with newer openssl version
until it's imported to base system so mpd5 won't link with two distinct
openssl versions.
PR: 231599
and add new way to deal with same problem
following upstream changes r22478, r2249:
- reverse r2246 as it used wrong way to deal with the problem;
- make it possible to override results of LCP negotiation
and set interface MTU to specified value with a command:
set iface mtu NNN override
New option "override" is not documented yet.
This is dangerous option as it can violate RFC 1661
and should be used with caution and at your own risk.
It is useful when you deal with broken PPP peer
negotiating too low value while higher MTU is known to work.
Add a fix from upstream r2246:
Fix bug #62: improve LCP MRU negotiation in some unusual cases
sometimes happening while using Orange S.A. services.
This allow us to setup PPPoE client connection interface using mtu=1492
instead of 1456.
Reported by: Xavier Lemaire
Tested by: Xavier Lemaire
Also the compat NO_INSTALL_MANPAGES shim.
Differential Revision: https://reviews.freebsd.org/D730
Reviewed by: antoine
With hat: portmgr
Sponsored by: Absolight
- Use PLIST_DIRSTRY to avoid warnings when deinstalling.
- Set PROVIDE to mpd5 in startup script (it currently confuses
bsdadminscripts' commands)
PR: 186531
Submitted by: olgeni
Approved by: maintainer timeout (120 days)
literal name_enable wherever possible, and ${name}_enable
when it's not, to prepare for the demise of set_rcvar().
In cases where I had to hand-edit unusual instances also
modify formatting slightly to be more uniform (and in
some cases, correct). This includes adding some $FreeBSD$
tags, and most importantly moving rcvar= to right after
name= so it's clear that one is derived from the other.
* New features:
* Added `set link redial-delay ...` command.
* Print global filters on `show iface|customer` commands.
* Added protocol/port forwarding support for NAT.
* Added utmpx support on 9-CURRENT.
* Bugfixes:
* Fix memory leaks on pptp and radius on some reason.
* Really make RESULT a mandatory option in ext-auth.
Submitted by: maintainer
Feature safe: yes
