Upstream has changed from providing a .war file to providing an unpackaged
webroot instead, so this won't upgrade cleanly. First pkg_delete unifi,
then rm -rf /usr/local/share/unifi/webapps/ROOT, then install. We can't
provide built packages of this port anyway, so this is relatively low
impact.
same machine. While this is of course done outside the rc.subr(8) framework,
the package's rc.d(8) script should not interfere with these instances.
Require exact pexp matches in order to achieve this. Patch by "nusenu",
developer of ansible-relayor, tested by me in various situations.
removing some which are no longer needed. ok pea@ (maintainer).
add a patch from upstream git (noticed after ok) fixing one of the default
config files.
Pyczmq exposes CZMQ functions via the cffi library's ABI access mode. No
compiler is required to use it. The czmq library is accessed with dlopen and a
set of parsed function declarations
ok jasper@
Notes:
- enable the use of -Wl,--as-needed, startup time is now "normal";
spotted by Ian McWilliam
- ntdb is now enabled but gets merged into the -main subpackage, since
upstream discusses its removal
- regedit disabled for now, because of a linker failure
- remove pidl Perl patches
ok sthen@
From Jan Klemkow (upstream and MAINTAINER).
pkg/DESCR:
This package contains additional tools for the UCSPI tool chains. It brings
support for using socks proxies on client side and tls on server and client
side.
fragmented ppp frames while reading from pty. From Sergey's mail:
=====
xl2tpd could not correctly handle fragmented ppp frame while read it
from pty if one of consequent read operation return an error.
That's happening because xl2tpd too often cleans the buffer and
assumes that each read frame operation is performed for new frame,
since xl2tpd uses single buffer for pty and for socket read operations.
The attached patch moves all pty related buffers into _call_ context
and corrects calculation of buffer position, what fixes issue with
handling of fragmented frames. Now we can use MTU > 1000. I tested
these changes with load (L2TP tunnel without IPSec) and all seems work
normally. No more "Protocol-Reject" frames.
=====
Performance is still poor (Sergey was seeing ~2Mb/s, I am seeing less
despite low cpu use) but the tunnel now seems reliable.
