Yeah, it's well hidden right now. I'll move openvpn-gui to the main
OpenVPN project on GitHub today / later this week, so it will be easier
to find.
--
Samuli Seppänen
Community Manager
OpenVPN Technologies, Inc
irc freenode net: mattock
Oh, OK then. I didn't know about the fork.
ValdikSS
Hi,
As the subject line says, there's now an IRC channel for the OpenVPN
Windows team on Freenode at #openvpn-windows.
--
Samuli Seppänen
Community Manager
OpenVPN Technologies, Inc
irc freenode net: mattock
On Mon, Nov 9, 2015 at 3:56 PM, Gert Doering wrote:
> On Mon, Nov 09, 2015 at 03:35:23PM -0500, Selva Nair wrote:
>> Requires the installer to update the registry during an upgrade of exisitng
>> installations, or notify the end user of the change in the default value of
>> disconnect_on_suspend (
Hi,
On Tue, Nov 10, 2015 at 11:18:55AM -0500, Selva Nair wrote:
> On Mon, Nov 9, 2015 at 3:56 PM, Gert Doering wrote:
> > On Mon, Nov 09, 2015 at 03:35:23PM -0500, Selva Nair wrote:
> >> Requires the installer to update the registry during an upgrade of exisitng
> >> installations, or notify the
Hi,
On Fri, Oct 30, 2015 at 03:07:47PM +0100, Holger Kummert wrote:
> Am 30.10.2015 um 14:58 schrieb Steffan Karger:
[..]
> > Seems I forgot the * before the second msg_bufpos, sorry. msg_bufpos
> > itself is a pointer indeed, but *msg_bufpos is (used as) an offset
> > into msg_buf:
> >
> > m
Your patch has been applied to the master branch.
commit 3671bd185a914dc1d91c8a967e1c3a14dedacd32
Author: Steffan Karger
List-Post: openvpn-devel@lists.sourceforge.net
Date: Fri Nov 6 08:42:39 2015 +0100
Fix (potential) memory leak in init_route_list()
Signed-off-by: Steffan Karger
Hi,
On Mon, Nov 09, 2015 at 07:11:16AM +0100, Arne Schwabe wrote:
> > I can see why we do this gc_addspecial() dance in socket.c (because the
> > elements returned by getaddrinfo() are being pointed at, not copied).
> >
> > This usage here seems unneeded, as the code will walk netlist, copy
> > o
ACK. It does increase the code slightly (because while the compiler can
optimize away the if() clause in most cases, for each FATAL msg(), the
new exit() leaves a few bytes...) - but the increase is truly marginal
(602843 to 603143 bytes "text segment" on my i386 linux) and it helps
with coverity
... and while at it, fix warning about losing "const" qualifier on
return.
Signed-off-by: Gert Doering
---
src/openvpn/tun.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/openvpn/tun.c b/src/openvpn/tun.c
index 24a61ec..6679330 100644
--- a/src/openvpn/tun.c
+++ b/s
The code always tried to copy-out a "struct sockaddr_in6" even for IPv4
results, which reads more bytes than getaddrinfo() is guaranteed to
allocate.
Now, look at ai->ai_family and only copy "struct sockaddr" for IPv4.
Also, reformat this block of code to comply to coding style.
This is a specif
On 10.11.2015 22:58, Gert Doering wrote:
> This is a specific 2.3 bug as the code in master (to be 2.4) has been
> completely rewritten to properly handle dual-stack and multiple responses
> from getaddrinfo() proper.
>
> Bug found by Daniel Hirche using "gcc -fsanitize=address". No possible
> exp
11 matches
Mail list logo
