similar to: Bug in mingw/device.c: addrinfo structure is freed too early

on systems that has separate IPv4/v6 socket layer (i.e. IPv4 packet does not get routed to AF_INET6 socket) rsync --daemon would accept IPv6 sessions only. open_socket_in() tries to deal with the situation, but it was not enough. here's the patch. (it is required on all *BSDs to accept both IPv4 and IPv6 connections with --daemon mode) itojun --- ? configure.lineno ? lib/dummy

Hi Arjen, in server/upsd.c r731, you moved the call conf_load() from after check_perms() (~ l.1020) to before setupsignals() (~ l.989). The problem is that conf_load() needs to open the ups driver socket, and it assumes that STATEPATH is the current working directory. The directory is only set in l.1016. Therefore, the first attempt to open a socket will always fail. From the user's point of

Hello, I use hosts that are dual stack configured (IPv4 and IPv6) and it happens that connectivity through one or the other is broken and timeouts. In these case connection to the SSH server can take quite some time as ssh waits for the first address to timeout before trying the next. So I gave a stab at implementing RFC 8305. This patch implements part of it in sshconnect.c. * It does not do

Hello all, I recently have a problem with multiple addresses and address families. Problem is simple, i have some hosts with IPv4 access only and some with IPv6 access. This wouldn't be big problem if I had a stable IP addresses. But sometimes I move to another network with complete different addresses. So I created patch which on option BindAddress accept list of addresses. With ip I solved

Hello all, I am wondering where one would submit a patch to OpenSSH cvs version? I have written a patch that allows the user to set a timeout the ssh clients connection attempt. I added this because many of us use ssh in automated scripts and in some cases machines may no longer be reachable, rather than wait you can set the timeout to say 3 seconds (-z switch). -- Larry Cashdollar

With the Ipv6 patch (r708), I get: upsclient.c: In function `upscli_connect': upsclient.c:469: `AI_ADDRCONFIG' undeclared (first use in this function) upsclient.c:469: (Each undeclared identifier is reported only once upsclient.c:469: for each function it appears in.) Even if it doesn't break IPv4 support, it may break portability, as IPv6 seems to require specific functions that are

- perform dns lookup on hostname, - randomize local tunnel port - bump rpm spec version updated to include danpb's feedback regarding getaddrinfo --- main.c | 36 ++++++++++++++++++------------------ ovirt-viewer.spec | 9 ++++++++- tunnel.c | 29 ++++++++++++++++++++--------- 3 files changed, 46 insertions(+), 28 deletions(-) diff --git a/main.c b/main.c index

I've documented a desire to do this for a while, time to actually follow through and support connecting as a client to a TCP server. Note that it is desirable to support the plugin connecting to an encrypted server over TCP, then exposing the raw data over a local Unix socket; that aspect requires yet more work, left for another day. But even allowing an old-style client to connect to an

Attached is a very small patch that allows the ssh clients to use the socks5 library. It should work with socks4 but is untested. Tested on linux only configure --with-socks configure --with-socks5 Also included is a configure option to disable scp statistics --disable-scp-stats modified files openssh-2.9.9p2/acconfig.h openssh-2.9.9p2/channels.c openssh-2.9.9p2/configure.in

>Submitter-Id: current-users >Originator: Alex Vasylenko >Organization: >Confidential: no >Synopsis: [patch] net/rsync: problems in client name lookup code >Severity: non-critical >Priority: low >Category: ports >Class: sw-bug >Release: FreeBSD 4.8-RELEASE-p16 i386 >Environment: System: FreeBSD 4.8-RELEASE-p16 >Description: rsync does reverse name

> Winton-- > > Excellent! Absolutely wonderful. > > I'm wondering which apps/encapsulators support 4A? This gets me > around > the DNS leakage problem quite nicely. > > Incidentally, we do need SOCKS5 support -- if for no other > reason, the > fact that there's *operating system* level support in OSX for SOCKS5 > redirection. So

Attached is a patch which adds a ConnectionTimeout option, and corrects the ConnectionAttempts documentation. Previously, ssh would try to make a connection ConnectionAttempts times, sleeping 1 second between tries. But each connection attempt could take a very long time to fail if the packets die before the get to the host. So if ssh is being run in a script or what-have-you, it might be

Hello, while openssh-1.2.1pre25 ran out-of-the box, in openssh-1.2.1pre27 the sshd fails to start with: "Host not found" returned... (gcc 2.9.5.2 under AIX 4.3.2 ML 2) Incompatibility of AIX-getaddrinfo ???? Since I'm not familiar with getaddrinfo, I've got no idea what goes wrong ??? Does the code below work on other architectures? Demo-Code built with values taken from a

Here are the new versions of this widely used patch for OpenSSH 3.2.2p1 and 3.2.3p1. The patch avoids waiting to long when using ssh() or scp() on a down host, it is usefull when you have to update many hosts via rsync or rdist themselves relying upon ssh(). It enables a new option 'ConnectTimeout' to control exactly the timeout value, so that it can be used even on slow links. These

Hi, I think we should try other AF for "x11_use_localhost" case. --- openssh-3.1p1/channels.c Tue Mar 5 10:57:45 2002 +++ openssh-3.1p1-fix/channels.c Thu May 2 21:26:28 2002 @@ -2356,6 +2356,13 @@ continue; } } +#ifdef IPV6_V6ONLY + if (ai->ai_family == AF_INET6) { + int on = 1; + if (setsockopt(sock, IPPROTO_IPV6, IPV6_V6ONLY, &on, sizeof(on)) < 0)

If bind() fails we _always_ should close socket. I sent this patch while ago to djm but I still don't see this fix in openssh_cvs. diff -urN openssh-2.3.0p1.org/sshd.c openssh-2.3.0p1/sshd.c --- openssh-2.3.0p1.org/sshd.c Sat Jan 6 19:54:11 2001 +++ openssh-2.3.0p1/sshd.c Sat Jan 6 19:55:48 2001 @@ -782,10 +782,10 @@ debug("Bind to port %s on %s.", strport, ntop); /*

When we make a TCP connection we have to make multiple underlying connect(2) calls, once for each address returned by getaddrinfo. Unfortunately this meant that we lost the errno from any of these calls: $ nbdsh -c 'h.connect_tcp ("localhost", "nbd")' nbd.Error: nbd_connect_tcp: connect: localhost:nbd: could not connect to remote host This commit saves the errno from

In the current implementation, ssh always uses the hostname supplied by the user directly for the SSHFP DNS record lookup. This causes problems when using the domain search path, e.g. I have "search example.com" in my resolv.conf and then do a "ssh host", I will connect to host.example.com, but ssh will query the DNS for an SSHFP record of "host.", not

Hi, We have updated our TIS authserv support patch for OpenSSH 2.5.1p2. You'll find it attached to my message. -- Bazsi PGP info: KeyID 9AF8D0A9 Fingerprint CD27 CFB0 802C 0944 9CFD 804E C82C 8EB1 -------------- next part -------------- diff -urN openssh-2.5.1p2/Makefile.in openssh-2.5.1p2-tis/Makefile.in --- openssh-2.5.1p2/Makefile.in Sun Feb 18 20:13:33 2001 +++

Hi, all -- Here's a small patch that gives better diagnostics on the daemon side if it fails to start up due to inability to create or bind the socket. Presently, it gives a log entry indicating that no sockets could be bound, but crucially does *not* have the system error message (i.e. errno) from the failed call, making it difficult or impossible to determine the problem. I do have a