This wasn't working at all, since we didn't do HMAC but just a plain hash.
Also, verification of packets failed because it was checking the whole packet,
not the packet minus the HMAC.
We clear the cached address used for UDP connections when a node becomes
unreachable. This also prevents host-up scripts from passing the old, cached
address from when the host becomes reachable again from a different address.
Before it would check all addresses, and not learn an address if another node
already claimed that address. This caused fast roaming to fail, the code from
commit 6f6f426b35 was never triggered.
The control socket code was completely different from how meta connections are
handled, resulting in lots of extra code to handle requests. Also, not every
operating system has UNIX sockets, so we have to resort to another type of
sockets or pipes for those anyway. To reduce code duplication and make control
sockets work the same on all platforms, we now just connect to the TCP port
where tincd is already listening on.
To authenticate, the program that wants to control a running tinc daemon must
send the contents of a cookie file. The cookie is a random 256 bits number that
is regenerated every time tincd starts. The cookie file should only be readable
by the same user that can start a tincd.
Instead of the binary-ish protocol previously used, we now use an ASCII
protocol similar to that of the meta connections, but this can still change.
Since event.h is not part of tinc, we include it in have.h were all other
system header files are included. We also ensure -levent comes before -lgdi32
when compiling with MinGW, apparently it doesn't work when the order is
reversed.
UNIX domain sockets, of course, don't exist on Windows. For now, when compiling
tinc in a MinGW environment, try to use a TCP socket bound to localhost as an
alternative.
In switch mode, if a known MAC address is claimed by a second node before it
expired at the first node, it is likely that this is because a computer has
roamed from the LAN of the first node to that of the second node. To ensure
packets for that computer are routed to the second node, the first node should
delete its corresponding Subnet as soon as possible, without waiting for the
normal expiry timeout.
If MTU probing discovered a node was not reachable via UDP, packets for it were
forwarded to the next hop, but always via TCP, even if the next hop was
reachable via UDP. This is now fixed by retrying to send the packet using
send_packet() if the destination is not the same as the nexthop.
Options should have a fixed width anyway, but this also fixes a possible MinGW
compiler bug where %lx tries to print a 64 bit value, even though a long int is
only 32 bits.
We now handle MAC Subnets in exactly the same way as IPv4 and IPv6 Subnets.
This also fixes a problem that causes unncessary broadcasting of unicast
packets in VPNs where some daemons run 1.0.10 and some run other versions.
When the HUP signal is sent while some outgoing connections have not been made
yet, or are being retried, a NULL pointer could be dereferenced resulting in
tinc crashing. We fix this by more careful handling of outgoing_ts, and by
deleting all connections that have not been fully activated yet at the HUP
signal is received.
This device works like /dev/tun on Linux, automatically creating a new tap
interface when a program opens it. We now pass the actual name of the newly
created interface in $INTERFACE.
This keeps NAT mappings for UDP alive, and will also detect when a node is not
reachable via UDP anymore or if the path MTU is decreasing. Tinc will fall back
to TCP if the node has become unreachable.
If UDP communication is impossible, we stop sending probes, but we retry if it
changes its keys.
We also decouple the UDP and TCP ping mechanisms completely, to ensure tinc
properly detects failure of either method.
Although it would be better to have the new defaults, only the most recent
releases of most of the platforms supported by tinc come with a version of
OpenSSL that supports SHA256. To ensure people can compile tinc and that nodes
can interact with each other, we revert the default back to Blowfish and SHA1.
This reverts commit 4bb3793e38.
Git's log and blame tools were used to find out which files had significant
contributions from authors who sent in patches that were applied before we used
git.
This feature is not necessary anymore since we have tools like valgrind today
that can catch stack overflow errors before they make a backtrace in gdb
impossible.
- Update year numbers in copyright headers.
- Add copyright information for Michael Tokarev and Florian Forster to the
copyright headers of files to which they have contributed significantly.
- Mention Michael and Florian in AUTHORS.
- Mention that tinc is GPLv3 or later if compiled with the --enable-tunemu
flag.
During the path MTU discovery phase, we might not know the maximum MTU yet, but
we do know a safe minimum. If we encounter a packet that is larger than that
the minimum, we now send it via TCP instead to ensure it arrives. We also
allow large packets that we cannot fragment or create ICMP replies for to be
sent via TCP.
The TAP-Win32 device is not a socket, and select() under Windows only works
with sockets. Tinc used a separate thread to read from the TAP-Win32 device,
and passed this via a local socket to the main thread which could then select()
from it. We now use a global mutex, which is only unlocked when the main thread
is waiting for select(), to allow the TAP reader thread to process packets
directly.
In light of the recent improvements of attacks on SHA1, the default hash
algorithm in tinc is now SHA256. At the same time, the default symmetric
encryption algorithm has been changed to AES256.
We used both rand() and random() in our code. Since it returns an int, we have
to use %x in our format strings instead of %lx. This fixes a crash under
Windows when cross-compiling tinc with a recent version of MinGW.
If two pointers do not belong to the same array, pointer subtraction gives
nonsensical results, depending on the level of optimisation and the
architecture one is compiling for. It is apparently not just subtracting the
pointer values and dividing by the size of the object, but uses some kind of
higher magic not intended for mere mortals. GCC will not warn about this at
all. Casting to void * is also a no-no, because then GCC does warn that strict
aliasing rules are being broken. The only safe way to query the ordering of two
pointers is to use the (in)equality operators.
The unsafe implementation of connection_compare() has probably caused the "old
connection_t for ... still lingering" messages. Our implementation of AVL trees
is augmented with a doubly linked list, which is normally what is traversed.
Only when deleting an old connection the tree itself is traversed.
If PMTUDiscovery is enabled, and we see a unicast packet that is larger than
the path MTU in switch mode, treat it just like we would do in router mode.
PMTUDiscovery was disabled in commit d5b56bbba5
because tinc did not handle packets larger than the path MTU in switch and hub
modes. We now allow it again in preparation of proper support, but default to
off.
Commit 5674bba5c5 introduced weighted Subnets,
but the weight was included in the SUBNET variable passed to subnet-up/down
scripts. This makes it harder to use in those scripts. The weight is now
stripped from the SUBNET variable and put in the WEIGHT variabel.
Grzegorz Dymarek noted that tinc segfaults at the stat() call in
execute_script() on the iPhone. We can omit the stat() call for the moment,
the subsequent call to system() will fail with just a warning.
This is a slightly modified patch from Grzegorz Dymarek that allows tinc to use
the tunemu device, which allows tinc to be compiled for iPhones and recent
iPods. To enable support for tunemu, the --enable-tunemu option has to be used
when running the configure script.
Valgrind caught tinc reading free'd memory during a purge(). This was caused by
first removing it from the main node tree, which will already call free_node(),
and then removing it from the UDP tree. This might cause spurious segmentation
faults.
Although we select() before we call recvfrom(), it sometimes happens that
select() tells us we can read but a subsequent read fails anyway. This is
harmless.
If there is an outstanding MTU probe event for a node which is not reachable
anymore, a UDP packet would be sent to that node, which caused a key request to
be sent to that node, which triggered a NULL pointer dereference. Probes and
other UDP packets to unreachable nodes are now dropped.
When chrooted, we either need to force-initialize resolver
and/or nsswitch somehow (no clean way) or resolve all the
names we want before entering chroot jail. The latter
looks cleaner, easier and it is actually safe because
we still don't talk with the remote nodes there, only
initiating outgoing connections.
This option can be set to low, normal or high. On UNIX flavours, this changes
the nice value of the process by +10, 0 and -10 respectively. On Windows, it
sets the priority to BELOW_NORMAL_PRIORITY_CLASS, NORMAL_PRIORITY_CLASS and
HIGH_PRIORITY_CLASS respectively.
A high priority might help to reduce latency and packet loss on the VPN.
If a host has multiple addresses on an interface, the source address of the TCP
connection(s) was picked by the operating system while the UDP packets used a
bound socket, i. e. the source address was the address specified by the user.
This caused problems because the receiving code requires the TCP connection and
the UDP connection to originate from the same IP address.
This patch adds support for the `BindToInterface' and `BindToAddress' options
to the setup of outgoing TCP connections.
Tested with Debian Etch on x86 and Debian Lenny on x86_64.
Signed-off-by: Florian Forster <octo@verplant.org>
If running without `--net', the (global) variable `netname' is NULL. This
creates a segmentation fault because this NULL-pointer is passed to strdup:
Program terminated with signal 11, Segmentation fault.
#0 0xb7d30463 in strlen () from /lib/tls/i686/cmov/libc.so.6
(gdb) bt
#0 0xb7d30463 in strlen () from /lib/tls/i686/cmov/libc.so.6
#1 0xb7d30175 in strdup () from /lib/tls/i686/cmov/libc.so.6
#2 0x0805bf47 in xstrdup (s=0x0) at xmalloc.c:118 <---
#3 0x0805be33 in setup_device () at device.c:66
#4 0x0805072e in setup_myself () at net_setup.c:432
#5 0x08050db2 in setup_network () at net_setup.c:536
#6 0x0805b27f in main (argc=Cannot access memory at address 0x0) at tincd.c:580
This patch fixes this by checking `netname' in `setup_device'. An alternative
would be to check for NULL-pointers in `xstrdup' and return NULL in this case.
Signed-off-by: Florian Forster <octo@verplant.org>
Add some logging about refused ADD_SUBNET
(it causes subsequent client disconnect so it's
important to know which subnet was at fault).
Maybe we should just ignore it completely.
First of all, the idea behind the TunnelServer option is to hide all other
nodes from each other, so we shouldn't forward broadcast packets from them
anyway. The other reason is that since edges from other nodes are ignored, the
calculated minimum spanning tree might not be correct, which can result in
routing loops.
Since compression can either grow or shrink a packet, the size of an MTU probe
after decompression might not reflect the real path MTU. Now we use the size
before decompression, which is independent of the compression algorithm, and
substract a safety margin such that the calculated path MTU will be safe even
for packets which grow as much as possible after compression.
Instead of a single, global decryption context, each node has its own context.
However, in send_ans_key(), the global context was initialised. This commit
fixes that and removes the global context completely.
Also only set status.validkey after all checks have been evaluated.
