To properly shut down charon-cmd with leak reports, only the main thread
should catch SIGUSR1 to shut down the application. Work threads should ignore
SIGUSR1 to avoid any hard application termination.
When doing a rekey for a CHILD_SA, the use counters get reset. An inactivity
job is queued for a time unrelated to the rekey time, so it might happen
that the inactivity job gets executed just after rekeying. If this happens,
inactivity is detected even if we had traffic on the rekeyed CHILD_SA just
before rekeying.
This change implies that inactivity checks can't handle inactivity timeouts
for rekeyed CHILD_SAs, and therefore requires that inactivity timeout is shorter
than the rekey time to have any effect.
When acting on files, we can use fstat() to estimate the buffer size. On
non-file FDs, we dynamically increase an allocated buffer.
Additionally we slightly change the function signature to properly handle
zero-length files and add appropriate unit tests.
The additional 6 bytes are not actually padding but are parsed by the
Cisco client as protocol and src and dst ports (each two bytes but
strangely only the first two in network order).
Don't add a selector to tunnel mode SAs, these might serve multiple
traffic selectors but with only one selector on the SA only the traffic
matching the first one would actually get tunneled.
The kernel implicitly creates an IPIP SA if an IPComp SA is installed.
This SA is used inbound for small packets that are not compressed.
Since the addresses are different (they are the tunnel addresses not
those of the tunneled traffic) additional rules are required if the
traffic selector does not cover the tunnel addresses (e.g. due to a NAT).
For SAs with multiple traffic selectors duplicate rules will get installed.
When changing async callbacks on streams, we have to make sure the watcher
callback is not currently active and has temporarily disabled callbacks. This
could have been the case, as we didn't explicitly removed any pending
watcher registration if both callbacks are NULL.
By enforcing the watcher unregistration, we are sure the watcher callback is
not active and currently is not mangling the callback hooks. This should make
sure we avoid any races for the callback variables.
In certain situations it is desirable to bind the send/receive sockets
for the DHCP address allocation to a particular interface. With this
patch the strongswan.conf option charon.plugins.dhcp.interface can be
used to restrict the DHCP communication to a configurable interface.
Signed-off-by: Thomas Egerer <thomas.egerer@secunet.com>
If a proposal string cannot be matched to a token using strcmp (e.g. if
you want to register a whole class of algorithms containing their ID,
like my_alg_2342), you can use the provided function to register a
parser that transforms the given string into a proposal token.
Signed-off-by: Thomas Egerer <thomas.egerer@secunet.com>
This patch exports the task manager's flush to allow flushing of all
queues with one function call from ike_sa->destroy. It allows the
access of intact children during task destructoin (see git-commit
e44ebdcf) and allows the access of the task manager in
child_state_change hook.
Signed-off-by: Thomas Egerer <thomas.egerer@secunet.com>
The hook data counts remaining buffer bytes, not used ones. Counting them
correctly fixes a crash for long hexdumps.
Further, print_in_hook() must return the number of bytes that would have been
written, not the actually written bytes. This is important, as we allocate a
dynamic buffer in bus that relies on the exact byte count. Fixes long hexdumps
that got truncated.
The two constants overflow time_t on i386 (they also produced a compiler
warning without type suffix) so the comparison with TIME_32_BIT_SIGNED_MAX
did not work as intended.
Fixes#477.
Our SipHash-2-4 implementation returns the result in host order, while
the test vectors are little-endian. Use a custom comparison function to
account for this.
Fixes#478.
The cast to a bool* cut of the actual value on big-endian systems
if bool was shorter than int because the bool argument to printf gets
promoted to an int.
Fixes#479.
On CentOS 6.5 the sys/capability.h header file defines _LINUX_TYPES_H
without actually including that header, preventing its later inclusion
here.
As library.h (via which the capabilities headers are included) is not
actually required in tun_device.[ch], moving the inclusion of tun_device.h
would not strictly be necessary. But it's probably a good idea to
include our own headers after system headers anyway, for if one of the
recursively included files at a later point includes library.h we'd have
the same problem again.
This fixes coverage reports, at least if leak detective is disabled.
If it is enabled the plugins are not unloaded so the destructor is not
executed until the process is destroyed, which seems not to be covered
by gcov.
This way the plugin does not have to be linked explicitly to the test
runner, which otherwise would require that the plugin is either always
enabled to build the tests or that ifdefs are added to the Makefile.
The TKM Diffie-Hellman plugin now maps IANA DH identifiers to TKM DH
algorithm identifiers. The mapping is specified in the daemon's
'dh_mapping' section in the strongswan.conf file:
dh_mapping {
iana_id1 = tkm_id1
iana_id2 = tkm_id2
iana_id3 = tkm_id3
...
}
Only the mapped IANA IDs are registered as supported DH groups.
Due to problems with the external libraries tkm_init/deinit can't be
called for each test case. Because of this leak detective has to be
disabled for these tests.
There are countless other such warnings because e.g. chunk_create() is called
with char*, but at least we prevent users from causing such warnings
inadvertently when using these macros.
Because the write lock was held while calling add_policies() on
child_sa_t, which finishes with a call to child_state_change() on bus_t,
a deadlock would ensue if CHILD_SAs are concurrently being established,
which also causes a call to child_state_change() that will require
the read lock in trap_manager_t.
No locks are now being held while creating the CHILD_SA and installing the
trap policies.
Other users may make use of the noinst libtest.la helper library to implement
unit tests. For libstrongswan, tests.[ch] provide the configuration for test
runner to perform unit tests in a simple manner.
Martin Willi