Tobias Brunner [Wed, 25 Nov 2015 16:05:56 +0000 (17:05 +0100)]
android: Switch to AppCompat/Material theme and use custom Toolbar as AppBar
Also includes some whitespace/formatting changes due to the switch to
Android Studio.
Tobias Brunner [Fri, 15 Apr 2016 15:43:58 +0000 (17:43 +0200)]
android: Ignore build/ in project directory
Tobias Brunner [Wed, 25 Nov 2015 15:07:55 +0000 (16:07 +0100)]
android: Update platform tools and pull in support libs
We'll have to change some stuff that Google deprecated (e.g. the tabs in
the ActionBar) and that requires changing the theme at least in activities.
Since that would look a bit inconsistent we'll change it globally and
use parts of the support library.
Tobias Brunner [Tue, 12 Apr 2016 15:10:11 +0000 (17:10 +0200)]
android: Update Android Gradle plugin and wrapper
Andreas Steffen [Tue, 26 Apr 2016 15:15:25 +0000 (17:15 +0200)]
testing: Use absolute path of imv_policy_manager
Andreas Steffen [Tue, 26 Apr 2016 15:13:49 +0000 (17:13 +0200)]
Updated products in IMV database
Andreas Steffen [Tue, 26 Apr 2016 15:13:20 +0000 (17:13 +0200)]
swanctl: list EAP type in --list-conns
Andreas Steffen [Mon, 25 Apr 2016 17:31:12 +0000 (19:31 +0200)]
testing: -D and -u options in sfdisk are not supported any more
Yannick Cann [Mon, 25 Apr 2016 08:39:41 +0000 (10:39 +0200)]
identification: Add support for dmdName RDN (2.5.4.54)
It's listed in RFC 2256 but was later removed with RFC 4519, but there
are still some certs that use it.
Closes strongswan/strongswan#43.
Andreas Steffen [Sun, 24 Apr 2016 21:34:44 +0000 (23:34 +0200)]
leak-detective: added _IO_file_doallocate to whitelist
Andreas Steffen [Sun, 24 Apr 2016 21:33:23 +0000 (23:33 +0200)]
swanctl: log errors to stderr
Andreas Steffen [Sun, 24 Apr 2016 11:36:22 +0000 (13:36 +0200)]
testing: updated testing.conf
Tobias Brunner [Mon, 18 Apr 2016 08:07:15 +0000 (10:07 +0200)]
pool: Use correct name to remove index for CHILD_SA TS in SQLite script
Fixes #1415.
Tobias Brunner [Wed, 6 Apr 2016 13:17:06 +0000 (15:17 +0200)]
kernel-pfkey: Add support for manual priorities
Also orders policies with equals priorities by their automatic priority.
Tobias Brunner [Mon, 11 Apr 2016 09:19:26 +0000 (11:19 +0200)]
kernel-pfkey: Update priority calculation formula to the new one in kernel-netlink
Since the selectors are not exactly the same (no port masks, no interface)
some small tweaks have been applied.
Tobias Brunner [Wed, 6 Apr 2016 12:40:28 +0000 (14:40 +0200)]
kernel-netlink: Order policies with equal priorities by their automatic priority
This allows using manual priorities for traps, which have a lower
base priority than the resulting IPsec policies. This could otherwise
be problematic if, for example, swanctl --install/uninstall is used while
an SA is established combined with e.g. IPComp, where the trap policy does
not look the same as the IPsec policy (which is now otherwise often the case
as the reqids stay the same).
It also orders policies by selector size if manual priorities are configured
and narrowing occurs.
Tobias Brunner [Fri, 15 Apr 2016 08:33:23 +0000 (10:33 +0200)]
Merge branch 'boringssl'
Adds some fixes to the openssl plugin to build against BoringSSL.
Fixes #1374.
Tobias Brunner [Mon, 11 Apr 2016 09:33:41 +0000 (11:33 +0200)]
curl: Add TLS support if libcurl is built against BoringSSL
We don't have to rely on the openssl plugin and its threading
initialization as BoringSSL is thread-safe out of the box.
Tobias Brunner [Fri, 8 Apr 2016 13:56:25 +0000 (15:56 +0200)]
openssl: BoringSSL does not support configuration
The other initialization functions are still defined but many are
apparently no-ops (this is also true for the threading initialization).
Tobias Brunner [Fri, 8 Apr 2016 13:55:05 +0000 (15:55 +0200)]
openssl: The member storing the DH exponent length has been renamed in BoringSSL
Tobias Brunner [Fri, 8 Apr 2016 13:54:08 +0000 (15:54 +0200)]
openssl: Use proper EVP macro to determine size of a hash
Tobias Brunner [Wed, 13 Apr 2016 07:31:50 +0000 (09:31 +0200)]
android: Remove OPENSSL_NO_EC* defines
Current versions of OpenSSL/BoringSSL shipped with Android support ECC.
Tobias Brunner [Fri, 8 Apr 2016 13:52:58 +0000 (15:52 +0200)]
android: OPENSSL_NO_ENGINE is now properly defined in the headers
Tobias Brunner [Fri, 8 Apr 2016 12:37:21 +0000 (14:37 +0200)]
curl: Handle LibreSSL like OpenSSL in regards to multi-threading
LibreSSL is API compatible so our openssl plugin does not need any
changes and it works fine with the curl plugin.
Tobias Brunner [Fri, 8 Apr 2016 12:35:47 +0000 (14:35 +0200)]
configure: Replace two remaining usages of AC_HAVE_LIBRARY with AC_CHECK_LIB
Tobias Brunner [Wed, 13 Apr 2016 09:58:32 +0000 (11:58 +0200)]
thread: Don't hold mutex when calling cleanup handlers while terminating
This could interfere with cleanup handlers that try to acquire
mutexes while other threads holding these try to e.g. cancel the threads.
As cleanup handlers are only queued by the threads themselves we don't need
any synchronization to access the list.
Fixes #1401.
Andreas Steffen [Tue, 12 Apr 2016 16:50:58 +0000 (18:50 +0200)]
testing: Added swanctl/rw-multi-ciphers-ikev1 scenario
Tobias Brunner [Mon, 11 Apr 2016 14:11:16 +0000 (16:11 +0200)]
Ignore Qt Creator project files
Closes strongswan/strongswan#32.
Andreas Steffen [Mon, 11 Apr 2016 08:24:12 +0000 (10:24 +0200)]
Version bump to 5.4.1dr1
Andreas Steffen [Mon, 11 Apr 2016 08:19:21 +0000 (10:19 +0200)]
Merge branch 'kernel-policies'
Andreas Steffen [Sat, 9 Apr 2016 14:46:37 +0000 (16:46 +0200)]
Extended IPsec kernel policy scheme
The kernel policy now considers src and dst port masks as well as
restictions to a given network interface. The base priority is
100'000 for passthrough shunts, 200'000 for IPsec policies,
300'000 for IPsec policy traps and 400'000 for fallback drop shunts.
The values 1..30'000 can be used for manually set priorities.
Andreas Steffen [Wed, 6 Apr 2016 14:08:38 +0000 (16:08 +0200)]
testing: Added swanctl/manual_prio scenario
Andreas Steffen [Mon, 28 Mar 2016 07:03:21 +0000 (09:03 +0200)]
Include manual policy priorities and restriction to interfaces in vici list-conn command
Andreas Steffen [Sun, 27 Mar 2016 08:18:19 +0000 (10:18 +0200)]
Implemented IPsec policies restricted to given network interface
Andreas Steffen [Thu, 24 Mar 2016 17:35:27 +0000 (18:35 +0100)]
Support manually-set IPsec policy priorities
Tobias Brunner [Mon, 4 Apr 2016 16:41:17 +0000 (18:41 +0200)]
peer-cfg: Use struct to pass data to constructor
Tobias Brunner [Mon, 4 Apr 2016 14:09:23 +0000 (16:09 +0200)]
child-cfg: Use struct to pass data to constructor
Tobias Brunner [Mon, 4 Apr 2016 13:06:44 +0000 (15:06 +0200)]
kernel-pfkey: Prefer policies with reqid over those without
Tobias Brunner [Mon, 4 Apr 2016 13:06:11 +0000 (15:06 +0200)]
kernel-pfkey: Only install templates for regular IPsec policies with reqid
Tobias Brunner [Fri, 1 Apr 2016 17:05:02 +0000 (19:05 +0200)]
testing: Add swanctl/net2net-gw scenario
Tobias Brunner [Mon, 4 Apr 2016 08:49:35 +0000 (10:49 +0200)]
shunt-manager: Install "outbound" FWD policy
If there is a default drop policy forwarded traffic might otherwise not
be allowed by a specific passthrough policy (while local traffic is
allowed).
Tobias Brunner [Fri, 1 Apr 2016 15:06:10 +0000 (17:06 +0200)]
kernel-netlink: Prefer policies with reqid over those without
This allows two CHILD_SAs with reversed subnets to install two FWD
policies each. Since the outbound policy won't have a reqid set we will
end up with the two inbound FWD policies installed in the kernel, with
the correct templates to allow decrypted traffic.
Tobias Brunner [Fri, 1 Apr 2016 14:51:51 +0000 (16:51 +0200)]
kernel-netlink: Only associate templates with inbound FWD policies
We can't set a template on the outbound FWD policy (or we'd have to make
it optional). Because if the traffic does not come from another (matching)
IPsec tunnel it would get dropped due to the template mismatch.
Tobias Brunner [Fri, 1 Apr 2016 14:41:05 +0000 (16:41 +0200)]
child-sa: Install "outbound" FWD policy
If there is a DROP shunt that matches outbound forwarded traffic it
would get dropped as the FWD policy we install only matches decrypted
inbound traffic. That's because the Linux kernel first checks the FWD
policies before looking up the OUT policy and SA to encrypt the packets.
Tobias Brunner [Fri, 1 Apr 2016 14:26:07 +0000 (16:26 +0200)]
kernel-netlink: Associate routes with IN policies instead of FWD policies
This allows us to install more than one FWD policy. We already do this
in the kernel-pfkey plugin (there the original reason was that not all
kernels support FWD policies).
Tobias Brunner [Thu, 31 Mar 2016 14:01:40 +0000 (16:01 +0200)]
kernel: Use structs to pass information to the kernel-ipsec interface
Tobias Brunner [Wed, 6 Apr 2016 11:56:36 +0000 (13:56 +0200)]
testing: List conntrack table on sun in ikev2/host2host-transport-connmark scenario
Tobias Brunner [Wed, 6 Apr 2016 09:17:27 +0000 (11:17 +0200)]
testing: Version bump to 5.4.0
References #1382.
Tobias Brunner [Tue, 24 Nov 2015 14:28:11 +0000 (15:28 +0100)]
testing: Disable leak detective when generating CRLs
GnuTLS, which can get loaded by the curl plugin, does not properly cleanup
some allocated memory when deinitializing. This causes invalid frees if
leak detective is active. Other invalid frees are related to time
conversions (tzset).
References #1382.
Tobias Brunner [Mon, 4 Apr 2016 13:35:43 +0000 (15:35 +0200)]
pkcs11: Skip zero-padding of r and s when preparing EC signature
They are zero padded to fill the buffer.
Fixes #1377.
Tobias Brunner [Mon, 4 Apr 2016 13:29:16 +0000 (15:29 +0200)]
chunk: Skip all leading zero bytes in chunk_skip_zero() not just the first
Tobias Brunner [Fri, 1 Apr 2016 17:03:22 +0000 (19:03 +0200)]
string: Gracefully handle NULL in str*eq() macros
Tobias Brunner [Thu, 31 Mar 2016 17:47:31 +0000 (19:47 +0200)]
byteorder: Explicitly check for htoXeXX macros
Some platforms have XetohXX macros instead of XeXXtoh macros, in which
case we'd redefine the htoXeXX macros.
Cameron McCord [Wed, 30 Mar 2016 11:01:37 +0000 (05:01 -0600)]
vici: Fix documentation of some dictionary keys of two request messages
Closes strongswan/strongswan#40.
Tobias Brunner [Wed, 30 Mar 2016 11:48:51 +0000 (13:48 +0200)]
proposal: Use standard integer types for static keywords
Tobias Brunner [Wed, 30 Mar 2016 07:12:09 +0000 (09:12 +0200)]
utils: Remove nonsensical typedefs for standard uint types
Andreas Steffen [Thu, 24 Mar 2016 19:58:32 +0000 (20:58 +0100)]
Use u_int32_t legacy type in blowfish header file
Andreas Steffen [Tue, 22 Mar 2016 12:22:01 +0000 (13:22 +0100)]
Use standard unsigned integer types
Shota Fukumori [Sat, 12 Mar 2016 15:05:01 +0000 (00:05 +0900)]
updown: Get value for PLUTO_MARK_{IN,OUT} from CHILD_SA
Or the invoked script will get a broken value when `mark=%unique` is
used in a configuration.
Closes strongswan/strongswan#37.
Tobias Brunner [Wed, 23 Mar 2016 13:40:29 +0000 (14:40 +0100)]
connmark: Explicitly include xt_mark.h for older kernels
Fixes #1365.
Tobias Brunner [Tue, 26 Jan 2016 16:26:04 +0000 (17:26 +0100)]
android: Enable 64-bit ABIs
Tobias Brunner [Tue, 26 Jan 2016 16:23:02 +0000 (17:23 +0100)]
android: Enable build against API level 21
While building against this level in general would break our app on
older systems, the NDK will automatically use this level for 64-bit
ABI builds (which are not supported in older levels). So to build
against 64-bit ABIs we have to support this API level.
Tobias Brunner [Wed, 23 Mar 2016 13:16:32 +0000 (14:16 +0100)]
libcharon: Add missing header file to Android.mk
Not really relevant, just to make sure both file lists are the same.
Tobias Brunner [Wed, 23 Mar 2016 13:13:07 +0000 (14:13 +0100)]
testing: Updated updown scripts in libipsec scenarios to latest version
Tobias Brunner [Tue, 22 Mar 2016 13:22:19 +0000 (14:22 +0100)]
ike-sa-manager: Avoid memory leak if IKE_SAs get checked in after flush() was called
A thread might check out a new IKE_SA via checkout_new() or
checkout_by_config() and start initiating it while the daemon is
terminating and the IKE_SA manager is flushed by the main thread.
That SA is not tracked yet so the main thread is not waiting for it and
the other thread is able to check it in and creating an entry after flush()
already terminated causing a memory leak.
Fixes #1348.
Thomas Egerer [Mon, 21 Mar 2016 13:46:11 +0000 (14:46 +0100)]
ha: Delete cache entry inside the locked mutex
Signed-off-by: Thomas Egerer <thomas.egerer@secunet.com>
Tobias Brunner [Tue, 22 Mar 2016 17:06:22 +0000 (18:06 +0100)]
swanctl: Fix documented directory name for remote pubkeys
Andreas Steffen [Tue, 22 Mar 2016 10:20:36 +0000 (11:20 +0100)]
Version bump to 5.4.0
Tobias Brunner [Fri, 11 Mar 2016 17:39:35 +0000 (18:39 +0100)]
kernel-netlink: Fix lookup of next hops for destinations with prefix
References #1347.
Tobias Brunner [Fri, 11 Mar 2016 13:40:55 +0000 (14:40 +0100)]
imc-os: Terminate buffer after fread(3) call to make Coverity happy
Tobias Brunner [Fri, 11 Mar 2016 13:22:19 +0000 (14:22 +0100)]
imc-os: Correctly check return value of ftell(2)
Tobias Brunner [Fri, 11 Mar 2016 10:25:25 +0000 (11:25 +0100)]
Fix some Doxygen issues
Andreas Steffen [Fri, 11 Mar 2016 10:31:02 +0000 (11:31 +0100)]
Updated NEWS
Tobias Brunner [Fri, 11 Mar 2016 09:24:36 +0000 (10:24 +0100)]
man: Updated default proposals in ipsec.conf(5)
Tobias Brunner [Fri, 11 Mar 2016 09:06:58 +0000 (10:06 +0100)]
identification: Make `written` signed to fix error checking when printing ranges
Tobias Brunner [Wed, 10 Feb 2016 11:09:37 +0000 (12:09 +0100)]
vici: Don't hold write lock while running or undoing start actions
Running or undoing start actions might require enumerating IKE_SAs,
which in turn might have to enumerate peer configs concurrently, which
requires acquiring a read lock. So if we keep holding the write lock while
enumerating the SAs we provoke a deadlock.
By preventing other threads from acquiring the write lock while handling
actions, and thus preventing the modification of the configs, we largely
maintain the current synchronous behavior. This way we also don't need to
acquire additional refs for config objects as they won't get modified/removed.
Fixes #1185.
Andreas Steffen [Fri, 11 Mar 2016 07:29:15 +0000 (08:29 +0100)]
Initialize ts variable
Tobias Brunner [Wed, 9 Mar 2016 11:10:12 +0000 (12:10 +0100)]
forecast: Compare the complete rules when deleting them
Same as the change in the connmark plugin.
References #1229.
Tobias Brunner [Mon, 7 Mar 2016 15:52:43 +0000 (16:52 +0100)]
connmark: Don't restore CONNMARK for packets that already have a mark set
This allows e.g. modified versions of xl2tpd to set the mark in
situations where two clients are using the same source port behind the
same NAT, which CONNMARK can't restore properly as only one conntrack entry
will exist with the mark set to that of the client that sent the last packet.
Fixes #1230.
Tobias Brunner [Mon, 7 Mar 2016 14:32:02 +0000 (15:32 +0100)]
connmark: Compare the complete rules when deleting them
By settings a matchmask that covers the complete rule we ensure that the
correct rule is deleted (i.e. matches and targets with potentially different
marks are also compared).
Since data after the passed pointer is actually dereferenced when
comparing we definitely have to pass an array that is at least as long as
the ipt_entry.
Fixes #1229.
Andreas Steffen [Thu, 10 Mar 2016 14:26:03 +0000 (15:26 +0100)]
Merge branch 'subnet-identities'
Implemented IKEv1 IPv4/IPv6 address subnet and range identities to
be used as owners for shared secrets.
swanctl supports configuration of traffic selectors with IPv4/IPv6
address ranges.
Andreas Steffen [Thu, 10 Mar 2016 11:00:56 +0000 (12:00 +0100)]
Support of IP address ranges in traffic selectors
Andreas Steffen [Thu, 10 Mar 2016 07:02:44 +0000 (08:02 +0100)]
Updated swanctl/rw-psk-ikev1 scenario
Andreas Steffen [Tue, 8 Mar 2016 21:27:30 +0000 (22:27 +0100)]
Implemented IPv4/IPv6 subnet and range identities
The IKEv1 IPV4_ADDR_SUBNET, IPV6_ADDR_SUBNET, IPV4_ADDR_RANGE and
IPV6_ADDR_RANGE identities have been fully implemented and can be
used as owners of shared secrets (PSKs).
Tobias Brunner [Thu, 10 Mar 2016 11:00:56 +0000 (12:00 +0100)]
Merge branch 'p-cscf'
This adds the p-cscf plugin that can request P-CSCF server addresses from
an ePDG via IKEv2 (RFC 7651). Addresses of the same families as requested
virtual IPs are requested if enabled in strongswan.conf for a particular
connection. The plugin currently writes received addresses to the log.
Tobias Brunner [Thu, 11 Feb 2016 14:47:32 +0000 (15:47 +0100)]
attr: Only enumerate attributes matching the IKE version of the current IKE_SA
Numerically configured attributes are currently sent for both versions.
Tobias Brunner [Fri, 5 Feb 2016 15:47:26 +0000 (16:47 +0100)]
attr: Add p-cscf keyword for P-CSCF server addresses
Tobias Brunner [Thu, 4 Feb 2016 17:41:14 +0000 (18:41 +0100)]
p-cscf: Make sending requests configurable and disable it by default
Tobias Brunner [Thu, 4 Feb 2016 17:35:07 +0000 (18:35 +0100)]
p-cscf: Only send requests if virtual IPs of the same family are requested
Tobias Brunner [Thu, 4 Feb 2016 17:26:29 +0000 (18:26 +0100)]
p-cscf: Add attribute handler for P-CSCF server addresses
Tobias Brunner [Wed, 3 Feb 2016 16:49:48 +0000 (17:49 +0100)]
p-cscf: Add plugin stub
Tobias Brunner [Wed, 3 Feb 2016 16:34:34 +0000 (17:34 +0100)]
payloads: Verify P-CSCF configuration attributes like others carrying IP addresses
Tobias Brunner [Wed, 3 Feb 2016 16:32:16 +0000 (17:32 +0100)]
attributes: Define P-CSCF address attributes described in RFC 7651
Tobias Brunner [Thu, 10 Mar 2016 10:48:12 +0000 (11:48 +0100)]
Merge branch 'mbb-reauth-online-revocation'
With these changes initiators of make-before-break reauthentications
suspend online revocation checks until after the new IKE_SA and all
CHILD_SAs are established. See
f1cbacc5d1be for details why that's
necessary.
Tobias Brunner [Thu, 10 Mar 2016 10:46:44 +0000 (11:46 +0100)]
NEWS: Added note on online revocation checks during make-before-break reauthentication
Tobias Brunner [Tue, 27 Oct 2015 16:42:45 +0000 (17:42 +0100)]
testing: Add ikev2/reauth-mbb-revoked scenario
Tobias Brunner [Tue, 27 Oct 2015 16:42:15 +0000 (17:42 +0100)]
testing: Generate a CRL that has moon's actual certificate revoked
Tobias Brunner [Wed, 28 Oct 2015 15:09:59 +0000 (16:09 +0100)]
ike-sa: Improve interaction between flush_auth_cfg and delayed revocation checks
Tobias Brunner [Tue, 27 Oct 2015 16:34:50 +0000 (17:34 +0100)]
ikev2: Delay online revocation checks during make-before-break reauthentication
We do these checks after the SA is fully established.
When establishing an SA the responder is always able to install the
CHILD_SA created with the IKE_SA before the initiator can do so.
During make-before-break reauthentication this could cause traffic sent
by the responder to get dropped if the installation of the SA on the
initiator is delayed e.g. by OCSP/CRL checks.
In particular, if the OCSP/CRL URIs are reachable via IPsec tunnel (e.g.
with rightsubnet=0.0.0.0/0) the initiator is unable to reach them during
make-before-break reauthentication as it wouldn't be able to decrypt the
response that the responder sends using the new CHILD_SA.
By delaying the revocation checks until the make-before-break
reauthentication is completed we avoid the problems described above.
Since this only affects reauthentication, not the original IKE_SA, and the
delay until the checks are performed is usually not that long this
doesn't impose much of a reduction in the overall security.
Tobias Brunner [Tue, 27 Oct 2015 16:31:43 +0000 (17:31 +0100)]
ikev2: Add task that verifies a peer's certificate
On failure the SA is deleted and reestablished as configured. The task
is activated after the REAUTH_COMPLETE task so a make-before-break reauth
is completed before the new SA might get torn down.