summaryrefslogtreecommitdiffstats
path: root/src/sbus
Commit message (Collapse)AuthorAgeFilesLines
* SBUS: Fix warning declaration shadows a global declarationLukas Slebodnik2014-05-161-4/+4
| | | | | | | | | src/sbus/sssd_dbus_meta.c: In function 'sbus_meta_find_signal': src/sbus/sssd_dbus_meta.c:43: warning: declaration of 'signal' shadows a global declaration /usr/include/signal.h:101: warning: shadowed declaration is here Reviewed-by: Stef Walter <stefw@redhat.com>
* SBUS: Allow registering paths with fallbackJakub Hrozek2014-05-131-4/+76
| | | | | | | | | | | | Some interfaces we support, such as domain might not be bound to a single path, but rather anything under a path 'directory'. This patch allows the SBUS to register a fallback path that would route any messages under a given anchor to a handler of an interface. The fallback interface is denoted with a trailing wildcard: /org/sssd/anchor/* Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* SBUS: remove unused variablesPavel Březina2014-05-131-4/+0
| | | | Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* SBUS: Refactor sbus_message_handler to retrieve caller IDJakub Hrozek2014-05-131-55/+111
| | | | | | | | | | | | | | | | sbus_message_handler() is refactored so that it only verifies the message matches any of the interfaces SSSD listens on in order to be able to return either of HANDLED or NOT_YET_HANDLED. If the interface is one of those the SSSD should handle, the caller ID is first resolved using sbus_get_sender_id_send and only then passed on to the appropriate special handler. If the sernder ID can't be retrieved, the message is dropped with an error. Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Stef Walter <stefw@redhat.com>
* SBUS: Add an async request to retrieve the caller IDJakub Hrozek2014-05-134-0/+221
| | | | | | | | | | | | | | | | | Adds an async request sbus_get_sender_id_{send,recv} that allows retrieval of UID based on "sender" as returned by dbus_message_get_sender(). The UID is an int64_t to be able to use "-1" to as a fallback value for uknown or error cases. The unit test is added as a standalone one, not part of the sbus_tests because the request, and by extension the unit test relies on being connected to the system bus, which is very unlikely to work in a build system. Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Stef Walter <stefw@redhat.com>
* SBUS: Add SBUS_CONN_TYPE_SYSBUSJakub Hrozek2014-05-132-3/+6
| | | | | | | | | We need to retrieve caller IDs for each call from the system bus. This commit adds a new SBUS connection type that identifies system bus connection. The connection is used in the IFP provider. Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Stef Walter <stefw@redhat.com>
* SBUS: Split out dbus_conn_sendJakub Hrozek2014-05-132-22/+45
| | | | | | | | | | Split dbus_conn_send out of sbus_conn_send to be able to call DBus messages without having a full sbus connection. This function is available to the sbus code only, consumers of sbus (responders and providers) should use sbus_conn_send. Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Stef Walter <stefw@redhat.com>
* SBUS: Add a convenience function sbus_error_newJakub Hrozek2014-05-132-0/+43
| | | | | | | | | Adds a convenience function that constructs a DBusError on top of a talloc context and as such can be used to mark an sbus request as failed without having to create a DBusError instance by the caller. Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Stef Walter <stefw@redhat.com>
* SBUS: Fix error handling conditionJakub Hrozek2014-05-131-1/+2
| | | | | Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Stef Walter <stefw@redhat.com>
* SBUS: several trivial style fixesJakub Hrozek2014-05-133-6/+9
| | | | | | | | | | In SSSD we tend to use {} brackets around single-line blocks, too to make sure we don't forget to add them should the block become larger. We also don't add a space between function name and the opening "(". Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Stef Walter <stefw@redhat.com>
* IFP: Close memstream handle in introspect destructorJakub Hrozek2014-05-061-0/+4
| | | | | | | There was a resource leak in the introspection code. This patch fixes the leak. Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* util: Change file check fns to use a mode maskSimo Sorce2014-04-282-4/+6
| | | | | | | | | | | | | | | Instead of using a custom way to chck file type, use the system provided macros and a mode mask to decide when we want to check. Additionally a mask also allows us to selectively check permissions. Related: https://bugzilla.redhat.com/1089098 Resolves: https://fedorahosted.org/sssd/ticket/2321 Signed-off-by: Simo Sorce <simo@redhat.com> Reviewed-by: Pavel Reichl <preichl@redhat.com>
* sbus request: fix error initializationPavel Březina2014-04-191-2/+4
| | | | | | fixes: sssd_dbus_request.c:28:1: error: missing initializer [-Werror=missing-field-initializers] sssd_dbus_request.c:28:1: error: (near initialization for 'error_internal.dummy1') [-Werror=missing-field-initializers]
* sbus: Add type-safe DBus method handlers and finish functionsStef Walter2014-04-196-33/+261
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Type safe method handlers allow methods not to have to do tedious unwrapping and wrapping of DBus method call messages or replies. Arguments of the following DBus types are supported in type-safe method handlers. In addition arrays of these are supported. y: uint8_t b: bool (but no arrays, yet) n: int16_t q: uint16_t i: int32_t u: uint32_t x: int64_t t: uint64_t d: double s: char * (utf8 string) o: char * (object path) As an exception, arrays of booleans are not supported, but could be added later. Other more complex types could be added later if desired. If a method has other argument types, then it must be marked as having a raw handler (see below). Internally each method can have a type specific invoker function which unpacks the incoming arguments and invokes the method handler with the correct arguments. Each method also has a finish which accepts the type-safe out arguments (ie: return values) and builds the reply message. Like other request 'finish' functions, these free the request talloc context, and are to be used in place of sbus_request_finish() or friends. Raw method handlers parse their own method arguments, and prepare their own reply (ideally using sbus_request_finish() helpers). They can also do strange things like have variable arguments. To mark a DBus method as having a raw method handler use the following annotation: <annotation name="org.freedesktop.sssd.RawHandler" value="true"/> Raw methods do not have invokers or finish functions. I've left all of the internal peer to peer communication using raw method handlers. No code changes here.
* SBUS: Create an sbus_method_meta instance for IntrospectionJakub Hrozek2014-04-193-0/+12
| | | | | Also fixes a warning about uninitialized 'method' as the 'method' variable was unused and not set previously when introspecting.
* SBUS: Generate introspection from the interface meta structureJakub Hrozek2014-04-153-3/+368
| | | | | | | | | | | | https://fedorahosted.org/sssd/ticket/2234 This patch generates the introspection data from the sbus interface meta structure. The generated XML conforms to http://dbus.freedesktop.org/doc/dbus-specification.html#introspection-format The XML description of the interface also always includes the org.freedesktop.DBus.Introspectable interface, which this patch also allows in the policy settings.
* sbus: Add the sbus_request_parse_or_finish() methodStef Walter2014-03-142-0/+155
| | | | | | | | | | | | | | | | | | | | | | Some DBus types returned from dbus_message_get_args() require memory to be released when done. We automatically attach these to the talloc struct sbus_request memory context in this function. This accepts varargs similar to dbus_message_get_args(), which are rather awkward. However instead of reworking them completely, future generated marshalling code will replace most uses of these varargs. If parsing the dbus message fails, then it responds to the DBus caller with an appropriate error such as o.f.D.Error.InvalidArgs. In these cases (ie: when it returns FALSE) the sbus_request is finished. Migrated some, but not all, uses of dbus_message_get_args() to the new function. Some instances have uncommon semantics such as terminating the connection upon failure to parse a message. Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* sbus: Make sbus_new_server() work for non-priveleged processesStef Walter2014-03-141-1/+1
| | | | | | | | | | I'd like to use this during testing. We should check that the socket ownership is the same as the process, rather than limiting it to root. Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* sbus: Refactor how we export DBus interfacesStef Walter2014-03-146-102/+54
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Most importantly, stop using per connection private data. This doesn't scale when you have more than one thing exporting or exported on a connection. Remove struct sbus_interface and expand sbus_conn_add_interface() function. Remove various struct sbus_interface args to connection initialization functions and make callers use sbus_conn_add_interface() directly. The old method was optimized for exporting one interface on a connection. We'll have connections that export zero, one or more interfaces. To export an interface on a DBus server, call sbus_conn_add_interface() from within the sbus_server_conn_init_fn. To export an interface on a DBus client, call sbus_conn_add_interface() after sbus_new_connection() returns. As before struct sbus_interface represents an object exported via DBus. However it is now talloc allocated. One can set instance data on the struct sbus_interface. This instance data is passed to the various handlers and used in their implementation. However, we now have type safe interface exporting in the various high level sss_process_init() sss_monitor_init() and so on. Introspection support was not in use, and is now gone until we implement it using the metadata (future patch). Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* sbus: Add struct sbus_request to represent a DBus invocationStef Walter2014-03-144-21/+221
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | struct sbus_request represents a request from a dbus client being handled by a dbus server implementation. The struct contains the message, connection and method (and in the future teh property) which is being requested. In the future it will contain caller information as well. sbus_request is a talloc memory context, and is a good place to attach any allocations and memory specific to the request. Each handler accepts an sbus_request. If a handler returns EOK, it is assumed that the handler will finish the request. Any of the sbus_request_*finish() methods can be used to complete the request and send back a reply. sbus_request_return_and_finish() uses the same argument varargs syntax as dbus_message_append_args(), which isn't a great syntax. Document it a bit, but don't try to redesign: The marshalling work (will follow this patch set) will remove the need to use varargs for most DBus implementation code. This patch migrates the monitor and data provider dbus code to use sbus_request, but does not try to rework the talloc context's to use it. Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* sbus: Generate constants from interface definitionsStef Walter2014-02-241-0/+21
| | | | | | | | | | | | | | This is not strictly necessary, but avoids duplicating data in mulitple places, and makes the interface definitions benefit dbus callers (a little). After applying this commit you may need to 'make clean' as the codegen has changed. Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Sumit Bose <sbose@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com> Reviewed-by: Simo Sorce <simo@redhat.com>
* sbus: Rework sbus to use interface metadata and vtablesStef Walter2014-02-242-35/+43
| | | | | | | | | | | | | | | | | | | | Previous commits added support for interface metadata and handler vtables. This commit ports sbus_dbus_connection to use them. Port the internal uses of dbus to use the new scheme in a very minimal way. Further cleanup is possible here. This commit provides basic definitions of the internal dbus interfaces. The interfaces aren't fully defined, as the handlers will continue to unpack manually, and often overload DBus methods with different arguments (which is rather unorthodox, but not the end of the world). Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Sumit Bose <sbose@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com> Reviewed-by: Simo Sorce <simo@redhat.com>
* sbus: Add sbus_vtable and update codegen to support itStef Walter2014-02-243-2/+56
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Each interface is a vtable structure derived from sbus_vtable, in the sense that it has an sbus_vtable struct as its first argument. This lets us upcast the interface vtable structure to an sbus_vtable and dispatch to it dynamically and cleanly. The interface metadata contains information about which vtable offset in the interface metadata should be dispatched to for a given function. This is a common scheme, not only among dbus implementations, but also compiled languages. Currently all the vtable functions are of type sbus_msg_handler_fn. These are the handlers we are familiar with and perform raw processing of the message. Later commits will introduce type safe handlers that levelage compile checking and automatic argument packing/unpacking. Although this may seem contrived now, the remainder of the dbus infrastructure work will build on this, including ofd.Properties, ofd.ObjectManager, ofd.Introspect, compiler checked type safe unpacking/packing, etc. The codegen now generates vtable structures for each interface along-side the metadata, and fills in vtable offsets appropriately. It is obviously still possible to hand-craft such vtables and metadata if needed for a special case. Once again examples output can be found at: src/tests/sbus_codegen_tests_generated.h Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Sumit Bose <sbose@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com> Reviewed-by: Simo Sorce <simo@redhat.com>
* sbus: Add meta data structures and code generatorStef Walter2014-02-243-0/+652
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | These metadata structures hold the information about all the details of a DBus interface. They are typically generated from the canonical XML form of the DBus interface, although they may also be hand crafted. Add some handy functions for looking up methods, props, signals, in the metadata of an interface. Currently lookups are just done by looking through an array. If performance becomes an issue (ie: very large interfaces) it would be really easy to sort things and use bsearch(). Later commits will include some definitions using this metadata and related functions. DBus interfaces are defined here: http://dbus.freedesktop.org/doc/dbus-specification.html#introspection-format The introspection data format has become the standard way to represent a DBus interface. For many examples see /usr/share/dbus-1/interfaces/ on a typical linux machine. A word about annotations. These are extra flags or values that can be assigned to anything. So far, the codegen supports this annotation: org.freedesktop.DBus.GLib.CSymbol - An annotation specified in the specification that tells us what C symbol to generate for a given interface or method. By default the codegen will build up a symbol name from the DBus name. It is possible to confuse the code generator into producing invalid C code (with strange method names, for example), but the C compiler catches such silliness right away. Add tests testing basic features of the codegen and poking through the metadata it creates. Also test the metadata lookup functions. Generated code is checked in for easy discovery. An example of the XML interface definitions can be found at: src/tests/sbus_codegen_tests.xml And an example of the generated header can be found here: src/tests/sbus_codegen_tests_generated.h Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Sumit Bose <sbose@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com> Reviewed-by: Simo Sorce <simo@redhat.com>
* Update DEBUG* invocations to use new levelsNikolai Kondrashov2014-02-124-59/+79
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Use a script to update DEBUG* macro invocations, which use literal numbers for levels, to use bitmask macros instead: grep -rl --include '*.[hc]' DEBUG . | while read f; do mv "$f"{,.orig} perl -e 'use strict; use File::Slurp; my @map=qw" SSSDBG_FATAL_FAILURE SSSDBG_CRIT_FAILURE SSSDBG_OP_FAILURE SSSDBG_MINOR_FAILURE SSSDBG_CONF_SETTINGS SSSDBG_FUNC_DATA SSSDBG_TRACE_FUNC SSSDBG_TRACE_LIBS SSSDBG_TRACE_INTERNAL SSSDBG_TRACE_ALL "; my $text=read_file(\*STDIN); my $repl; $text=~s/ ^ ( .* \b (DEBUG|DEBUG_PAM_DATA|DEBUG_GR_MEM) \s* \(\s* )( [0-9] )( \s*, ) ( \s* ) ( .* ) $ / $repl = $1.$map[$3].$4.$5.$6, length($repl) <= 80 ? $repl : $1.$map[$3].$4."\n".(" " x length($1)).$6 /xmge; print $text; ' < "$f.orig" > "$f" rm "$f.orig" done Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Stephen Gallagher <sgallagh@redhat.com> Reviewed-by: Simo Sorce <simo@redhat.com>
* Make DEBUG macro invocations variadicNikolai Kondrashov2014-02-124-82/+82
| | | | | | | | | | | | | | | | | | | | | | | | Use a script to update DEBUG macro invocations to use it as a variadic macro, supplying format string and its arguments directly, instead of wrapping them in parens. This script was used to update the code: grep -rwl --include '*.[hc]' DEBUG . | while read f; do mv "$f"{,.orig} perl -e \ 'use strict; use File::Slurp; my $text=read_file(\*STDIN); $text=~s#(\bDEBUG\s*\([^(]+)\((.*?)\)\s*\)\s*;#$1$2);#gs; print $text;' < "$f.orig" > "$f" rm "$f.orig" done Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Stephen Gallagher <sgallagh@redhat.com> Reviewed-by: Simo Sorce <simo@redhat.com>
* Include ext headers with #include <foo.h> - contPavel Reichl2013-11-041-1/+1
| | | | Changing style of including header files from outside of sssd tree - from "header.h" to <header.h>
* Include external headers with #include <foo.h>Jakub Hrozek2013-10-225-7/+10
| | | | | | I find it more readable to include headers from outside the sssd tree with <foo.h>, not "foo.h". The latter should be used for in-tree headers only.
* Remove unused codeJakub Hrozek2013-09-122-69/+0
|
* Fix pointer formattingLukas Slebodnik2013-09-111-5/+5
|
* Fixing critical format string issues.Lukas Slebodnik2013-05-201-2/+3
| | | | | | --missing arguments. --format '%s', but argument is integer. --wrong format string, examle: '%\n'
* Move some debug lines to new debug log levelsStef Walter2012-06-201-1/+1
| | | | | | | * These are common lines of debug output when starting up sssd https://bugzilla.redhat.com/show_bug.cgi?id=811113
* Catch cases where D-Bus connection is NULLJakub Hrozek2012-04-051-0/+4
| | | | https://fedorahosted.org/sssd/ticket/1270
* SBUS: Fix DEBUG log matchingStephen Gallagher2011-11-101-7/+8
| | | | | | This log message should only be displayed at the most verbose of log levels. Since it didn't match, it was resulting in a Coverity error warning of the printing of an uninitialized value (fd).
* Report on errno, not return code in create_socket_symlinkJakub Hrozek2011-10-141-0/+2
| | | | https://fedorahosted.org/sssd/ticket/1044
* Fix off-by-one error in remove_socket_symlink()Jakub Hrozek2011-10-141-1/+1
| | | | https://fedorahosted.org/sssd/ticket/1043
* Append PID to sbus server socket name, let clients use a symlinkJakub Hrozek2011-10-134-18/+172
| | | | https://fedorahosted.org/sssd/ticket/1034
* Add option to follow symlinks to check_file()Jakub Hrozek2011-10-132-2/+2
|
* New DEBUG facility - conversionPavel Březina2011-08-251-1/+1
| | | | | | | | | | https://fedorahosted.org/sssd/ticket/925 Conversion of the old debug_level format to the new one. (only where it was necessary) Removed: SSS_DEFAULT_DEBUG_LEVEL (completely replaced with SSSDBG_DEFAULT)
* Defer sbus_dispatch() for 30ms during reconnectSumit Bose2010-05-201-1/+2
|
* Properly handle dbus send attempts on a closed connectionStephen Gallagher2010-03-152-0/+81
| | | | | | | | dbus_connection_send_with_reply() will report success and return a NULL pending_reply when the connection is not open for communication. This patch creates a new wrapper around dbus_connection_send_with_reply() to properly detect this condition and report it as an error.
* Add forgotten \n in DEBUG statementsMartin Nagy2010-03-041-2/+2
| | | | | | Logs from confdb with missing '\n' in the DEBUG statements annoyed me so I decided to fix them. I also made a quick grep through the code and found other places so I fixed them too.
* Check and set permissions on SBUS socketsSumit Bose2010-02-232-0/+40
|
* Rename server/ directory to src/Stephen Gallagher2010-02-187-0/+1651
Also update BUILD.txt