| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
These functions were used when reconnecting to the DP after losing
the connection. Since there is no DP any longer, there's no reason
to have these functions.
|
|
The way we were processing errors from the provider caused offline
authentication to stop working. Previously the problem was masked
by a bug in the data provider that always returned "Success" for any
operation no matter what the actual return code was. when DP got
removed the bug became evident.
|
|
Turn the backend process into data provider servers
Make Frontends (pam, nss) directly attach to the backends
|
|
We have converted to using dhash in place of btreemap everywhere
in the code.
|
|
This should fix #218
It should also prevent us from leaking memory in case the original request times
out and should prevent races with the callbacks beeing freed after sdp_req is
freed and thus dereferencing freed memory in the callbacks detructors.
|
|
Loop control variable was not being incremented.
I also converted a goto loop into a do...while loop to make it
easier to follow the logic.
|
|
Introduces a new option --debug-to-files which makes SSSD output its
debug information to a file instead of stderr, which is still the
default.
Also introduces a new confdb option debug_to_files which does the same,
but can be specified per-service in the config file.
The logfiles are stored in /var/log/sssd by default.
Changes the initscript to log to files by default.
|
|
This converts a great many configuration options to the new
standard format.
|
|
This reverts commit 8c50bd085c0efe5fde354deee2c8118887aae29d.
Amended: commit 1016af2b1b97ad4290ccce8fa462cc7e3c191b2e also made
use of the SYSLOG_ERROR() macro, so those portions of that code
also needed to be reverted.
|
|
This is just a band-aid until ELAPI is fully functional and ready to
use.
|
|
|
|
- the client sends the PID as uint32_t and sssd will use uint32_t too
- fix a possible type issue where a uint32_t is sent as int32 in internal
dbus communication
|
|
The patch that added check_cache() broke them, no results returned for any group
with actual members ...
|
|
Fixes: #138
|
|
This timeout specifies the lifetime of a cache entry before it is
updated out-of-band. When this timeout is hit, the request will
still complete from cache, but the SSSD will also go and update
the cached entry in the background to extend the life of the
cache entry and reduce the wait time of a future request.
|
|
getpwnam, getpwuid, getgrnam and getgrgid will now use a common
function, check_cache, for determining whether to return a cached
value or to go to the provider.
|
|
The special persistent local database retains the original name.
All other backends now have their own cache-NAME.ldb file.
|
|
Make counter for used messages explicit.
|
|
|
|
Also remove legacy memberuid support
|
|
- with the boolean option filterUsersInGroups it can be controlled
wether filtered users appear in groups or not.
- fixed an error which prevented the display of groups with filtered
members
- removed some tab indents
|
|
|
|
Remove redundant reconnection code that was interfeering with the sbus
reconnection code.
Consolidate include files for sbus relates operations.
Make pamsrv code similar to nsssrv code.
|
|
Many of the functions in responder_dp.c were originally NSS-
specific and were moved there from the NSS responder code. Since
they are now generic to any responder, rename them to sss_dp_*
|
|
Currently, if an additional request comes in for a cache entry
while that same entry is already in the process of being refreshed,
we start a duplicate cache update request. This patch adds allows
the cache to maintain a hash table of all in-progress requests and
queue up multiple callbacks for updates in progress. Once the data
is returned, all of these callbacks will fire.
|
|
D-BUS handles timeouts itself and reports DBUS_ERROR_NO_REPLY if
a timeout fires, so we can rely on this instead of having an
explicit timeout ourselves. Furthermore, the two timeouts present
a potential race condition.
|
|
Mirrors what we have done with the monitor.
|
|
Let services identify themselves voiluntarily as the first operation
instead of polling from the monitor.
Also consolidate some common functions and make them available as monitor
helpers.
|
|
The child processes call prctl() and when their parent process is
killed, they are sent SIGTERM using prctl. This is currently
Linux-specific, for non-Linuxes, a similar effect is achieved by
catching a set of common termination signals and sending SIGTERM to the
process group.
|
|
Previously, we had hardcoded the paths for the NSS, PAM and
private PAM sockets to /var/lib/sss/pipes. With this patch, we
will specify the sockets with --with-pipe-path.
|
|
Make as much as possible static, and remove use of talloc_reference and
allocation/deallocation of memory when not necessary.
Fix also responder use of rctx->conn, was mistakenly used for both
monitor and dp connections.
|
|
This reduce code duplication as it allows to use one set of watch and timeout
functions, and at the same time also allow not to use a secondary structure just
to unify these functions.
|
|
Rationalize and rename connection names in preparatoin for merging of server and
connection structures.
|
|
Simplify code by removing stuff that is never used or redundant.
|
|
|
|
Fix incorrect error code return in local_handler_callback
|
|
|
|
|
|
This is part of a set of patches to rewrite sysdb to a hopefully better
API, that will also let use use tevent_req async style calls to manipulate
our cache.
|
|
This sysdb_req has always really been a transaction handle and not
a request.
This is part of a set of patches to rewrite transaction support in sysdb to a
hopefully better API, that will also let use use tevent_req async style to
manipulate our cache.
|
|
|
|
|
|
|
|
The previous patch to fix an enumeration bug found with group enumeration
inadvertently introduced a bug with user enumeration.
Yeah, almost funny!
|
|
|
|
If a backend had all its results filtered in fill_pwent or fill_grent
then we would return an empty result, which means "end of results" to
the client.
Now we return ENOENT and let callers decide what to do.
Also make sure we do not grow packets unless we are going to fill them
as that's a recipe for killing the client as the size passed to
sss_packet_grow is used to determine the size of the final packet.
|
|
The timeout of the data provider call (in ms) got overwritten by a cache
timeout (in s).
|
|
|
|
|
|
Convert auth modules to do the caching themselves
|
