summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* IPA: Resolve IPA user groups' overrideDN in non-default viewJakub Hrozek2015-01-301-8/+312
| | | | | | | | | | | When the client is in a non-default view, we need to store the override data, in particular the overrideDN as well. Resolves: https://fedorahosted.org/sssd/ticket/2571 Reviewed-by: Sumit Bose <sbose@redhat.com> (cherry picked from commit b2c3722b9a1eaf265f6b102043958f6d4378788c)
* LDAP: Add UUID when saving incomplete groupsJakub Hrozek2015-01-305-12/+29
| | | | | | | | Related to: https://fedorahosted.org/sssd/ticket/2571 Reviewed-by: Sumit Bose <sbose@redhat.com> (cherry picked from commit 108db0e3b9e06e530364ef8228634f5e3f6bd3b5)
* pysss: Fix double freeLukas Slebodnik2015-01-301-3/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The talloc context is removed in destructor. ==1695== Invalid read of size 4 ==1695== at 0x1243D0CD: talloc_chunk_from_ptr (talloc.c:372) ==1695== by 0x1243D0CD: _talloc_free (talloc.c:1559) ==1695== by 0x117B18C3: PySssLocalObject_dealloc (pysss.c:836) ==1695== by 0x117B1AEE: PySssLocalObject_new (pysss.c:898) ==1695== by 0x4ED5522: type_call (typeobject.c:729) ==1695== by 0x4E7F902: PyObject_Call (abstract.c:2529) ==1695== by 0x4F15584: do_call (ceval.c:4328) ==1695== by 0x4F15584: call_function (ceval.c:4133) ==1695== by 0x4F15584: PyEval_EvalFrameEx (ceval.c:2753) ==1695== by 0x4F16BE5: fast_function (ceval.c:4196) ==1695== by 0x4F16BE5: call_function (ceval.c:4131) ==1695== by 0x4F16BE5: PyEval_EvalFrameEx (ceval.c:2753) ==1695== by 0x4F183FF: PyEval_EvalCodeEx (ceval.c:3342) ==1695== by 0x4EA46BC: function_call (funcobject.c:526) ==1695== by 0x4E7F902: PyObject_Call (abstract.c:2529) ==1695== by 0x4F1504F: ext_do_call (ceval.c:4423) ==1695== by 0x4F1504F: PyEval_EvalFrameEx (ceval.c:2792) ==1695== by 0x4F183FF: PyEval_EvalCodeEx (ceval.c:3342) ==1695== Address 0x112d4560 is 64 bytes inside a block of size 96 free'd ==1695== at 0x4C2ACE9: free (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so) ==1695== by 0x1243D2F2: _talloc_free_internal (talloc.c:1057) ==1695== by 0x1243D2F2: _talloc_free (talloc.c:1581) ==1695== by 0x117B1ABF: PySssLocalObject_new (pysss.c:876) ==1695== by 0x4ED5522: type_call (typeobject.c:729) ==1695== by 0x4E7F902: PyObject_Call (abstract.c:2529) ==1695== by 0x4F15584: do_call (ceval.c:4328) ==1695== by 0x4F15584: call_function (ceval.c:4133) ==1695== by 0x4F15584: PyEval_EvalFrameEx (ceval.c:2753) ==1695== by 0x4F16BE5: fast_function (ceval.c:4196) ==1695== by 0x4F16BE5: call_function (ceval.c:4131) ==1695== by 0x4F16BE5: PyEval_EvalFrameEx (ceval.c:2753) ==1695== by 0x4F183FF: PyEval_EvalCodeEx (ceval.c:3342) ==1695== by 0x4EA46BC: function_call (funcobject.c:526) ==1695== by 0x4E7F902: PyObject_Call (abstract.c:2529) ==1695== by 0x4F1504F: ext_do_call (ceval.c:4423) ==1695== by 0x4F1504F: PyEval_EvalFrameEx (ceval.c:2792) ==1695== by 0x4F183FF: PyEval_EvalCodeEx (ceval.c:3342) Reviewed-by: Pavel Reichl <preichl@redhat.com> (cherry picked from commit 3cd7275c3c41a03eb65769c2bf4e472d1de7b8c0)
* PROXY: Fix use after freeLukas Slebodnik2015-01-301-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The dbus_req and associated talloc context are no longer valid after execution of the function sbus_request_return_and_finish even if error code was returned. ==32479== Invalid read of size 8 ==32479== at 0x131F275F: client_registration (proxy_init.c:474) ==32479== by 0x529709E: sbus_request_invoke_or_finish (sssd_dbus_request.c:69) ==32479== by 0x52949B3: sbus_handler_got_caller_id (sssd_dbus_connection.c:555) ==32479== by 0x89B27E3: tevent_common_loop_immediate (tevent_immediate.c:135) ==32479== by 0x89B70CD: epoll_event_loop_once (tevent_epoll.c:907) ==32479== by 0x89B57D6: std_event_loop_once (tevent_standard.c:114) ==32479== by 0x89B1FBC: _tevent_loop_once (tevent.c:530) ==32479== by 0x89B215A: tevent_common_loop_wait (tevent.c:634) ==32479== by 0x89B5776: std_event_loop_wait (tevent_standard.c:140) ==32479== by 0x529E255: server_loop (server.c:668) ==32479== by 0x40DBC5: main (data_provider_be.c:2915) ==32479== Address 0xb700858 is 104 bytes inside a block of size 136 free'd ==32479== at 0x4C2AD17: free (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so) ==32479== by 0x8BBE462: _talloc_free (in /usr/lib64/libtalloc.so.2.1.1) ==32479== by 0x52971A4: sbus_request_finish (sssd_dbus_request.c:95) ==32479== by 0x529731A: sbus_request_return_and_finish (sssd_dbus_request.c:119) ==32479== by 0x131F264D: client_registration (proxy_init.c:443) ==32479== by 0x529709E: sbus_request_invoke_or_finish (sssd_dbus_request.c:69) ==32479== by 0x52949B3: sbus_handler_got_caller_id (sssd_dbus_connection.c:555) ==32479== by 0x89B27E3: tevent_common_loop_immediate (tevent_immediate.c:135) ==32479== by 0x89B70CD: epoll_event_loop_once (tevent_epoll.c:907) ==32479== by 0x89B57D6: std_event_loop_once (tevent_standard.c:114) ==32479== by 0x89B1FBC: _tevent_loop_once (tevent.c:530) ==32479== by 0x89B215A: tevent_common_loop_wait (tevent.c:634) Resolves: https://fedorahosted.org/sssd/ticket/2573 Reviewed-by: Pavel Březina <pbrezina@redhat.com> (cherry picked from commit 33889b2ad764beb6b129f5211b1fab9790da8884)
* SELINUX: Set and reset umask when caling set_seuser from deamon codeJakub Hrozek2015-01-272-1/+21
| | | | | | | https://fedorahosted.org/sssd/ticket/2563 Reviewed-by: Michal Židek <mzidek@redhat.com> (cherry picked from commit 8f78b6442f3176ee43aa06704a3adb9f4ac625d6)
* SELINUX: Call setuid(0)/setgid(0) to also set the real IDs to rootJakub Hrozek2015-01-271-1/+17
| | | | | | | | | | | https://fedorahosted.org/sssd/ticket/2564 libselinux uses many access(2) calls and access() uses the real UID, not the effective UID for the check. Therefore, the setuid selinux_child, which only has effective UID of root would fail the check. Reviewed-by: Michal Židek <mzidek@redhat.com> (cherry picked from commit 486f0d5227a9b81815aaaf7d9a2c39aafcbfdf6a)
* IPA: Use attr's dom for users, tooJakub Hrozek2015-01-261-11/+11
| | | | | | | | | | | | The 'dom' pointer points to domain of the main object being saved. In case of group, dom points to the domain where the group resides. But when saving members, each members might be from a different domain, so we need to find every member's domain based on the attributes. Also don't use Yoda style in conditions. Reviewed-by: Sumit Bose <sbose@redhat.com> (cherry picked from commit b2c5e98def89a0c3d16f5cf7e07ce2020338b540)
* IPA: process_members() add ghosts only onceSumit Bose2015-01-261-2/+4
| | | | | | | | | | | | Since ghost entries might not be properly removed on the IPA server (https://fedorahosted.org/sssd/ticket/2567) chances are that during extdom group lookups a single user is returned multiple time. This patch removes the duplicates before trying to write the data to the cache. Related to https://fedorahosted.org/sssd/ticket/2159 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 60f11e2fa1f63cd40ebace525ad823b0360fac94)
* IPA: resolve IPA group-memberships for AD usersSumit Bose2015-01-262-3/+44
| | | | | | | | | | | | | | | | So far only for initgroups requests the IPA group memberships where resolved for AD users and due to 6fac5e5f0c54a0f92872ce1450606cfcb577a920 those memberships are not overridden by other request. But it turned out that the originalMemberOf attributes related to the IPA group memberships can be overridden by user lookups. Since the originalMemberOf attribute is important in the HBAC evaluation this patch makes sure that the originalMemberOf attribute is not removed but updated during user lookups. Related to https://fedorahosted.org/sssd/ticket/2560 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 63748c69a2c6785d949c82f94749704e0408e5a7)
* AD: support for AD site overridePavel Reichl2015-01-265-7/+37
| | | | | | | | | | Override AD site found during DNS discovery. Resolves: https://fedorahosted.org/sssd/ticket/2486 Reviewed-by: Pavel Březina <pbrezina@redhat.com> (cherry picked from commit e438fbf102c3d787902504bdae177e84230cbbc9)
* AD: add new option ad_sitePavel Reichl2015-01-265-0/+19
| | | | | | | | | | This option overrides a result of the automatic site discovery. Resolves: https://fedorahosted.org/sssd/ticket/2486 Reviewed-by: Pavel Březina <pbrezina@redhat.com> (cherry picked from commit b22e0da9e644f5eb84ee0c8986979fec3fe7eb56)
* sysdb: remove ghosts in all sub-domains as wellSumit Bose2015-01-231-2/+3
| | | | | | | | | | | | | | If a user is a member is a group in a different sub-domain, e.g with universal groups in AD, the ghost attribute might not be properly removed from the group object if the user is resolved. The reason is that only groups from the domain of the user were search for ghost attributes. This patch increases the search-base to all sub-domains of the configured SSSD domain. Resolves https://fedorahosted.org/sssd/ticket/2567 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit fc2146c108e28d50bbf691925cedf9592142dd14)
* nss: fix SID lookupsSumit Bose2015-01-231-1/+9
| | | | | | https://fedorahosted.org/sssd/ticket/2566 Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* IPA: properly handle mixed-case trusted domainsSumit Bose2015-01-234-2/+133
| | | | | | | | | | | | In the SSSD cache domain names are handled case-sensitive. As a result fully-qualified names in RDN contain the domain part in the original spelling. When IPA client lookup up group-memberships on the IPA server via the extdom plugin the names returned are all lower case. To make sure new DNs are generated correctly the domain part must adjusted. Related to https://fedorahosted.org/sssd/ticket/2159 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com>
* views: fix GID overrride for mpg domainsSumit Bose2015-01-211-1/+20
| | | | | | | | | | | | When adding a user sysdb internally adds a value to SYSDB_GIDNUM for mpg domain which might cause conflicts with the one we added to users git GID overrides. With this patch the override GID is added after the user is created but in the same transaction Releted to https://fedorahosted.org/sssd/ticket/2514 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit ba818cc39dfe94c2b8613f4badf7912811f0f737)
* Open the PAC socket from krb5_child before dropping rootJakub Hrozek2015-01-213-0/+27
| | | | | | | | | | | The PAC responder by default allows only connections from the root user. This patch opens the socket to the PAC responder before the krb5_child drops privileges so the connection seemingly comes from root. https://fedorahosted.org/sssd/ticket/2559 Reviewed-by: Sumit Bose <sbose@redhat.com> (cherry picked from commit 858e750c3d4fe54e50616a1ed1e101469503c070)
* nss: Add original DN and memberOf to origbyname requestSumit Bose2015-01-202-2/+5
| | | | | | | | | | IPA HBAC evaluation relies on the original values for DN and memberOf attributes. Resolves https://fedorahosted.org/sssd/ticket/2560 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 7543052f562f157f7b17fdc46a6777d80c0cb3bd)
* nss: refactor fill_orig()Sumit Bose2015-01-201-67/+66
| | | | | | | The two loops in fill_orig were almost identical. Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit a4d64002b5ca763622bde240d27797d361ba0388)
* nss: make fill_orig() multi-value awareSumit Bose2015-01-202-21/+198
| | | | | Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 5f4d896ec8e06476f4282b562b1044de14c48ecf)
* krb5: fix entry order in MEMORY keytabSumit Bose2015-01-192-28/+172
| | | | | | | | | | | | | | | Since krb5_kt_add_entry() adds new entries at the beginning of a MEMORY type keytab and not at the end a simple copy into a MEMORY type keytab will revert the order of the keytab entries. Since e.g. the sssd_krb5 man page give hints about where to add entries into keytab files to help SSSD to find a right entry we have to keep the order when coping a keytab into a MEMORY type keytab. This patch fixes this by doing a second copy to retain the original order. Resolves https://fedorahosted.org/sssd/ticket/2557 Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com> Reviewed-by: Jakub Hrozek <jhrozek@redhat.com>
* MAN: add dots as valid character in domain namesPavel Reichl2015-01-151-1/+1
| | | | | | | | | | Add dots into a set of allowed characters for domain names. Resolves: https://fedorahosted.org/sssd/ticket/2527 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 9a15eb105d01d9e100e69e9d66fb8e880b228246)
* MAN: dyndns_iface supports only one interfacePavel Reichl2015-01-152-0/+6
| | | | | | | | Resolves: https://fedorahosted.org/sssd/ticket/2548 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 702176303382b5a385e90fe68ad2c32bd708ebf1)
* GPO: add systemd-user to gpo default permit listPavel Reichl2015-01-152-1/+8
| | | | | | | | Resolves: https://fedorahosted.org/sssd/ticket/2556 Reviewed-by: Stephen Gallagher <sgallagh@redhat.com> (cherry picked from commit b49c6abe12721ee8442be1c1bd6c15443b518ca2)
* krb5_child: Return ERR_NETWORK_IO on KRB5_KDCREP_SKEWJakub Hrozek2015-01-141-0/+1
| | | | | | | Previously, we were only handling KRB5KRB_AP_ERR_SKEW Reviewed-by: Sumit Bose <sbose@redhat.com> (cherry picked from commit 9b2cd4e5e451c07cb2f04cdbaea2b94ccb5fb2ee)
* IPA: set SYSDB_INITGR_EXPIRE for RESP_USER_GROUPLISTSumit Bose2015-01-141-0/+14
| | | | | | | | Since RESP_USER_GROUPLIST contains all group memberships it is effectively an initgroups request hence SYSDB_INITGR_EXPIRE will be set. Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 62d919aea98edd1095f6a22241903d4c045b46ed)
* IPA: resolve missing membersSumit Bose2015-01-141-14/+48
| | | | | Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 3cd287313d93e29f9754feb46017dba2a039affd)
* IPA: rename ipa_s2n_get_groups_send() to ipa_s2n_get_fqlist_send()Sumit Bose2015-01-141-50/+50
| | | | | Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit f1f22df95996390f63266ebacb624e521d934592)
* IPA: process_members() optionally return missing members listSumit Bose2015-01-141-18/+67
| | | | | Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 942ebb62c8df766a22271103abd518ddae02ea3a)
* IPA: add missing breakSumit Bose2015-01-141-0/+1
| | | | | | | | The current request already returned the SID, we do not need to request it separately. Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit e6046d23b3e90102fb3c796737ced03fb5a60fea)
* IPA: make version check more preciseSumit Bose2015-01-141-2/+2
| | | | | | | | | | The call protected by the check does not only expect the version 1 of the extdom plugin is used but a specific response type as well. Since version 1 can return older response types as well we want to be on the safe side. Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 2fc12875f7d51248799016c19c1298b85e06a286)
* IPA: do not look up overrides on client with default viewSumit Bose2015-01-141-14/+49
| | | | | | | | | The IPA extdom plugin returns the data with the default view already applied hence it is on needed to look up the override data if the client has the default view assigned. Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit d8ceb194023a2cdc8bc183acc322e9a7fb6fe2b1)
* IPA: ipa_resolve_user_list_send() take care of overridesSumit Bose2015-01-131-10/+8
| | | | | | | | Currently ipa_resolve_user_list_send() only looks up the related user objects but do not check for overrides. This patch tries to fix this. Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit eab17959df71341073f946c533f59fc5e593b35c)
* sysdb: fix group members with overridden namesSumit Bose2015-01-132-11/+63
| | | | | Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit fbcdc08722aa8ed17c4b114e01fbb37c02cfb2fe)
* IPA: resolve ghost members if a non-default view is appliedSumit Bose2015-01-132-0/+213
| | | | | | | Related to https://fedorahosted.org/sssd/ticket/2481 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 765d9075bb1e10ae0f09b6c2701bfd50aeb423d4)
* IPA: add get_be_acct_req_for_user_name()Sumit Bose2015-01-132-0/+18
| | | | | | | Related to https://fedorahosted.org/sssd/ticket/2481 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit d32b165fad7b89462f49c82349e1df5a2343afa2)
* GPO: Extract server hostname after connectingJakub Hrozek2015-01-131-29/+29
| | | | | | | | | | | | https://fedorahosted.org/sssd/ticket/2543 The LDAP URI is not valid prior to connecting to LDAP. Moreover, reconnecting to a different server might invalidate the URI. Move reading the URI after the connection has been established. Reviewed-by: Sumit Bose <sbose@redhat.com> (cherry picked from commit ccff8e75940963a0f68f86efcddc37133318abfa)
* GPO: Don't use stdout for output in gpo_childJakub Hrozek2015-01-133-6/+12
| | | | | | | | | | | Resolves: https://fedorahosted.org/sssd/ticket/2544 Use a dedicated fd instead to work around https://bugzilla.samba.org/show_bug.cgi?id=11036 Reviewed-by: Sumit Bose <sbose@redhat.com> (cherry picked from commit f00a61b6079d8de81432077a59daf015d85800d2)
* UTIL: Allow dup-ing child pipe to a different FDJakub Hrozek2015-01-137-22/+39
| | | | | | | | | | | Related to: https://fedorahosted.org/sssd/ticket/2544 Adds a new function exec_child_ex and moves setting the extra_argv[] to exec_child_ex() along with specifying the input and output fds. Reviewed-by: Sumit Bose <sbose@redhat.com> (cherry picked from commit 16cb0969f0a9ea71524d852077d6a480740d4f12)
* GPO: Set libsmb debugging to stderrJakub Hrozek2015-01-131-0/+1
| | | | | | | | | | | | | libsmb logs to stdout by default. It's much more reasonable to log to stderr by default. Please also note: https://bugzilla.samba.org/show_bug.cgi?id=11036 and: https://fedorahosted.org/sssd/ticket/2544 Reviewed-by: Sumit Bose <sbose@redhat.com> (cherry picked from commit bb7ddd2be9847bfb07395341c7623da1b104b8a6)
* logrotate: Fix warning file size changed while zippingLukas Slebodnik2015-01-091-0/+1
| | | | | | | | | | | | | Postpone compression of the previous log file to the next rotation cycle. This only has effect when used in combination with compress. We need to use it because we cannot tell sssd to close log files and thus sssd processes might continue writing to the previous log file for some time. Resolves: https://fedorahosted.org/sssd/ticket/2547 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> (cherry picked from commit 565eb6fa43e74e2fbfff00dc29fdb20c5544a3d2)
* spec: sifp requires sssd-dbusPavel Březina2015-01-091-0/+1
| | | | | | | | Resolves: https://fedorahosted.org/sssd/ticket/2550 Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com> (cherry picked from commit ce6ba48c5a0723d9c8db6d960d2dfbcb6ffdd673)
* TESTS: Cover sysdb_gpo.c with unit testsJakub Hrozek2015-01-082-0/+80
| | | | | | | Untested code is risky to change. Reviewed-by: Pavel Reichl <preichl@redhat.com> (cherry picked from commit ee8dccf5f0a7de4aba16ab73a53872df9a65175c)
* GPO: Ignore ENOENT result from sysdb_gpo_get_gpo_result_setting()Jakub Hrozek2015-01-081-1/+4
| | | | | | | | | | | https://fedorahosted.org/sssd/ticket/2542 If the GPO result object was missing completely, we would error out with a fatal error code. It's more user-friendly to treat the missing object as if the requested attribute was missing on the provider level. Reviewed-by: Pavel Reichl <preichl@redhat.com> (cherry picked from commit fc2cc91a5b645180e53d46436b0d08011aac8d74)
* Updating the version to the 1.12.4 releaseJakub Hrozek2015-01-081-1/+1
|
* Updating translations for the 1.12.3 releasesssd-1_12_3Jakub Hrozek2015-01-0838-24796/+26484
|
* Add zanata.xml file for integration with Zanata command line clientCarlos A. Munoz2015-01-081-0/+106
| | | | | | | To set up and use the Zanata client, follow: http://zanata.org/help/cli/cli-configuration/ Reviewed-by: Jakub Hrozek <jhrozek@redhat.com>
* TESTS: typo in 'assert message'Pavel Reichl2015-01-081-2/+2
| | | | Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* monitor: Service restart fixesStephen Gallagher2015-01-071-20/+74
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are actually two bugs here: 1) When either the kill(SIGTERM) or kill(SIGKILL) commands returned failure (for any reason), we would talloc_free(svc) which removed it from being eligible for restart, resulting in the service never starting again without an SSSD service restart. 2) There is a fairly wide race condition where it's possible for a SIGKILL timer to "catch up" to the child exit handler between us noticing the termination and actually restarting it. The race happens because we re-enter the mainloop and add a restart timeout to avoid a quick failure if we keep restarting due to a transitory issue (the mt_svc object, and therefore the SIGKILL timer, were never freed until we got to the actual service restart). We can minimize this race by recording the timer_event for the SIGKILL timeout in the mt_svc object. This way, if the process exits via SIGTERM, we will immediately remove the timer for the SIGKILL. Additionally, we'll catch the special-case of an ESRCH response from the kill(SIGKILL) and assume that it means that the process has exited. The only other two possible errors are * EINVAL: (an invalid signal was specified) - This should be impossible, obviously. * EPERM: This process doesn't have permission to send signals to this PID. If this happens, it's either an SELinux bug or else the process has terminated and a new process that SSSD doesn't control has taken the ID over. So in the incredibly unlikely case that one of those occurs, we'll just go ahead and try to start a new process. This patch also removes the incorrect talloc_free(svc) calls on the kill() failures and replaces them with an attempt to just start up the service again and hope for the best. Resolves: https://fedorahosted.org/sssd/ticket/2525 Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* Add test for sysdb_store_overrideSumit Bose2015-01-071-0/+70
| | | | Reviewed-by: Jakub Hrozek <jhrozek@redhat.com>
* CONFDB: Typo in debug messagePavel Reichl2015-01-051-1/+1
| | | | Reviewed-by: Sumit Bose <sbose@redhat.com>
generated by cgit (git 2.34.1) at 2025-09-04 05:05:52 +0000