summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* TESTS: Add a test for parallel execution of klistreviewJakub Hrozek2017-05-232-1/+27
| | | | | | | | | Integration test for: https://pagure.io/SSSD/sssd/issue/3372 With https://pagure.io/SSSD/sssd/issue/3372 still broken, the unit test wold fail because one of the concurrent klist commands would trigger a race condition in the KCM queue code, crashing the KCM responder.
* KCM: Fix the per-client serialization queueJakub Hrozek2017-05-231-60/+122
| | | | | | | | | | | | | | | Resolves: https://pagure.io/SSSD/sssd/issue/3372 Fixes a race condition between one client request adding an operation to the hash table value, which was previously a linked list of operations, while another concurrent operation would remove the last remaining linked list element through its callback. Instead, the hash table value is now a separate 'queue head' structure which is only changed in a tevent request to make sure is is not processes concurrently with adding to the queue (which is also a tevent request).
* INTG: Do not use configure time option enable-files-domainLukas Slebodnik2017-05-232-1/+2
| | | | | | | | The implicit_files was started with each test even though was not required. Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* UTIL: Drop unused error code ERR_MISSING_CONFLukas Slebodnik2017-05-233-8/+0
| | | | | Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* CONFDB: Use default configuration with missing sssd.confLukas Slebodnik2017-05-232-6/+4
| | | | | | | | Resolves: https://pagure.io/SSSD/sssd/issue/3339 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* pam: properly support UPN logon namesSumit Bose2017-05-232-3/+80
| | | | | | | | | | | | | | | | Many logon applications like /bin/login or sshd canonicalize the user name before they call pam_start() and hence the UPN is not seen by SSSD's pam responder. But some like e.g. gdm don't and authentication might fail if a UPN is used. The reason is that currently the already parsed short name of the user was used in the cache_req and hence the cache_req was not able to fall back to the UPN lookup code. This patch uses the name originally provided by the user as input to allow the fallback to the UPN lookup. Resolves https://pagure.io/SSSD/sssd/issue/3240 Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* test: make sure p11_child is build for pam-srv-testsSumit Bose2017-05-231-0/+3
| | | | Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* cache_req: use the right negative cache for initgroups by upnSumit Bose2017-05-231-2/+2
| | | | Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* KRB5: Advise the user to inspect the krb5_child.log if the child doesn't ↵Jakub Hrozek2017-05-231-0/+3
| | | | | | | | | | | | | return a valid response If the child returns a runtime error, it is often not clear from the domain debug logs what to do next. This patch adds a DEBUG message that tells the admin to look into the krb5_child.log Resolves: https://pagure.io/SSSD/sssd/issue/2955 Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* AD SUBDOMAINS: Fix search bases for child domainsMichal Židek2017-05-192-0/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When using direct AD integration, child domains did not respect the sssd.conf configuration of search bases. There were few issues all of which are fixed in this small patch. First problem was that the sdap domain list was not properly inherited from the parent in the child domains and the children always created their own sdap domains lists that were disconnected from the parent context and never used. Second issue was that the child domain did not call the function to reinit the search bases after the sdap_domain was added to the list of sdap domains. This caused that child domains always used automatically detected search bases and never used the configured ones even though they were properly read into the ID options context attached to the subdomain. Also there has been an issue that the sdap search bases were rewritten by the new child domain initialization (this only happened with more than one child domain) because the sdap domain list was 'updated' every time a new child domain was initialized, which caused that only the main domain and the last child domain had proper search bases, the others only the auto-discovered ones (because they were overwritten with the 'update'). Resolves: https://pagure.io/SSSD/sssd/issue/3397 Reviewed-by: Sumit Bose <sbose@redhat.com>
* AD: Add debug messagesMichal Židek2017-05-191-0/+6
| | | | | | Add debug messages when 1way or 2way trusts are created. Reviewed-by: Sumit Bose <sbose@redhat.com>
* CACHE_REQ: Ensure the domains are updated for "filter" related callsFabiano Fidêncio2017-05-171-5/+81
| | | | | | | | | | | | | | | | | | | | | | As contacting the infopipe responder on a "filter" related call may lead to the situation where the cr_domains' list is not populated yet (as the domains and subdomains lists from the data provider are not processed yet), let's explicitly call sss_dp_get_domains() for those cases and avoid returning a wrong result to the caller. This situation may happen only because the schedule_get_domains_task(), that's called when the infopipe responder is initialized, may take some time to run/finish. While I'm not exactly sure whether it's the best solution to avoid the "race", it seems to be sane enough to avoid the issues. Resolves: https://pagure.io/SSSD/sssd/issue/3387 Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com> Reviewed-by: Sumit Bose <sbose@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* CACHE_REQ: Avoid using of uninitialized valueFabiano Fidêncio2017-05-151-6/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | Commit 4ef0b19a introduced the following warning, as "req" may be used without being initialized: src/responder/common/cache_req/cache_req_search.c: In function 'cache_req_search_done': src/responder/common/cache_req/cache_req_search.c:467:9: error: 'req' may be used uninitialized in this function [-Werror=maybe-uninitialized] tevent_req_error(req, ret); ^ src/responder/common/cache_req/cache_req_search.c:424:24: note: 'req' was declared here struct tevent_req *req; ^ cc1: all warnings being treated as errors In order to fix the issue above, let's just allocate tmp_ctx after "req" is already set. Related: https://pagure.io/SSSD/sssd/issue/3362 Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com> Co-Author: Lukáš Slebodník <lslebodn@redhat.com> Reviewed-by: Sumit Bose <sbose@redhat.com>
* sss_nss_getlistbycert: return results from multiple domainsSumit Bose2017-05-104-12/+192
| | | | | | | | | | Currently only the results from one domain were returned although all domains were searched and the results were available. Unit tests are updated to cover this case as well. Resolves https://pagure.io/SSSD/sssd/issue/3393 Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* test_ldap.py: Add test for filter_{users,groups}Lukas Slebodnik2017-05-101-0/+96
| | | | | | | | | Related: https://pagure.io/SSSD/sssd/issue/3362 Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* CACHE_REQ: Make use of cache_req_ncache_filter_fn()Fabiano Fidêncio2017-05-108-39/+165
| | | | | | | | | | | | | | | | | | | | | This patch makes use of cache_req_ncache_filter_fn() in order to process the result of a cache_req search and then filter out all the results that are present in the negative cache. The "post cache_req search" result processing is done basically in two different cases: - plugins which don't use name as an input token (group_by_id, user_by_id and object_by_id), but still can be affected by filter_{users,groups} options; - plugins responsible for groups and users enumeration (enum_groups and enum_users); Resolves: https://pagure.io/SSSD/sssd/issue/3362 Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* CACHE_REQ_RESULT: Introduce cache_req_create_ldb_result_from_msg_list()Fabiano Fidêncio2017-05-102-0/+40
| | | | | | | | | | | | | | | Similarly to what cache_req_create_ldb_result_from_msg() does this new function creates a new ldb_result from a list of ldb_message. It's going to be used in the follow-up patch where some messages from ldb_result may be filtered and then a new ldb_result has to be created. Related: https://pagure.io/SSSD/sssd/issue/3362 Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* CACHE_REQ: Add a new cache_req_ncache_filter_fn() plugin functionFabiano Fidêncio2017-05-1021-0/+33
| | | | | | | | | | | | | | | | | This function will be responsible for filtering out all the results that we have that are also present in the negative cache. This is useful mainly for plugins which don't use name as an input token but can still be affected by filter_{users,groups} options. For now this new function is not being used anywhere. Related: https://pagure.io/SSSD/sssd/issue/3362 Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* NSS: Move output name formatting to utilsNikolai Kondrashov2017-05-107-49/+87
| | | | | | | Move NSS nss_get_name_from_msg and the core of sized_output_name to the utils to make them available to provider and other responders. Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* DP: Reduce Data Provider log level noiseJustin Stephenson2017-05-093-4/+13
| | | | | | | | | | | | | | | Certain operations are not supported with certain providers causing informational Data Provider log messages to be logged as errors or failures. This patch lowers the log level to reduce overall log noise and ensure only critical log messages are logged when a low debug_level value is used. Resolves: https://pagure.io/SSSD/sssd/issue/3287 https://pagure.io/SSSD/sssd/issue/3278 Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com>
* MAN: AD Provider GSSAPI clarificationJustin Stephenson2017-05-091-1/+4
| | | | | | | | | | Explicitly state that the AD provider uses Kerberos and GSSAPI for encrypting traffic to avoid attempted custom configurations with SSL/TLS Resolves: https://pagure.io/SSSD/sssd/issue/3377 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com>
* KCM: include missing header fileLukas Slebodnik2017-05-091-0/+1
| | | | | | | | | | | | | | | | | | man 2 readv says that the header file "sys/uio.h" must be included for the functions readv/writev Previously, "sys/uio.h" was included in "sys/socket.h" in glibc. It worked just by a change. But it will be changed in glibc-2.26. https://sourceware.org/bugzilla/show_bug.cgi?id=21426 src/responder/kcm/kcmsrv_cmd.c: In function 'kcm_iovec_op': src/responder/kcm/kcmsrv_cmd.c:75:15: error: implicit declaration of function 'readv'; did you mean 'read'? [-Werror=implicit-function-declaration] src/responder/kcm/kcmsrv_cmd.c:77:15: error: implicit declaration of function 'writev'; did you mean 'write'? [-Werror=implicit-function-declaration] Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SERVER_MODE: Update sdap lists for each ad_ctxMichal Židek2017-05-041-0/+36
| | | | | | | | | | | | | | | | | We use separate AD context for each subdomain in the server mode. Every such context has it's own sdap_domain list witch represents sdap options such as filter and search bases for every domain. However AD context can only fully initialize sdap_domain structure for the same domain for which the whole context was created, which resulted in the other sdap_domain structures to be have automaticily detected settings. This can cause problems if user is member of groups from multiple domains. Resolves: https://pagure.io/SSSD/sssd/issue/3381 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com>
* PAM: check matching certificates from all domainsSumit Bose2017-05-041-6/+63
| | | | | | | | | | Although the cache_req lookup found matching in multiple domains only the results from the first domain were used. With this patch the results from all domains are checked. Resolves https://pagure.io/SSSD/sssd/issue/3385 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com>
* LDAP/AD: Do not fail in case rfc2307bis_nested_groups_recv() returns ENOENTFabiano Fidêncio2017-05-031-1/+7
| | | | | | | | | | | | | | | Commit 25699846 introduced a regression seen when an initgroup lookup is done and there's no nested groups involved. In this scenario the whole lookup fails due to an ENOENT returned by rfc2307bis_nested_groups_recv(), which leads to the user removal from sysdb causing some authentication issues. Resolves: https://pagure.io/SSSD/sssd/issue/3331 Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com> Reviewed-by: Sumit Bose <sbose@redhat.com>
* SPEC: Use %license macroLukas Slebodnik2017-05-031-18/+24
| | | | | | | | | | | | Starting with rpm 4.11, it is possible to install the license using a new file macro %license, this will separate the license files from documents and install them in a special directory in /usr/share rpm -q -l -p ./sssd-1.15.3-0.el7.x86_64.rpm /usr/share/licenses/sssd-1.15.3 /usr/share/licenses/sssd-1.15.3/COPYING Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Use macro python_provide conditionallyLukas Slebodnik2017-05-031-10/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The rpm macro python_provide is defined only in fedora and epel. This is the reason why we have fallback definition in the beginning of spec file otherwise build on rhel would fail. This macro is defined in file /usr/lib/rpm/macros.d/macros.python provided by package python-rpm-macros. sh$ rpm -qf /usr/lib/rpm/macros.d/macros.python python-rpm-macros-3-20.fc26.noarch sh$ grep python_provide /usr/lib/rpm/macros.d/macros.python %python_provide() %{lua: print("%python_provide: ERROR: ") But this package is not installed in minimal chroot and therefore build dependencies cannot be extracted from spec file. sh$ mock --clean --shell 'rpm -q python-rpm-macros' 2>/dev/null package python-rpm-macros is not installed sh$ mock --shell 'rpm --eval "%{python_provide python-test}"' 2>/dev/null %{python_provide python-test} sh$ mock --resultdir . --rebuild sssd-1.15.3-0.fc26.src.rpm ... error: line 295: Unknown tag: %{python_provide python2-sssdconfig} ... This is the reason why it has to be used conditionally in fedora as it is shown in example common spec file in python fedora packaging guidelines http://fedoraproject.org/wiki/Packaging:Python#Example_common_spec_file sh$ rpm -q --whatrequires python-rpm-macros python2-devel-2.7.13-5.fc26.x86_64 python3-devel-3.6.0-22.fc26.x86_64 This patch reduce differences between upstream and fedora spec file. Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Call ldconfig in libsss_certmap scriptletsLukas Slebodnik2017-05-031-0/+4
| | | | | | We do it for other libraries. Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Move kcm scriptlets to systemd sectionLukas Slebodnik2017-05-031-12/+12
| | | | Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Move files provider files within packageLukas Slebodnik2017-05-031-4/+3
| | | | | | | | | It's a cosmetic change to group similar files together (e.g. man pages). The same order is in fedora downstream spec file. It simplifies comparison of changes between spec files. Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Use correct package for translated sssd-kcm man pagesLukas Slebodnik2017-05-031-2/+5
| | | | | | | Resolves: https://pagure.io/SSSD/sssd/issue/3327 Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Use correct package for translated sss_certmap man pagesLukas Slebodnik2017-05-031-4/+9
| | | | | | | | | This patch also moved sss_certmap.5 from sssd-common to libsss_certmap Resolves: https://pagure.io/SSSD/sssd/issue/3327 Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Use correct package for translated idmap_sss man pagesLukas Slebodnik2017-05-031-3/+6
| | | | | | | Resolves: https://pagure.io/SSSD/sssd/issue/3327 Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Use correct package for translated sssctl man pagesLukas Slebodnik2017-05-031-0/+3
| | | | | | | Resolves: https://pagure.io/SSSD/sssd/issue/3327 Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Use correct package for translated sss_ssh* man pagesLukas Slebodnik2017-05-031-0/+3
| | | | | | | Resolves: https://pagure.io/SSSD/sssd/issue/3327 Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Move man page for sss_rpcidmapd to the right packageLukas Slebodnik2017-05-031-4/+7
| | | | | | | | | Patch also fixes location of translated manual pages Resolves: https://pagure.io/SSSD/sssd/issue/3327 Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Use correct package for translated sssd-ifp man pageLukas Slebodnik2017-05-031-3/+6
| | | | | | | Resolves: https://pagure.io/SSSD/sssd/issue/3327 Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Add missing scriptlets for package sssd-dbusLukas Slebodnik2017-05-031-0/+9
| | | | Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Move systemd service sssd-ifp.service to right packageLukas Slebodnik2017-05-031-1/+3
| | | | | | | The sssd-ifp.service was installed even though sssd_ifp was not installed on systemd. Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* SPEC: Update processing of translation in %installLukas Slebodnik2017-05-031-4/+22
| | | | Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com>
* AD: Make ad_account_can_shortcut() reusable by SSSD on an IPA serverJakub Hrozek2017-05-021-78/+84
| | | | | | | | | | | | | | | | | | | | | Resolves: https://pagure.io/SSSD/sssd/issue/3318 The ad_account_can_shortcut() function is helpful to avoid unnecessary searches when SSSD is configured with an Active Directory domain that uses ID-mapping in the sense that if we find that an ID is outside our range, we can just abort the search in this domain and carry on. This function was only used in the AD provider functions which are used when SSSD is enrolled direcly with an AD server. This patch moves the function to a codepath that is shared between directly enrolled SSSD and SSSD running on an IPA server. Apart from moving the code, there are some minor changes to the function signature, namely the domain is passed as as struct (previously the domain name from the DP input was passed). Reviewed-by: Michal Židek <mzidek@redhat.com>
* SDAP: Fix handling of search basesMichal Židek2017-05-024-16/+30
| | | | | | | | | | | | | We were rewriting the sdap_domain's search bases for only the first sdap_domain in the list, which does not work for subdomains. Also when search bases were already initialized in sdap_domain_subdom_add, we should only rewrite them when they were explicitly set in sssd.conf. Resolves: https://pagure.io/SSSD/sssd/issue/3351 Reviewed-by: Jakub Hrozek <jhrozek@redhat.com>
* CONTRIB: Force single-thread install to workaround concurrency issuesFabiano Fidêncio2017-05-021-1/+2
| | | | | | | | @see also https://bugzilla.redhat.com/show_bug.cgi?id=1260190 Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* LDAP: Remove duplicated debug messageFabiano Fidêncio2017-05-021-1/+0
| | | | | Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* NSS: Fix typo inigroups -> initgroupsFabiano Fidêncio2017-05-022-2/+2
| | | | | Signed-off-by: Fabiano Fidêncio <fidencio@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* ssh tools: Split connect and communication phasesSimo Sorce2017-04-281-8/+22
| | | | | | | | | | | | | | | | | We can fallback after a connect error, but we cannot easily fall back once we start sending data as we may have consumed part of the buffer so reconnecting and sending what's left would not make sense. Therefore we now fallback on connect errors, but we issue a hard fail if error happens after communication has been established. Resolves: https://pagure.io/SSSD/sssd/issue/1498 Merges: https://pagure.io/SSSD/sssd/pull-request/3383 Signed-off-by: Simo Sorce <simo@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* ssh tools: Fix issues with multiple IP addressesSimo Sorce2017-04-281-1/+7
| | | | | | | | | | | | | | | Cycle through all resolved address until one succeed or all fail. This is needed for dual stack systems where either IPv4 or IPv6 are improperly configured or selectively filtered at some point along the route. Resolves: https://pagure.io/SSSD/sssd/issue/1498 Merges: https://pagure.io/SSSD/sssd/pull-request/3383 Signed-off-by: Simo Sorce <simo@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* ssh tools: The ai structure is not an array,Simo Sorce2017-04-281-3/+3
| | | | | | | | | | | | | This structure is actually a linked list, so do not mislead readers by treating it as an array. Resolves: https://pagure.io/SSSD/sssd/issue/1498 Merges: https://pagure.io/SSSD/sssd/pull-request/3383 Signed-off-by: Simo Sorce <simo@redhat.com> Reviewed-by: Lukáš Slebodník <lslebodn@redhat.com>
* cache_req: Avoid bool in switch/caseLukas Slebodnik2017-04-281-11/+7
| | | | | | | | Pair-Programmed-With: Michal Židek <mzidek@redhat.com> Reviewed-by: Fabiano Fidêncio <fidencio@redhat.com> Reviewed-by: Pavel Březina <pbrezina@redhat.com> Reviewed-by: Michal Židek <mzidek@redhat.com>
* Use correct spelling of overrideRené Genz2017-04-286-8/+8
| | | | Reviewed-by: Jakub Hrozek <jhrozek@redhat.com>
generated by cgit (git 2.34.1) at 2025-09-02 07:14:01 +0000