Age | Commit message (Collapse) | Author | Files | Lines |
|
The file has been split in three:
sdap_async_users.c
sdap_async_groups.c
sdap_async_initgroups.c
https://fedorahosted.org/sssd/ticket/864
|
|
This patch deletes memory context parameter in those places in sysdb
where it is not necessary. The code using modified functions has been
updated. Tests updated as well.
|
|
The patch also updates code using modified functions. Tests have also
been adjusted.
|
|
|
|
https://fedorahosted.org/sssd/ticket/951
|
|
|
|
https://fedorahosted.org/sssd/ticket/916
|
|
|
|
|
|
|
|
|
|
https://fedorahosted.org/sssd/ticket/878
|
|
Instead of issuing N LDAP requests when processing a group with N users,
utilize the dereference functionality to pull down all the members in a
single LDAP request.
https://fedorahosted.org/sssd/ticket/799
|
|
This patch splits checking cache and hash tables into standalone
functions. This will make it easy to reuse the code in a new branch that
uses dereferencing.
|
|
Instead of downloading complete user data which is potentionally very
slow, only download the necessary minimum information and store the
users as dummy entries.
|
|
RFC2307bis code relies heavily on originalDN, so the fake users need to
have an option to store it, too.
|
|
|
|
|
|
|
|
|
|
We were not fully compliant with section 5.3 of RFC 2307 which
states:
An account's GECOS field is preferably determined by a value of the
gecos attribute. If no gecos attribute exists, the value of the cn
attribute MUST be used. (The existence of the gecos attribute allows
information embedded in the GECOS field, such as a user's telephone
number, to be returned to the client without overloading the cn
attribute. It also accommodates directories where the common name
does not contain the user's full name.)
|
|
|
|
|
|
If the loop ran through at least one
sdap_process_missing_member_2307() call and errored out later, we
were not canceling the transaction.
|
|
|
|
|
|
This routine will replace the use of sysdb_attrs_to_list() for any
case where we're trying to get the name of the entry. It's a
necessary precaution in case the name is multi-valued.
|
|
|
|
https://fedorahosted.org/sssd/ticket/822
|
|
|
|
Groups in ldap with multiple values for their groupname attribute
will now be compared against the RDN of the entry to determine the
"primary" group name. We will save only this primary group name to
the ldb cache.
|
|
Groups in ldap with multiple values for their groupname attribute
will now be compared against the RDN of the entry to determine the
"primary" group name. We will save only this primary group name
to the ldb cache.
|
|
Users in ldap with multiple values for their username attribute
will now be compared against the RDN of the entry to determine the
"primary" username. We will save only this primary name to the ldb
cache.
|
|
|
|
|
|
|
|
https://fedorahosted.org/sssd/ticket/824
|
|
The IPA provider was not properly removing groups in the cache
that the user was no longer a member of.
https://fedorahosted.org/sssd/ticket/803
|
|
|
|
Sometimes, a value in LDAP will cease to exist (the classic
example being shadowExpire). We need to make sure we purge that
value from SSSD's sysdb as well.
https://fedorahosted.org/sssd/ticket/750
|
|
Nested groups where not unrolled completely during the first enumeration
run because not all where present in the cache.
|
|
|
|
The user may not be a direct member of their primary group, but
we still want to make sure that group is cached on the system.
|
|
After asking the cache for the list of member DNs for groups
during an initgroups request, we were passing it through the
sanitization function. Since this had already been done before
they were saved to the cache, this meant that it was corrupting
the results. It is safe to pass the returned DN directly into the
sysdb_group_dn_name() function.
|
|
|
|
We were not enclosing group processing in a transaction, which was
resulting in extremely high numbers of disk-writes. This patch
adds a transaction around the sdap_process_group code to ensure
that these actions take place within a transaction.
This patch also adds a check around the missing member code for
RFC2307bis so we don't go back to the LDAP server to look up
entries that don't exist (since the enumeration first pass would
already have guaranteed that we have all real users cached)
|
|
|
|
Otherwise fallback to the default modifyTimestamp indicator
|
|
|
|
|