| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| ... | |
| |
|
|
| |
Summary: Handle syncing add opererations that have a ntuniqueid present.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bug Description: MMR: Directory updates on same object
Reviewed by: nhosoi (Thanks!)
Fix Description: The problem does appear to be concurrency. I think the original intention of
the urp fixup code was that it should only be run inside the database lock, so
that the database could be restored to a consistent state before the next
operation was processed. However, this requires the database code to know when
the database is already locked, so that if e.g. a modrdn operation needs to
call an internal delete, the database should not be locked again. The flag
OP_FLAG_REPL_FIXUP is used to denote both that the operation is such an
internal operation, and that the database should not be locked again.
There are a couple of cases where these operations can be called from outside
of the database lock:
urp_fixup_rename_entry is called from multimaster_postop_modrdn and
multimaster_postop_delete, both of which are front end post op plugins, not
called from within the database lock. Same with urp_fixup_delete_entry and
urp_fixup_modify_entry. In other cases, such as urp_fixup_add_entry, and other
places where urp_fixup_rename_entry and urp_fixup_modify_entry are called, they
are called from a bepostop plugin function, which is called after the original
database operation has been processed, within the database lock. So the
solution appears to be to move the urp_* functions to the bepostop plugin
functions. One of these functions does an internal search -
urp_get_min_naming_conflict_entry - but it does not appear that search locks
the database, so there was nothing to be done to make it "reentrant".
Without this patch, I can crash the server in a matter of minutes (x86_64
rhel5) using the latest Fedora DS 1.1 code. With the patch, the server runs
for several hours (maybe longer, I had to stop the test).
Also, to really exercise the urp code, I added a rename operation between the
add and delete e.g.
add("ou=test");
rename("ou=test", "ou=test2");
delete("ou=test2");
The server still runs for several hours with no problems.
Platforms tested: RHEL5 x86_64
Flag Day: no
Doc impact: no
|
| |
|
|
|
|
|
|
|
|
|
| |
Bug Description: Show-Stopper - Migration from DS 6.21 to DS80
Reviewed by: nhosoi (Thanks!)
Fix Description: Added a new function migrateNetscapeRoot which will create a temporary LDIF file from the given NetscapeRoot.ldif file. The function will look for all \bNetscape\b occurances and convert them to @capbrand@ where that is defined as the capitalized brand name in configure. It will then import this temporary LDIF file and delete it.
Platforms tested: RHEL5 x86_64
Flag Day: no
Doc impact: no
QA impact: should be covered by regular nightly and manual testing
New Tests integrated into TET: none
|
| |
|
|
|
|
|
|
|
| |
Bug Description: crash at startup with new ldap sdk on 64-bit platform
Reviewed by: nkinder (Thanks!)
Fix Description: Use LBER_SOCKET as the socket type instead of int. There was also a place in ldclt where we were doing the same thing.
Platforms tested: RHEL5 x86_64
Flag Day: no
Doc impact: no
|
| |
|
|
|
|
|
|
| |
Summary: verify-db.pl (db_verify) does not work on a little endian machine
(comment #1-#4)
Description: 1) introducing dbverify mode to ns-slapd.
2) providing new script dbverify to call "ns-slapd dbverify"
3) fixing verify-db.pl to call dbverify instead of db_verify from BDB
|
| |
|
|
|
|
|
|
|
|
|
| |
Bug Description: Pass in schema and config LDIF files to setup
Reviewed by: nhosoi (Thanks!)
Fix Description: 1) Allow multi-valued parameters in .inf files and command line. These values will be represented internally as an array ref. No existing parameters allow being multi-valued (e.g. you can't use Suffix=o=foo and Suffix=o=bar)
2) Add two new .inf parameters - SchemaFile and ConfigFile. The files listed in SchemaFile will be copied into the schema subdirectory of the new instance, so they must already be named appropriately (e.g. 60foo.ldif). The files listed in ConfigFile must be LDIF files with one or more whole entries to be added to the initial dse.ldif. These could be additional suffixes/databases to create, plugin configuration, replication configuration, or anything else.
Right now, if you have an LDIF file that relies on custom schema, you cannot use the InstallLdifFile directive during setup. SchemaFile allows you to do that.
Platforms tested: RHEL5
Flag Day: no
Doc impact: Will need to document the two additional parameters.
|
| |
|
|
|
|
| |
Summary: Replication monitor CGI seems misconfigured (comment #1)
1) get_replica: missing return value for the success case
2) support the case "shadowport" is not passed
|
| |
|
|
|
|
|
|
|
|
|
| |
Bug Description: 'Operation Error' encountered during DS6.21 to DS80 cross-platform migration
Reviewed by: nhosoi (Thanks!)
Fix Description: For cross platform migration, we should not migrate the nsslapd-idl-switch setting, we should just use the new default value which is "new".
Platforms tested: RHEL4
Flag Day: no
Doc impact: no
QA impact: should be covered by regular nightly and manual testing
New Tests integrated into TET: none
|
| |
|
|
| |
Summary: Corrected search scope used to find entries to sync in winsync total update protocol.
|
| |
|
|
|
| |
Summary: Migration script references a non-existing directory (comment #22)
Description: introduced ignoreOldEntries not to migrate obsolete config entries.
|
| |
|
|
| |
Summary: Handle syncing multi-valued street attribute to AD.
|
| |
|
|
|
|
|
|
|
|
|
| |
Bug Description: Migration script does not migrate nsDS5ReplicaCredentials correctly.
Reviewed by: nhosoi (Thanks!)
Fix Description: This was a big endian vs. little endian issue. We only use name based UUID generation with the reversible password code. This code was not doing the ntoh with the numeric values generated. I'm sure there is probably a compiler warning about this on some platform.
Platforms tested: RHEL5 x86_64, Solaris 9 64-bit
Flag Day: no
Doc impact: no
QA impact: should be covered by regular nightly and manual testing
New Tests integrated into TET: none
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bug Description: Migration script references a non-existing directory
Reviewed by: nhosoi (Thanks!)
Fix Description: This fixes a couple of problems.
1) Use the inst_dir from the directory server as the instance dir where the ldif2db script is found.
2) The password for migratecred should be quoted before being passed to the shell, in case there are shell meta chars in there
3) If using cross platform migration, and no LDIF files were found to migrate, this will cause an error message to be printed and migration will be aborted.
Platforms tested: RHEL4 i386, RHEL5 x86_64
Flag Day: no
Doc impact: no
QA impact: should be covered by regular nightly and manual testing
New Tests integrated into TET: none
|
| |
|
|
|
| |
Summary: Migration script references a non-existing directory (comment #5,#7)
Description: 3 more fixes for the with-fhs-opt case
|
| |
|
|
|
| |
Summary: Migration script references a non-existing directory
Description: added an if clause for the with_fhs_opt case
|
| |
|
|
| |
Summary: Check if passwords are already hashed before sync'ing with AD.
|
| |
|
|
|
|
|
|
|
| |
Bug Description: enabling chain-on-update causes replica to act as a master
Reviewed by: nkinder (Thanks!)
Fix Description: This fix is only minor - we must use sym_load() to look up plugins, including the entry distribution plugin we use for chain on update. But I don't believe this was causing the problem. Chain on update does not work if you BIND as directory manager. You must bind as a regular user. It may be difficult to change this. We need to do more testing to see if, in general, proxy BIND and operations work with directory manager. The chaining backend cannot use directory manager as the proxy user.
Platforms tested: RHEL5
Flag Day: no
Doc impact: Yes, we need to make sure we document exactly how chain on update is to be used.
|
| |
|
|
| |
Summary: Trim initials attribute value when sync'ing to AD.
|
| |
|
|
| |
Summary: Support FHS opt layout for perldir and propertydir.
|
| | |
|
| |
|
|
| |
Summary: Check if pkgconfig is available before trying to use it.
|
| |
|
|
|
|
| |
Summary: cyclic dependency from getpwnam() in log rotation code
Description: Moved getpwnam call to the startup time, store the info in
slapdFrontendConfig to reuse.
|
| |
|
|
|
|
|
|
| |
Bug Description: Does not build on Fedora 8
Fix Description: If using the O_CREAT flag with open(), the file mode must also be given. Also, the bdb calls to use ->open() must use parentheses around the function pointer access e.g. (DB->open)(args...) instead of just DB->open(args).
Platforms tested: RHEL4, Fedora 8
Flag Day: no
Doc impact: no
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bug Description: 7.1 to 8.0 Migration Bug
Reviewed by: nhosoi (Thanks!)
Fix Description: Lots of issues
1) Instead of using Net::Domain::hostname() for the hostname to use for server identifiers, we must use the leftmost component of the FullMachineName specified by the user. One of the reasons is that hostname() and hostfqdn() can give different results such that hostname() is not the leftmost component of hostfqdn(). And we should just use whatever the user specifies. This required several changes to the mapfiles, and a change to the maptable processing, to process the perl code to eval last, so that we can use token substitutions from the inf file and from hard coded strings.
2) We need to add the global preferences stuff, during migration, in order for the console to function. We cannot rely on the migrated o=NetscapeRoot data because we have to make sure we specify the new jar files to use. In addition, we need to migrate over any customizations that the user has made to these preferences. This is handled by the new updateConsoleInfo function called during admin server migration.
3) There were several resources for messages missing. These have been added.
4) With the new perl only ds instance creation code, there will actually be data in the database even though the server is not started. This is what I believe was causing the error_removing_temp_db_files error. So now, the code will only attempt to remove plain files, not directories that could be globbed.
5) Use /opt/brand-ds for the old server root instead of /opt/pkgname.
6) The migration log should use [Migration} not [Setup]
7) migration should not report a fatal error upon success
Platforms tested: RHEL4
Flag Day: no
Doc impact: no
QA impact: should be covered by regular nightly and manual testing
New Tests integrated into TET: none
|
| |
|
|
|
|
|
|
|
|
|
| |
Bug Description: 7.1 to 8.0 Migration Bug
Reviewed by: nhosoi (Thanks!)
Files: see diff
Branch: HEAD
Fix Description: Although the directory server always creates key/cert databases, the admin server does not. So, just log a message and skip if they do not exist. Also was missing some message resources for various errors.
Platforms tested: RHEL4
Flag Day: no
Doc impact: no
|
| |
|
|
|
|
|
|
|
|
| |
Bug Description: Change filesystem path naming from "fedora-ds" to "dirsrv"
Reviewed by: nkinder (Thanks!)
Fix Description: Mostly just changing the package name in configure.ac, and making sure we consistently use that in path naming (e.g. /etc/@PACKAGE_NAME@ or /etc/@PACKAGE_NAME_BASE@ for adminserver).
Platforms tested: RHEL4, FC6
Flag Day: no
Doc impact: Oh yes.
QA impact: Any existing tests that depend on /path/brand-ds will need to change to use dirsrv. It is highly encouraged to use a macro or variable for the package name in any scripts to minimize the impact of future package name changes.
|
| |
|
|
| |
those first in the link line, so that the build finds the dirsec versions of those components and not the mozilla/seamonkey ones in libdir
|
| |
|
|
|
| |
Description: Update dsktune for 1.1
Updated the date.
|
| |
|
|
|
| |
Description: update dsktune for 1.1
These are the latest Solaris patches.
|
| |
|
|
| |
be ignored, which causes rpm packaging errors
|
| | |
|
| |
|
|
| |
Summary: Changed ldap-agent to find location of stats file dynamically.
|
| |
|
|
| |
Windows.
|
| |
|
|
|
|
| |
Summary: ds_remove cannot remove/rename directories (Comment #2)
Description: when creating inst_dir and config_dir, make sure the parent dirs
(lib/<brand>-ds and etc/<brand>-ds) have the specified gid.
|
| |
|
|
|
| |
Description: Update dsktune for 1.1
Fix Description: This is for HP-UX 11.23 IPF 64bit
|
| |
|
|
|
|
| |
Description: Move DS Admin Code into Admin Server - dsktune
Fix Description: This adds the dsktune dialog to the initial list of setup dialogs.
Platforms tested: RHEL4
|
| |
|
|
| |
Summary: not all the addresses associated with listenhost are bound to listen sockets (comment #10)
|
| |
|
|
|
|
|
|
|
|
| |
Bug Description: improve perldap script execution ability on bundled platforms
Reviewed by: nkinder (Thanks!)
Fix Description: Most platforms will just use perl from PATH. However, on Solaris and HP-UX, we have to use special 64 bit versions to execute perldap, since perldap is 64 bit on those platforms. Also, if bundling all of the dependent components into the single package, we need to make sure the perl library path is set correctly to find perldap.
The last step will be to build our version of perldap on the bundled platforms to use rpath to point to the correct runtime library location.
Platforms tested: RHEL4, HP-UX 11.23 IPF 64 bit
Flag Day: no
Doc impact: no
|
| |
|
|
|
| |
Summary: rsearch - make search timeout a configurable parameter
Description: Introduced a new option "-o <search time limit>"
|
| |
|
|
| |
Summary: Implement a script registering DS to the Configuration DS (comment#5)
|
| |
|
|
|
| |
Summary: restart-slapd has an unreplaced token {{INST-DIR}}
Description: hash key, value: INST-DIR => $inf->{slapd}->{inst_dir} is added
|
| |
|
|
|
| |
Description: setup-ds.pl should create default org entries
Fix Description: Make the default in interactive mode to be to add those entries.
|
| |
|
|
|
|
|
|
|
| |
Bug Description: cn equality index missing by default
Reviewed by: nhosoi (Thanks!)
Fix Description: When creating the database instance during dse.ldif processing, we do not create the user defined indexes from the defaults for this backend. This used to work in the old setup code because that code would always add the configuration for the indexes for the new instance. The way it is supposed to work is that a new instance should just copy the default indexes for that backend. This works fine when adding an instance via LDAP but not during startup. I just added a call to have this done during startup. I also removed some obsolete indexes from the default indexes.
Platforms tested: RHEL4
Flag Day: no
Doc impact: no
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
| |
Bug Description: quick install failed when login userid doesn't match install user's id
Reviewed by: me
Fix Description: getlogin returns the _login_ ID which is the initial login id (from utmp). We want the username of the effective user ID, so I just needed to change getLogin to do that.
Platforms tested: RHEL4
Flag Day: no
Doc impact: no
|
| |
|
|
|
|
|
|
|
| |
Bug Description: Replace ds_newinst binary with perl script - remove dead code
Reviewed by: nhosoi (Thanks!)
Fix Description: This is my favorite part - removing all of the now dead code. I also had to fix the exit code so that you could use $? to test for the success or failure of the new setup scripts.
Platforms tested: RHEL4, FC6
Flag Day: no
Doc impact: no
|
| |
|
|
|
|
| |
Description: CLU: dbgen.pl fails to load data files
Reviewed by: nhosoi (Thanks!)
Fix Description: dbgen.pl is hardcoded to look for data files in ../data. Instead, we should allow the user to provide the location to the data files e.g. -l /path/to/data, and we should also use share/brand-ds/data as the default directory.
|
