I don't remember if we posted this already or not. We're prepping 2.4.6 for release; please test RE24 and submit any problems you encounter to the ITS.
Likewise, RE23 looks about ready. If you're interested in 2.3.39 please test and report results for RE23 as well.
Currently all tests in RE24 pass for me on OpenSUSE 10.2 x86_64 and FedoraCore6 x86_64.
--On Thursday, October 18, 2007 9:16 PM -0700 Howard Chu hyc@symas.com wrote:
I don't remember if we posted this already or not. We're prepping 2.4.6 for release; please test RE24 and submit any problems you encounter to the ITS.
Likewise, RE23 looks about ready. If you're interested in 2.3.39 please test and report results for RE23 as well.
Currently all tests in RE24 pass for me on OpenSUSE 10.2 x86_64 and FedoraCore6 x86_64.
I seem to hit a bug with gcc 4.1.1 and -O2, where test008 will fail about 50% of the time. When only using -O1, it passes every time. This is on x86_64 CentOS 5. So beware gcc 4.1.1 and -O2.
--Quanah
--
Quanah Gibson-Mount Principal Software Engineer Zimbra, Inc -------------------- Zimbra :: the leader in open source messaging and collaboration
Is it usual/expected to get failed binds during "make test" in RE24?
PID=25508 - Bind(1000): base="ou=People,dc=example,dc=com", filter="(userPassword=*)" attr="userPassword". PID=25508 - Bind base="ou=People,dc=example,dc=com" filter="(userPassword=*)" got 4 values. PID=25527 - Search(500): base="dc=example,dc=com" scope=sub filter="slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) (cn=Bjorn Jensen)" attrs=cn (more...). PID=25546 - Read(1000): entry="cn=James A Jones 1, ou=Alumni Association, ou=People, dc=example,dc=com". slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49)
On Tue, 23 Oct 2007, Aaron Richton wrote:
On Thu, 18 Oct 2007, Howard Chu wrote:
Likewise, RE23 looks about ready. If you're interested in 2.3.39 please test and report results for RE23 as well.
RE23 back-bdb/hdb/meta/ldap tests good (Sun Studio 11, sparcv7 and sparcv9 ABI). I'll run RE24 today...
--On Tuesday, October 23, 2007 3:05 PM -0400 Aaron Richton richton@nbcs.rutgers.edu wrote:
Is it usual/expected to get failed binds during "make test" in RE24?
Yep. That's one of the tests.
--Quanah
--
Quanah Gibson-Mount Principal Software Engineer Zimbra, Inc -------------------- Zimbra :: the leader in open source messaging and collaboration
Quanah Gibson-Mount wrote:
--On Tuesday, October 23, 2007 3:05 PM -0400 Aaron Richton richton@nbcs.rutgers.edu wrote:
Is it usual/expected to get failed binds during "make test" in RE24?
Yep. That's one of the tests.
Note that in general the test suite allows one to define what errors are acceptable and what are not (the -i switch, which allows textual form for errors that should be ignored: just remove the LDAP_ prefix from usual macros). Since one can't tell with enough granularity what error is expected from each single operation when a mix of success and failure are possible, the designer of the test should be careful. I'll probably extend the format of the test inputs to allow to optionally define the expected result set for each operation, as soon as required by any specific test. In the above case, we force some bind failures by attempting binds with all users with password defined in the database, and by denying some of them auth access to their password.
p.
Ing. Pierangelo Masarati OpenLDAP Core Team
SysNet s.r.l. via Dossi, 8 - 27100 Pavia - ITALIA http://www.sys-net.it --------------------------------------- Office: +39 02 23998309 Mobile: +39 333 4963172 Email: pierangelo.masarati@sys-net.it ---------------------------------------
Aaron Richton wrote:
Is it usual/expected to get failed binds during "make test" in RE24?
From slapd-bind, yes - some invocations search the directory for a bunch of users and then randomly select users from the result set to Bind against. Some of the results will include entries that are being Added and Deleted by the slapd-adddel clients, so occasionally those entries will be missing and the Binds will fail. Perfectly normal and the slapd-bind client will continue running in those cases. If there was an error worth worrying about the client programs would exit with a failure status.
PID=25508 - Bind(1000): base="ou=People,dc=example,dc=com", filter="(userPassword=*)" attr="userPassword". PID=25508 - Bind base="ou=People,dc=example,dc=com" filter="(userPassword=*)" got 4 values. PID=25527 - Search(500): base="dc=example,dc=com" scope=sub filter="slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) (cn=Bjorn Jensen)" attrs=cn (more...). PID=25546 - Read(1000): entry="cn=James A Jones 1, ou=Alumni Association, ou=People, dc=example,dc=com". slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49) slapd-bind PID=25508: ldap_sasl_bind_s: Invalid credentials (49)
On Tue, 23 Oct 2007, Aaron Richton wrote:
On Thu, 18 Oct 2007, Howard Chu wrote:
Likewise, RE23 looks about ready. If you're interested in 2.3.39 please test and report results for RE23 as well.
RE23 back-bdb/hdb/meta/ldap tests good (Sun Studio 11, sparcv7 and sparcv9 ABI). I'll run RE24 today...
-
Aaron Richton -
Howard Chu -
Pierangelo Masarati -
Quanah Gibson-Mount