openldap-bugs June 2009

openldap-bugs@openldap.org

41 participants
115 discussions

Re: (ITS#6177) bdb delete locking issue
by hyc＠symas.com 18 Jun '09

18 Jun '09

richton(a)nbcs.rutgers.edu wrote: > Howard has me experimenting with moving the lock from cache.c to delete.c. > With this experiment in place, I'm getting a failure with > test008/back-bdb. Under libumem, the crash is in bdb_search line 752. New fix in HEAD, please test. Thanks. -- -- Howard Chu CTO, Symas Corp. http://www.symas.com Director, Highland Sun http://highlandsun.com/hyc/ Chief Architect, OpenLDAP http://www.openldap.org/project/

1 0

(ITS#6181) slapo-unique: slaptest fails in case unique_uri is prepended the keywords "strict" or "ignore"
by daniel＠pluta.biz 18 Jun '09

18 Jun '09

Full_Name: Daniel Pluta Version: OPENLDAP_STABLE OS: Linux URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (91.12.203.84) Hi, from slapo-unique's manpage: "... It is possible to set strict checking for the uniqueness domain by prepending the keyword strict. ..." I've tried to use "unique_uri strict ldap:///....." and/or "unique_uri ignore ldap:///..." configuration syntax within slapd.conf. Both settings cause slaptest to fail with the error message: line 86: <strict> invalid ldap urilist my slapd.conf's line 86 looks (similar) to this one: unique_uri strict ldap:///dc=example,dc=com?uid?sub?(objectClass=account) The error message disapears and slapd starts up only in case the keywords "strict" and/or "ignore" get removed from slapd.conf again. After a quick look into unique's source I think the problem's cause seems to be located within/around the ConfigDriver function "unique_cf_uri", especially regarding the section "case LDAP_MOD_ADD:" if ( c->line ) rc = unique_new_domain ( &domain, c->line, c ); else rc = unique_new_domain ( &domain, c->argv[1], c ); I've added the following Debug statement Debug ( LDAP_DEBUG_TRACE, "c->line: %s\n", c->line, NULL, NULL ); just before the above condition. The output results in the following trace output: ==> unique_db_init line 86 (unique_uri strict ldap:///dc=example,dc=com?uid?sub?(objectClass=account)) c->line: (null) ==> unique_new_domain <strict> Because c->line is (null) only the else branch get's traversed... I don't understand exactly why. I've always used c->argv[n] (never used of c->line) in the past. Regarding the function: unique_new_domain( unique_domain **domainp, char *domain_spec, ConfigArgs *c ) if ( strncasecmp ( uri_start, "ignore ", STRLENOF( "ignore " ) ) == 0 ) { domain->ignore = 1; uri_start += STRLENOF( "ignore " ); } if ( strncasecmp ( uri_start, "strict ", STRLENOF( "strict " ) ) == 0 ) { domain->strict = 1; uri_start += STRLENOF( "strict " ); if ( !domain->ignore && strncasecmp ( uri_start, "ignore ", STRLENOF( "ignore " ) ) == 0 ) { domain->ignore = 1; uri_start += STRLENOF( "ignore " ); } } I'm not sure whether I've understand the design idea correctly but in my opinion the above strncasecmp-condition block shouldn't be located in this function. I would suggest to not handle the parsing and differentiation between "strict | ignore | ldap_uri" here. In my opinion this could also/better be handled in advance; perhaps directly within the ConfigDriver function by using the already parsed configuration tokens c->argv[n] instead of "c->line"? BTW: 1.) initially using "unique_uri ldap:///...." in slapd.conf 2.) converting slapd.conf into slapd.d-format 3.) Starting slapd 4.) Editing olcUniqueURI online via cn=config: inserting "strict" here work's fine (at least no error message is displayed ;-)) I've not tested whether the "strict" logic works correctly after these 4 steps. Cheers Daniel

1 0

(ITS#6180) section 3 manpages need RETURN VALUE section
by hyc＠OpenLDAP.org 18 Jun '09

18 Jun '09

Full_Name: Howard Chu Version: 2.4 OS: URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (76.91.220.157) Submitted by: hyc The return value from libldap/liblber functions is described in the DESCRIPTION section; in most manpages this text is in its own RETURN VALUE section. Even though the information is there, without the header it is easy to miss.

1 0

Re: (ITS#6177) bdb delete locking issue
by richton＠nbcs.rutgers.edu 18 Jun '09

18 Jun '09

Howard has me experimenting with moving the lock from cache.c to delete.c. With this experiment in place, I'm getting a failure with test008/back-bdb. Under libumem, the crash is in bdb_search line 752. I see that rs->sr_un.sru_search.r_entry = 0xbaddcafebaddcafe -- uninitialized per libumem -- is being used as an argument to is_entry_subentry(). Backtrace: https://www.nbcs.rutgers.edu/~richton/test008-bt-200906181020.txt -d -1: https://www.nbcs.rutgers.edu/~richton/test008-slapdlog-200906181020.txt

1 0

Re: (ITS#6173) Error while making
by hyc＠symas.com 18 Jun '09

18 Jun '09

Usage questions belong on the openldap-software mailing list. There is no bug here, this ITS will be closed. Bipin.Nair(a)iflexsolutions.com wrote: > This is a multi-part message in MIME format. > > ------_=_NextPart_001_01C9EFB5.22156E9E > Content-Type: text/plain; > charset="iso-8859-1" > Content-Transfer-Encoding: quoted-printable > > Thanks. > Do you think that the OpenSSL and Berkley DB version also should change = > when I go with a upgraded OpenLDAP version? > =20 > Also,is it mandatory to have Cyrus sasl also installed prior to = > OpenLDAP? -- -- Howard Chu CTO, Symas Corp. http://www.symas.com Director, Highland Sun http://highlandsun.com/hyc/ Chief Architect, OpenLDAP http://www.openldap.org/project/

1 0

(ITS#6179) ACL expand style requres dn.regex in what clause
by rein＠OpenLDAP.org 18 Jun '09

18 Jun '09

Full_Name: Rein Tollevik Version: 2.4.16 OS: Irrelevant URL: Submission from: (NULL) (2a01:600:0:1:21c:23ff:feab:61cd) Submitted by: rein Using the expand style (as in "set.expand") in an ACL requires the <what> clause to include a dn.regex style. This is far too restrictive now that the other dn.* styles and attrs val.regex also provides submatches that can be expanded. The test could be extended to also include these cases, but should in that case never fail since there is always an implicit dn.regex=^.*$ <what> present when no DN is specified. So, I'd prefer to remove this check, as in the patch at the end. As of now I use "set.regex=" and overlook the deprecated warning that gives me... Rein Tollevik Basefarm AS --- aclparse.c 15 May 2009 11:45:31 +0200 1.2 +++ aclparse.c 18 Jun 2009 16:21:02 +0200 @@ -851,16 +851,6 @@ } } - /* expand in <who> needs regex in <what> */ - if ( ( sty == ACL_STYLE_EXPAND || expand ) - && a->acl_dn_style != ACL_STYLE_REGEX ) - { - Debug( LDAP_DEBUG_CONFIG | LDAP_DEBUG_ACL, "%s: line %d: \"expand\" style " - "or modifier used in conjunction with a non-regex <what> clause.\n", - fname, lineno, 0 ); - goto fail; - } - if ( strncasecmp( left, "real", STRLENOF( "real" ) ) == 0 ) { is_realdn = 1; bdn = &b->a_realdn;

1 0

Re: (ITS#6177) bdb delete locking issue
by richton＠nbcs.rutgers.edu 18 Jun '09

18 Jun '09

This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. ---559023410-851401618-1245331022=:1022 Content-Type: TEXT/PLAIN; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE On Thu, 18 Jun 2009, Michael Str=F6der wrote: > richton(a)nbcs.rutgers.edu wrote: >> 2.4.17 prerelease testing revealed SEGVs in test039's slapd.1. > > Maybe that's related to ITS#6170 I've filed? If I'm reading correctly, 6170 and 5733 are connection handling? But then= =20 again, we're talking some sort of corruption, it could manifest itself=20 many ways... ---559023410-851401618-1245331022=:1022--

1 0

Re: (ITS#6178) test048-syncrepl-multiproxy failed (exit 1)
by michael＠stroeder.com 18 Jun '09

18 Jun '09

You can download testrun/ with SLAPD_DEBUG=16389 here: http://www.stroeder.com/temp/openldap/its6178-test048-testrun.tar.gz

1 0

(ITS#6178) test048-syncrepl-multiproxy failed (exit 1)
by michael＠stroeder.com 18 Jun '09

18 Jun '09

Full_Name: Michael Ströder Version: HEAD OS: openSUSE Linux 11.1 URL: Submission from: (NULL) (84.163.104.134) >>>>> Starting test048-syncrepl-multiproxy ... running defines.sh Starting master slapd on TCP/IP port 9011... Using ldapsearch to check that master slapd is running... Using ldapadd to create the context prefix entry in the master... Starting P1 slave slapd on TCP/IP port 9012... Using ldapsearch to check that P1 slave slapd is running... Starting R1 slave slapd on TCP/IP port 9013... Using ldapsearch to check that R1 slave slapd is running... 1 > Using ldapadd to populate the master directory... Waiting 3 seconds for syncrepl to receive changes... 1 < Comparing retrieved entries from master and P1 slave... 1 < Comparing retrieved entries from master and R1 slave... 2 > Stopping the provider, sleeping 7 seconds and restarting it... Using ldapsearch to check that master slapd is running... Using ldapmodify to modify master directory... Waiting 3 seconds for syncrepl to receive changes... 2 < Comparing retrieved entries from master and P1 slave... 2 < Comparing retrieved entries from master and R1 slave... 3 > Stopping slaves to test recovery... Modifying more entries on the master... Restarting P1 slave... Restarting R1 slave... Waiting 3 seconds for syncrepl to receive changes... 3 < Comparing retrieved entries from master and P1 slave... test failed - master and slave databases differ >>>>> ./scripts/test048-syncrepl-multiproxy failed (exit 1) make[2]: *** [bdb-mod] Error 1 make[2]: Leaving directory `/usr/src/michael/openldap/HEAD/openldap/tests' make[1]: *** [test] Error 2 make[1]: Leaving directory `/usr/src/michael/openldap/HEAD/openldap/tests' make: *** [test] Error 2

1 0

Re: (ITS#6177) bdb delete locking issue
by michael＠stroeder.com 18 Jun '09

18 Jun '09

richton(a)nbcs.rutgers.edu wrote: > 2.4.17 prerelease testing revealed SEGVs in test039's slapd.1. Maybe that's related to ITS#6170 I've filed? Ciao, Michael.

1 0

← Newer
1
2
3
4
5
6
7
8
...
12
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

openldap-bugs June 2009