openldap-bugs

openldap-bugs@openldap.org

1 participants
20965 discussions

Re: (ITS#8303) An Asynchronous meta back-end for OpenLDAP
by nivanova＠symas.com 03 May '16

03 May '16

This is a patch that removes some unused attributes from olcAsyncMetaConfig. These attributes were removed when some configuration options were removed for asyncmeta, being no longer applicable, but removing them from the class definition was missed. ftp://ftp.openldap.org/incoming/nadezhda-ivanova-160503.patch

1 0

(ITS#8417) mdb_load could use an option to specify the mapsize.
by eriixblaike+git＠gmail.com 03 May '16

03 May '16

Full_Name: Eric R. Monson Version: git 3fe9678baa1be8067e7e0bbb94007a096952e62c OS: Linux URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (72.193.242.31) The lack of a mapsize option available in mdb_load can make it inconvenient to load datasets. This is particularly true when you are using the plaintext format specified by '-T', since it has no header you could add an appropriate mapsize to. My patch resolves this by adding a mapsize option ('-m') which can be used to manually specify a mapsize of your choosing, overriding both the default and any value specified in the header of the database you're loading. The patch also modifies the man page for mdb_load to reflect the new functionality. I don't know if the code is good enough for inclusion, but I figured providing my solution was better than nothing. I, Eric Monson, hereby place the following modifications to OpenLDAP Software (and only these modifications) into the public domain. Hence, these modifications may be freely used and/or redistributed for any purpose with or without attribution and/or other notice. https://eriix.org/download_file/eric-monson-16-05-03.patch

1 0

(ITS#8416) Cross compiling failed
by vystrcil.martin＠gmail.com 03 May '16

03 May '16

Full_Name: Martin Vystrcil Version: 2.4.4 OS: Linux Mint URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (147.32.219.136) While cross compiling it is not able to strip final binaries. My config command: ./configure --host=arm-linux-gnueabihf CC=/home/martin/Development/SMON3/testing/gcc-linaro-arm-linux-gnueabihf-4.8-2014.04_linux/bin/arm-linux-gnueabihf-gcc STRIP=/home/martin/Development/SMON3/testing/gcc-linaro-arm-linux-gnueabihf-4.8-2014.04_linux/bin/arm-linux-gnueabihf-strip --with-yielding_select=yes --prefix=/home/martin/Development/SMON3/testing/ldap What it prints (fail with) make[2]: Entering director%6`/home/martin/Development/SMON3/testing/openldap-2.4.44/clients/tools' ../../build/shtool mkdir -p /home/martin/Development/SMON3/testing/ldap/bin ../../build/shtool install -c -m 755 -s ldapsearch /home/martin/Development/SMON3/testing/ldap/bin/ldapsearch strip: Unable to recognise the format of the input file `/home/martin/Development/SMON3/testing/ldap/bin/#INST@30215#' Why this happen ? In file build/shtool is following construction. if [ ".$opt_s" = .yes ]; then if [ ".$opt_t" = .yes ]; then echo "strip $dsttmp" 1>&2 fi strip $dsttmp || shtool_exit $? fi In my opinion there should be strip based on config. In config there it is set correctly. Please lememe know, h t to fix this correctly, I'm not such an expert for automake.

1 0

Re: (ITS#8415) Documentation has hard coded date
by quanah＠zimbra.com 02 May '16

02 May '16

--On Monday, May 02, 2016 8:57 PM +0200 Hallvard Breien Furuseth <h.b.furuseth(a)usit.uio.no> wrote: > openldap-etc.git:misc/crupdate.sh needs a fix so it also > recognizes these two dates, and/or the .sdf copyright notices > can be modified so crupdate catches them. Right, that's the goal here. --Quanah -- Quanah Gibson-Mount Platform Architect Zimbra, Inc. -------------------- Zimbra :: the leader in open source messaging and collaboration A division of Synacor, Inc

1 0

Re: (ITS#8415) Documentation has hard coded date
by h.b.furuseth＠usit.uio.no 02 May '16

02 May '16

On 02/05/16 16:37, quanah(a)openldap.org wrote: > The admin guide has a hard coded year in plain.sdf and preamble.sdf. > > We need to look at using an include file or similar that can have its dates > updated when we do copyright updates. "Doing copyright updates" _is_ updating hard-coded years in the source. I'm pretty sure they ought to stay hard-coded. openldap-etc.git:misc/crupdate.sh needs a fix so it also recognizes these two dates, and/or the .sdf copyright notices can be modified so crupdate catches them.

1 0

(ITS#8415) Documentation has hard coded date
by quanah＠openldap.org 02 May '16

02 May '16

Full_Name: Quanah Gibson-Mount Version: N/A OS: N/A URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (75.111.52.177) The admin guide has a hard coded year in plain.sdf and preamble.sdf. We need to look at using an include file or similar that can have its dates updated when we do copyright updates.

1 0

(ITS#8414) test008 is non portable due to use of "time" command
by quanah＠openldap.org 29 Apr '16

29 Apr '16

Full_Name: Quanah Gibson-Mount Version: 2.4.44 OS: Linux 2.6 URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (75.111.52.177) test008 uses the "time" command as part of the test script. However, that is not portable as the default /bin/sh shell on Debian and Ubuntu does not have "time" as a built-in. "time" is only present there for debugging, and is not required for the test itself. Time to remove time.

1 0

Re: (ITS#8404) Fix for an assertion failure when olcDbRewrite is modified on a meta or asyncmeta database
by hyc＠symas.com 28 Apr '16

28 Apr '16

nivanova(a)symas.com wrote: > Full_Name: Nadezhda Ivanova > Version: 2.5 > OS: Ubuntu 14.04 > URL: ftp://ftp.openldap.org/incoming/nadezhda-ivanova-160414.patch > Submission from: (NULL) (78.83.54.234) > > > The problem was reproduced when attempting to modify the dynamic configuration > of meta and asyncmeta rewrite engine. On a meta database that has rewrite > enabled and some rewrite rules configured, execute ldapmodify with the following > LDIF: > dn: olcMetaSub={0}uri,olcDatabase={1}meta,cn=config > replace: olcDbRewrite > olcDbRewrite: rewriteEngine "off" Hm, multiple problems with this patch: > diff --git a/servers/slapd/back-asyncmeta/config.c b/servers/slapd/back-asyncmeta/config.c > index 1801a55..a4fbb91 100644 > --- a/servers/slapd/back-asyncmeta/config.c > +++ b/servers/slapd/back-asyncmeta/config.c > @@ -1797,14 +1797,66 @@ asyncmeta_back_cf_gen( ConfigArgs *c ) > > case LDAP_BACK_CFG_SUFFIXM: /* unused */ > case LDAP_BACK_CFG_REWRITE: > - if ( mt->mt_rwmap.rwm_bva_rewrite ) { > - ber_bvarray_free( mt->mt_rwmap.rwm_bva_rewrite ); > - mt->mt_rwmap.rwm_bva_rewrite = NULL; > - } > - if ( mt->mt_rwmap.rwm_rw ) > + { > + char *argv0; > + if ( c->valx >= 0 ) { > + int i; > + > + for ( i = 0; !BER_BVISNULL( &mt->mt_rwmap.rwm_bva_rewrite[ i ] ); i++ ) Need a semicolon for the empty for-loop. Or, if it's intended that the following if-block is part of the loop, it needs to be properly enclosed in braces and indented. > + > + if ( c->valx >= i ) { > + rc = 1; > + break; > + } > + ber_memfree( mt->mt_rwmap.rwm_bva_rewrite[ c->valx ].bv_val ); > + for ( i = c->valx; !BER_BVISNULL( &mt->mt_rwmap.rwm_bva_rewrite[ i + 1 ] ); i++ ) > + { > + mt->mt_rwmap.rwm_bva_rewrite[ i ] = mt->mt_rwmap.rwm_bva_rewrite[ i + 1 ]; > + } > + BER_BVZERO( &mt->mt_rwmap.rwm_bva_rewrite[ i ] ); > + > rewrite_info_delete( &mt->mt_rwmap.rwm_rw ); > - break; > + assert( mt->mt_rwmap.rwm_rw == NULL ); > + > + rc = asyncmeta_rwi_init( &mt->mt_rwmap.rwm_rw ); > + > + for ( i = 0; !BER_BVISNULL( &mt->mt_rwmap.rwm_bva_rewrite[ i ] ); i++ ) > + { > + ConfigArgs ca = { 0 }; > + > + ca.line = mt->mt_rwmap.rwm_bva_rewrite[ i ].bv_val; > + ca.argc = 0; Unnecessary initialization, you already set all of ca = {0}. > + init_config_argv( &ca ); > + config_parse_ldif( &ca ); > + > + argv0 = ca.argv[ 0 ]; What is argv0 being saved for? > + if ( !strcasecmp( ca.argv[0], "suffixmassage" )) { > + rc = asyncmeta_suffixm_config( &ca, ca.argc, ca.argv, mt ); > + } else { > + rc = rewrite_parse( mt->mt_rwmap.rwm_rw, > + c->fname, c->lineno, ca.argc, ca.argv ); > + } > + > + ca.argv[ 0 ] = argv0; Why is argv0 being restored? All you're doing after this is freeing ca.argv anyway. > + > + ch_free( ca.tline ); > + ch_free( ca.argv ); > > + assert( rc == 0 ); > + } > + > + } else if ( mt->mt_rwmap.rwm_rw != NULL ) { > + if ( mt->mt_rwmap.rwm_bva_rewrite ) { > + ber_bvarray_free( mt->mt_rwmap.rwm_bva_rewrite ); > + mt->mt_rwmap.rwm_bva_rewrite = NULL; > + } > + if ( mt->mt_rwmap.rwm_rw ) > + rewrite_info_delete( &mt->mt_rwmap.rwm_rw ); > + > + asyncmeta_rwi_init( &mt->mt_rwmap.rwm_rw ); > + } > + } > + break; > case LDAP_BACK_CFG_MAP: > if ( mt->mt_rwmap.rwm_bva_map ) { > ber_bvarray_free( mt->mt_rwmap.rwm_bva_map ); -- -- 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#8413) Assertion in back-mdb/search.c during replication
by quanah＠openldap.org 27 Apr '16

27 Apr '16

Full_Name: Quanah Gibson-Mount Version: 2.4.44 OS: Linux URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (75.111.52.177) During replication from the accesslog DB, in a 4-way MMR setup, various masters periodically crash with slapd: search.c:1246: oc_filter: Assertion `f != ((void *)0)' failed. This is back-mdb/search.c, not slapd-search.c This is triggered when a NULL filter is passed through. However, it should be impossible for the filter generated by str2filter to ever fail.

1 0

ITS#8410 published: SECURITY: LDAP connection issue with
by openldap-its＠OpenLDAP.org 26 Apr '16

26 Apr '16

Private ITS#8410 is now publicly readable URL: http://www.OpenLDAP.org/its/?findid=8410

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

openldap-bugs