openldap-bugs

openldap-bugs@openldap.org

1 participants
21001 discussions

(ITS#9070) Failing to include openssl1.1.1c with openLDAP
by vishal.bafna307＠gmail.com 29 Aug '19

29 Aug '19

Full_Name: Vishal Bafna Version: 2.4.42 OS: Linux URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (134.204.193.36) I am trying to build openLDAP using openssl1.1.1c. It is failing to locate the SSL/TLS package. I have given the openssl lib and include paths in LDFLAGS and CPPFLAGS respectively in the Makefile of LDAP. The Make file also has --with-tls=openssl \ mentioned as config option snapshot of config option in the ldap Makefile CONFIG_OPTS= --host=$(TARGET) --prefix=${LINUX_ROOT} --exec_prefix=${LINUX_ROOT} --disable-static --enable-shared --with-yielding_select=no --disable-slapd --enable-syslog --disable-bdb --disable-hdb --enable-debug --with-tls=openssl --target=$(TARGET) LDFLAGS="-L${LINUX_ROOT}/lib -L$(LINUX_ROOT)/ssl/lib" CPPFLAGS="-I${LINUX_ROOT}/include -I$(LINUX_ROOT)/ssl/include" \ checking for SSL_library_init in -lssl... no checking for ssl3_accept in -lssl... no configure: error: Could not locate TLS/SSL package Makefile:27: recipe for target 'configure_arm-none-linux-gnueabi-2.4.42' failed Could someone please help with the issue

1 0

Re: (ITS#9069) Stop setting custom GnuTLS mutex functions
by hyc＠symas.com 28 Aug '19

28 Aug '19

Ryan Tandy wrote: > On Wed, Aug 28, 2019 at 01:34:31PM +0100, Howard Chu wrote: >> This is not a safe assumption on Windows, unfortunately. While libldap uses native >> Windows mutexes on Windows, a lot of code built using gcc/g++ uses a posixthreads >> emulation library for thread and mutex support, and the user-visible structures are >> not compatible with native mutexes. (I've found this to be a significant problem >> for g++ std::mutex, std::thread, etc.) > > Thanks for mentioning that. GnuTLS also uses native Windows threads when building for Windows (including mingw-ish environments), so it should still be OK; but > I'd better go and figure out a smoke test on Windows too. > > How does this work on the OpenSSL side? For 1.1.0+ it sounds like the same situation, with OpenSSL using the Windows native API as well. > > Does "incompatible" just mean that a single mutex object can't be shared between the two APIs? (I don't think that would be an issue for our case.) Or is it > something stronger like a pthreads mutex not being able to synchronize native threads, or outright not being able to use both interfaces in a single process? > Passing a native structure to the posixthreads interface (and vice versa) will crash the program (or worse - silently corrupt memory without crashing the program....) -- -- 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

Re: (ITS#9069) Stop setting custom GnuTLS mutex functions
by ryan＠nardis.ca 28 Aug '19

28 Aug '19

On Wed, Aug 28, 2019 at 01:34:31PM +0100, Howard Chu wrote: >This is not a safe assumption on Windows, unfortunately. While libldap uses native >Windows mutexes on Windows, a lot of code built using gcc/g++ uses a posixthreads >emulation library for thread and mutex support, and the user-visible structures are >not compatible with native mutexes. (I've found this to be a significant problem >for g++ std::mutex, std::thread, etc.) Thanks for mentioning that. GnuTLS also uses native Windows threads when building for Windows (including mingw-ish environments), so it should still be OK; but I'd better go and figure out a smoke test on Windows too. How does this work on the OpenSSL side? For 1.1.0+ it sounds like the same situation, with OpenSSL using the Windows native API as well. Does "incompatible" just mean that a single mutex object can't be shared between the two APIs? (I don't think that would be an issue for our case.) Or is it something stronger like a pthreads mutex not being able to synchronize native threads, or outright not being able to use both interfaces in a single process?

1 0

Re: (ITS#9069) Stop setting custom GnuTLS mutex functions
by hyc＠symas.com 28 Aug '19

28 Aug '19

ryan(a)openldap.org wrote: > Full_Name: Ryan Tandy > Version: master > OS: Debian > URL: https://github.com/rtandy/openldap/compare/openldap:master...rtandy:remove-… > Submission from: (NULL) (70.66.128.207) > Submitted by: ryan > > > The GnuTLS documentation > <https://gnutls.org/manual/html_node/Core-TLS-API.html#gnutls_005fglobal_005…> > states: > >> Do not call this function from a library, or preferably from any application > unless really needed to. > Additionally, the custom mutex functions are never reset, so if libldap is > unloaded (for example via dlclose()) after calling this, its code may be > unmapped and the application could crash when GnuTLS calls the mutex functions. > > On typical systems, GnuTLS system mutexes are probably the same as what libldap > uses anyway. This is not a safe assumption on Windows, unfortunately. While libldap uses native Windows mutexes on Windows, a lot of code built using gcc/g++ uses a posixthreads emulation library for thread and mutex support, and the user-visible structures are not compatible with native mutexes. (I've found this to be a significant problem for g++ std::mutex, std::thread, etc.) > > IPR: > > The attached patch file is derived from OpenLDAP Software. All of the > modifications to OpenLDAP Software represented in the following patch were > developed by Ryan Tandy <ryan(a)nardis.ca>. I have not assigned rights and/or > interest in this work to any party. > > I, Ryan Tandy, 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. > > -- -- 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

Re: (ITS#9069) Stop setting custom GnuTLS mutex functions
by ondra＠mistotebe.net 28 Aug '19

28 Aug '19

On Wed, Aug 28, 2019 at 01:04:25AM +0000, ryan(a)openldap.org wrote: > The GnuTLS documentation states: >> Do not call this function from a library, or preferably from any application >> unless really needed to. > > I disobeyed that guidance in commit 829027945, because I wasn't sure that > GnuTLS's own threading support would cover all the platforms libldap does. This > choice caused some bugs, e.g. <https://bugs.debian.org/803197> and > <https://www.openldap.org/its/?findid=8797>. > > I don't know how to find out for sure whether anyone builds libldap with GnuTLS > on a system where it lacks native mutexes. I think at this point I would rather > fix the known broken cases, over the risk of potentially breaking a theoretical > setup I'm not sure actually exists. > > Therefore: I now propose applying this change for 2.5. A similar change (making tlso_thr_init a no-op) has been introduced in ITS#8533 when compiling with OpenSSL 1.1.0+, so I gather it should be fine for GnuTLS as well. Regards, -- Ondřej Kuzník Senior Software Engineer Symas Corporation http://www.symas.com Packaged, certified, and supported LDAP solutions powered by OpenLDAP

1 0

(ITS#9069) Stop setting custom GnuTLS mutex functions
by ryan＠openldap.org 28 Aug '19

28 Aug '19

Full_Name: Ryan Tandy Version: master OS: Debian URL: https://github.com/rtandy/openldap/compare/openldap:master...rtandy:remove-… Submission from: (NULL) (70.66.128.207) Submitted by: ryan The GnuTLS documentation <https://gnutls.org/manual/html_node/Core-TLS-API.html#gnutls_005fglobal_005…> states: > Do not call this function from a library, or preferably from any application unless really needed to. I disobeyed that guidance in commit 829027945, because I wasn't sure that GnuTLS's own threading support would cover all the platforms libldap does. This choice caused some bugs, e.g. <https://bugs.debian.org/803197> and <https://www.openldap.org/its/?findid=8797>. I don't know how to find out for sure whether anyone builds libldap with GnuTLS on a system where it lacks native mutexes. I think at this point I would rather fix the known broken cases, over the risk of potentially breaking a theoretical setup I'm not sure actually exists. Therefore: I now propose applying this change for 2.5. I made this change as a distro patch in Debian 10 (buster), and no Debian or Ubuntu users have reported any regressions so far. I also did some local testing with slapd and slapd-mtread on the GnuTLS versions shipped in the last few releases of Debian. I have not tested any non-Linux platforms yet. For completeness, I should probably try this on at least FreeBSD and Solaris. I kind of assume GnuTLS system mutexes are going to work anywhere LMDB already does. Rationale for the change, from the commit message: Since GnuTLS moved to implicit initialization on library load, calling this function deinitializes GnuTLS and then re-initializes it. When GnuTLS uses /dev/urandom as an entropy source (getrandom() not available, or older versions of GnuTLS), and the application closed all file descriptors at startup, this could result in GnuTLS opening /dev/urandom over one of the application's file descriptors when re-initialized. Additionally, the custom mutex functions are never reset, so if libldap is unloaded (for example via dlclose()) after calling this, its code may be unmapped and the application could crash when GnuTLS calls the mutex functions. On typical systems, GnuTLS system mutexes are probably the same as what libldap uses anyway. IPR: The attached patch file is derived from OpenLDAP Software. All of the modifications to OpenLDAP Software represented in the following patch were developed by Ryan Tandy <ryan(a)nardis.ca>. I have not assigned rights and/or interest in this work to any party. I, Ryan Tandy, 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.

1 0

Re: (ITS#8753) Public key pinning support in libldap
by ryan＠nardis.ca 27 Aug '19

27 Aug '19

The gnutls_digest_get_id function was added in GnuTLS 3.2.2: https://gitlab.com/gnutls/gnutls/blob/gnutls_3_2_2/NEWS That was released in 2013, so I think it's OK to depend on it by now. Please consider applying this patch to update the configure.in check: https://github.com/openldap/openldap/compare/master...rtandy:its8753-gnutls… Currently master fails to compile with older GnuTLS (e.g. on Debian 7/wheezy) after configure succeeded.

1 0

(ITS#9068) mdb_load corrupts escaped backslash
by hyc＠openldap.org 26 Aug '19

26 Aug '19

Full_Name: Howard Chu Version: 0.9.24 OS: URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (80.233.39.186) Submitted by: hyc When reloading a printable dump (from mdb_dump -p), actual backslashes in the input (represented as '\\') will be corrupted. This has been broken since mdb_load was written in 2004.

1 0

Re: (ITS#9067) preferredDeliveryMethod syntax evaluation incorrect
by hyc＠symas.com 26 Aug '19

26 Aug '19

bremer(a)univention.de wrote: > Full_Name: Julia Bremer > Version: 2.4.45 > OS: debian / UCS 4.4 > URL: http://updates.software-univention.de/download/temp/openldap/fix-syntax-eva… > Submission from: (NULL) (82.198.197.8) > > > Trying to set the value of the attribute preferredDeliveryMethod to a sequence > of items, separated by " $ ", e.g "telephone $ videotex" fails with > > ldap_modify: Invalid syntax (21) > additional info: preferredDeliveryMethod: value #0 invalid per syntax > > even though the syntax is correct according to the RFC4517 > (Syntax 1.3.6.1.4.1.1466.115.121.1.14). > https://tools.ietf.org/html/rfc4517#section-3.3.5 > > This is due to an error in the method deliveryMethodValidate in > servers/slapd/schema_init.c which causes the function to check the string > backwards after a space character occurs. > The attached patch fixes this, so that the syntax is validated correctly. Thanks for the report, committed to master. Looks like this code has never been used since Kurt wrote it in 2004. I've tested it locally now but would ask whether you have time to also add a check into the test suite. Thanks. > > > Univention 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. > > -- -- 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#9067) preferredDeliveryMethod syntax evaluation incorrect
by bremer＠univention.de 26 Aug '19

26 Aug '19

Full_Name: Julia Bremer Version: 2.4.45 OS: debian / UCS 4.4 URL: http://updates.software-univention.de/download/temp/openldap/fix-syntax-eva… Submission from: (NULL) (82.198.197.8) Trying to set the value of the attribute preferredDeliveryMethod to a sequence of items, separated by " $ ", e.g "telephone $ videotex" fails with ldap_modify: Invalid syntax (21) additional info: preferredDeliveryMethod: value #0 invalid per syntax even though the syntax is correct according to the RFC4517 (Syntax 1.3.6.1.4.1.1466.115.121.1.14). https://tools.ietf.org/html/rfc4517#section-3.3.5 This is due to an error in the method deliveryMethodValidate in servers/slapd/schema_init.c which causes the function to check the string backwards after a space character occurs. The attached patch fixes this, so that the syntax is validated correctly. Univention 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.

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