openldap-bugs

openldap-bugs@openldap.org

21001 discussions

(ITS#5534) Samba4 needs internal transactions/consistancy
by abartlet＠samba.org 27 May '08

27 May '08

Full_Name: Andrew Bartlett Version: CVS HEAD OS: Fedora 9 URL: http://www.openldap.org/lists/openldap-technical/200803/msg00101.html Submission from: (NULL) (59.167.251.137) For Samba4, I need a few things, detailed in the attached URL. This ITS is for internal transactions and validation - the ability to have a openldap overlay roll back all the changes so far, because a precondition is not met. I need the memberOf and refint modules to ensure that no dangling links ever exist, even over subtree renames and invalid modifies, and that a transaction ensures this is always the case. This needs to occur even between databases on the server, but I won't ask that it occur outside the known trees.

1 0

Re: (ITS#5318) response msgid <= 0 mishandled in libldap
by hyc＠symas.com 27 May '08

27 May '08

h.b.furuseth(a)usit.uio.no wrote: > Full_Name: Hallvard B Furuseth > Version: HEAD > OS: > URL: > Submission from: (NULL) (129.240.6.233) > Submitted by: hallvard > > > libraries/libldap/result.c:try_read1msg() accesses 'lr' uninitialized > if 'id' (message ID) from line 577's 'ber_get_int( ber,&id )' is<= 0. > > I'm not sure if the client should terminate the connection when it > receives message id< 0, or if it should just toss the response like > it does with unknown message IDs. RFC4511 isn't really explicit here, although it does say that the connection should be dropped for unparsable messages. Anyway, I've patched HEAD to toss the messages for now. > With message ID 0, the code reaches this statement with 'lr' uninitialized: > Debug( LDAP_DEBUG_TRACE, > "read1msg: ld %p msgid %ld message type %s\n", > (void *)ld, (long)lr->lr_msgid, ldap_int_msgtype2str( tag ) ); > As far as I can tell, normally lr->lr_msgid == id. I haven't tracked what > those values are with LDAP_CONNECTIONLESS at the 'nextresp2:' label. For CONNECTIONLESS, it can only jump back there because there were multiple responses to the current request. So the lr is the same as for the first response. No problem there. (Remember this is all within a single datagram. Responses to multiple different requests cannot be interleaved at this point.) > A 700-line function with 5 labels, yuck. > Anyway, I wonder why taht statement and the statement below: > if ( id == 0 ) { > doesn't use the same value, either id or lr->lr_msgid for both. -- -- 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#5533) test035-meta crashes --without-threads
by h.b.furuseth＠usit.uio.no 27 May '08

27 May '08

quanah(a)zimbra.com writes: > Isn't this just a duplicate of ITS#5489? Apparently not. Got it again at 4th run with daemon.c 1.380.2.12. Also in a somewhat patched HEAD with the same configuration. Hm, the crashes happen for extended operations. So far 6 times "Passwd ExOp failed (1)!", once "WhoAmI failed (49)!". -- Hallvard

1 0

Re: (ITS#5533) test035-meta crashes --without-threads
by quanah＠zimbra.com 27 May '08

27 May '08

--On Tuesday, May 27, 2008 8:05 PM +0000 h.b.furuseth(a)usit.uio.no wrote: > Full_Name: Hallvard B Furuseth > Version: RE24 > OS: > URL: ftp://ftp.openldap.org/incoming/Hallvard-Furuseth-080527.tgz.1 > Submission from: (NULL) (129.240.6.233) > Submitted by: hallvard > Isn't this just a duplicate of ITS#5489? --Quanah -- Quanah Gibson-Mount Principal Software Engineer Zimbra, Inc -------------------- Zimbra :: the leader in open source messaging and collaboration

1 0

(ITS#5533) test035-meta crashes --without-threads
by h.b.furuseth＠usit.uio.no 27 May '08

27 May '08

Full_Name: Hallvard B Furuseth Version: RE24 OS: URL: ftp://ftp.openldap.org/incoming/Hallvard-Furuseth-080527.tgz.1 Submission from: (NULL) (129.240.6.233) Submitted by: hallvard slapd crashes in test035 after 10 runs: daemon.c:979: assert( SLAP_SOCK_IS_ACTIVE( s )); RE24 + libraries/libldap_r/thr_stub.c rev 1.35 ./configure --quiet --without-threads --enable-meta --enable-ldap 'CFLAGS=-O0 -g' (declare -i n=0; while n=n+1 && echo "run #$n" && ./run test035; do sleep 5; done) Testrun directory enclosed, with files OUTPUT (from tty) and slapd.*.GDB. Removed database directories though. (Argh, I messed up with passive mode again. Oh well.)

1 0

Re: (ITS#5524) Additions to the Security section of the Admin Guide
by ghenry＠OpenLDAP.org 27 May '08

27 May '08

<quote who="Kurt(a)OpenLDAP.org"> > > On May 26, 2008, at 12:48 PM, andrew.findlay(a)skills-1st.co.uk wrote: > >> On Mon, May 26, 2008 at 01:07:22PM +0100, Gavin Henry wrote: >> >>> I've patched security.sdf and I'm abotu to clean up some typos and >>> formating. >> >> Thanks. >> >>> Should we also mention that CRYPT is platform specific? >> >> I put in a note about the glibc2 version, but I was not aware of any >> platform oddities on the traditional 13-character version. Are there >> other MD5 variants? >> >> We could also mention the LANMAN scheme which I forgot to include as >> it is a compile-time option. >> >>> Lastly, should I also put: >>> >>> Portions Copyright 2008 Andrew Findlay <andrew.findlay(a)skills-1st.co.uk >>> > >>> >>> into doc/guide/COPYRIGHT ? > > No. > > This file is a copy of the main OpenLDAP Software COPYRIGHT file. > That file is intended to contain notices of copyright holders which > hold signfiicant rights in OpenLDAP Software. Notices for copyright > holders, as noted in the COPYRIGHT file, are generally to be placed in > the individual source files which the holder holds significant rights > in. That is, this notice belongs in security.sdf itself. I have so > updated the source documents. Good, this clears things up and make it clear going forward. Thanks. > Though the preface does clearly say that admin guide is part of > OpenLDAP Software, I've also updated the preface to note that portions > of the document may be copyright by others as indicated in source files. Thanks again.

1 0

Re: (ITS#5529) test048-syncrepl-multiproxy --without-threads hangs
by h.b.furuseth＠usit.uio.no 27 May '08

27 May '08

ando(a)sys-net.it writes: >h.b.furuseth(a)usit.uio.no wrote: >> RE24 configured with --quiet --without-threads --enable-ldap 'CFLAGS=-O0 -g' Sorry, that's RE24 + the ITS#5519 (--without-thread miscompiles) patch: cvs diff -r1.34 -r1.35 -I'[$]OpenLDAP[$:]' libraries/libldap_r/thr_stub.c > Probably, sync replication should not be allowed when building > --without-threads. Or there could be a warning that --without-threads breaks some features. Looking closer I see the server replicates from itself in this test. I can see how that'd be a problem in single-threaded mode... Most other syncrepl tests succeed. Though test050 crashes: daemon.c:979: slapd_set_read: Assertion `((slap_daemon.sd_index[(s)]) != -1)' failed. (Argh, it refuses to dump core for some reason. I've done ulimit -c unlimited and put it in defines.sh too.) -- Hallvard

1 0

Re: (ITS#5524) Additions to the Security section of the Admin Guide
by Kurt＠OpenLDAP.org 27 May '08

27 May '08

On May 26, 2008, at 12:48 PM, andrew.findlay(a)skills-1st.co.uk wrote: > On Mon, May 26, 2008 at 01:07:22PM +0100, Gavin Henry wrote: > >> I've patched security.sdf and I'm abotu to clean up some typos and >> formating. > > Thanks. > >> Should we also mention that CRYPT is platform specific? > > I put in a note about the glibc2 version, but I was not aware of any > platform oddities on the traditional 13-character version. Are there > other MD5 variants? > > We could also mention the LANMAN scheme which I forgot to include as > it is a compile-time option. > >> Lastly, should I also put: >> >> Portions Copyright 2008 Andrew Findlay <andrew.findlay(a)skills-1st.co.uk >> > >> >> into doc/guide/COPYRIGHT ? No. This file is a copy of the main OpenLDAP Software COPYRIGHT file. That file is intended to contain notices of copyright holders which hold signfiicant rights in OpenLDAP Software. Notices for copyright holders, as noted in the COPYRIGHT file, are generally to be placed in the individual source files which the holder holds significant rights in. That is, this notice belongs in security.sdf itself. I have so updated the source documents. Though the preface does clearly say that admin guide is part of OpenLDAP Software, I've also updated the preface to note that portions of the document may be copyright by others as indicated in source files. -- Kurt

1 0

(ITS#5532) incorrect timestamp of slapd replica log
by hai.zhao＠gmail.com 27 May '08

27 May '08

Full_Name: Zhao Hai Version: 2.3.41 OS: Linux 2.4.21 arm URL: ftp://ftp.openldap.org/incoming/zhaohai-080527.patch Submission from: (NULL) (205.209.140.4) Problem: race condition makes incorrect timestamp in replogfile, cause certain modification of entries not replicate to slurp slaves. replica: 180.0.10.2:1234 replica: 180.0.10.3:1234 time: 1211855467 ^^^^^^^^^^ this timestamp How to reproduce the problem: 1) run under very slow machines (my environ: arm 266MHz) 2) slapd is configed to generate replogfile 3) ldapadd about 5 entries, then ldapmodify 2 entries without delay. Patch: Please look the attached patch. It works for me.

1 0

Re: (ITS#5530) OpenLDAP fails to compile
by hyc＠symas.com 26 May '08

26 May '08

mike(a)tedder.cc wrote: > Full_Name: Mike Tedder > Version: 2.4.9 > OS: powerpc-linux > URL: ftp://ftp.openldap.org/incoming/ > Submission from: (NULL) (210.170.158.22) > > > The back-bdb backend for the OpenLDAP slapd server fails to compile when being > built using Berkeley DB 4.7: This was already reported in ITS#5523. -- -- 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

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