openldap-bugs September 2008

openldap-bugs@openldap.org

39 participants
206 discussions

(ITS#5678) slurpd stops processing the replication log if consecutive blank lines are encountered
by mgwin＠escp-eap.net 03 Sep '08

03 Sep '08

Full_Name: Michael Gwin Version: 2.3.30 OS: linux (debian etch x86_64) URL: Submission from: (NULL) (195.85.247.248) Hi, Replication of our ldap database often stops working, with pending changes stacking up and not getting processed by slurpd. /var/lib/ldap/replog (as configured in slapd.conf by the "replog" directive) fills up with entries, and slurpd doesn't pick up on them. Whilst trying to debug the issue, I noticed there were consecutive blank lines seperating entries in the replication log (both /var/lib/ldap/replog and /var/spool/slurpd/replica/slurpd.replog). Running slurpd in one shot mode on either of these files would result in any entries before the consecutive blank lines being processed, but not the entries following them. After reducing each set of consecutive blank lines to one blank line, slurpd processes all entries correctly. The replicas are set up in slapd.conf as follows: replogfile /var/lib/ldap/replog replica uri=ldap://host1.domain.tld:389 starttls=critical bindmethod=simple binddn="cn=replicationuser,o=MYORG" credentials=XYZ replica uri=ldaps://host2.domain.tld:636 bindmethod=simple binddn="cn=replicationuser,o=MYORG" credentials=XYZ attrs=account,MYORGGroup,MYORGPerson,\ organization,organizationalRole,\ organizationalUnit,posixAccount,posixGroup,\ simpleSecurityObject,top NB. - the "attrs" directive of the second replica is one line in slapd.conf. - some of the (confidential) values have been changed. version info: - openldap 2.3.30 on all hosts (debian package: slapd-2.3.30-5+etch1) - all hosts are running debian etch x86_64 Not all entries are seperated by multiple blank lines. I don't know if it's ok in theory for slapd to write multiple blank lines to /var/lib/ldap/replog, but slurpd should handle them properly, or if slapd shouldn't be writing them at all. How to reproduce: Insert consecutive blank lines between two entries in a replication log, and run slurpd in one-shot mode on it. Processing stops when the consecutive blank lines are encountered.

1 0

Re: (ITS#5661) contextCSN gets corrupted on the stand by mirror
by ando＠sys-net.it 02 Sep '08

02 Sep '08

ali.pouya(a)free.fr wrote: > I made a new directory with only one contextCSN (SID=002) as you recommended, > and reproduced the contextCSN corruption problem several times. > > Example1 : > contextCSN:: 0L0NojA5MDIxMjU5NDkuNzMwMjg1WiMwMDAwMDAjMDAyIzAwMDAwMA== > > The four corrupted bytes at the beginning are : D0 BD 02 A2 (hex) > > Example2 : > contextCSN:: 4I54oTA5MDIxNTE5MTYuMjYzNDIxWiMwMDAwMDAjMDAyIzAwMDAwMA== > > The four corrupted bytes at the beginning are : E0 8E 78 A1 (hex) > > > I insist on the fact that the problem heppens ONLY if I use TWO syncrepl > directives as recommended in the Admin Guide. > If I use only ONE syncrepl directive, I don't reproduce the problem and the > mirrors get synchronized correctly (whichever mirror side I use for writing). > Also the problem happens on the stand by mirror only when therer are write > operations on the active mirror (> 1000 writes per minute). > > I do not understand the interest of using TWO syncrepl directives for > mirrormode. Well, going back to your initial posting, I think you are somehow correct. Rather than not seeing the point of having two syncrepl statements (of which only one is supposed to be active), I see it as an inconsistent and potentially dangerous configuration. In fact, the only advantage of having two syncrepl statements is related to being able to share the same configuration among two symmetric servers (mirror mode, multimaster, ...), using the serverID directive to determine what is the "right" one. But in that case, you'd need to have multiple serverID directives as well, with the URI field set. I set up a test system with your configuration, and loaded it very heavily, while running the server that's supposed to screw up under valgrind. I haven't seen any issue yet, though. 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 Fax: +39 0382 476497 Email: ando(a)sys-net.it -----------------------------------

1 0

(ITS#5676) Possible dereferencing of uninitialized pointer in bdb_entry_get()
by ando＠sys-net.it 02 Sep '08

02 Sep '08

Full_Name: Pierangelo Masarati Version: HEAD/re24 OS: irrelevant URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (81.72.89.40) Submitted by: ando In bdb_entry_get(), if oex->oe_key == bdb then boi is NULL; however, it could be dereferenced few lines later (it happened to me under heavy load). A fix is coming. p.

1 0

(ITS#5675) Incorrect contextCSN detection during consumer initialization?
by ando＠sys-net.it 02 Sep '08

02 Sep '08

Full_Name: Pierangelo Masarati Version: HEAD (earlier?) OS: irrelevant URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (81.72.89.40) Submitted by: ando See <http://www.openldap.org/lists/openldap-bugs/200809/msg00012.html> and, significantly, its follow-up. The most recent contextCSN should be used instead of just the first one. p.

1 0

Re: (ITS#5673) Replication delete problem
by miguel.jinez＠gmail.com 02 Sep '08

02 Sep '08

------=_Part_7335_29974836.1220369704755 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline I was making a test deleting 8600 users in my ldap DIT, but I think meanwhile Master A perform the actions the synchronization doesn't respect the hierachy and deletes fathers but not his sons. Why I said that, because I try to upload again the users and in some cases they alredy exist Migue 2008/9/1 <ghenry(a)openldap.org> > Miguel Jinez wrote: > > It is no the same problem, is a new problem > > On list please. They are related. > > > -- > Kind Regards, > > Gavin Henry. > OpenLDAP Engineering Team. > > E ghenry(a)OpenLDAP.org > > Community developed LDAP software. > > http://www.openldap.org/project/ > > > ------=_Part_7335_29974836.1220369704755 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline <div dir="ltr">I was making a test deleting 8600 users in my ldap DIT, but I think meanwhile Master A perform the actions the synchronization doesn't  respect the hierachy and deletes fathers but not his sons.<br>Why I said that, because I try to upload again the users and in some cases they alredy exist<br> <br>Migue<br><br><div class="gmail_quote">2008/9/1 <span dir="ltr"><<a href="mailto:ghenry@openldap.org">ghenry(a)openldap.org</a>></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> <div><div></div><div class="Wj3C7c">Miguel Jinez wrote:<br> > It is no the same problem, is a new problem<br> <br> On list please. They are related.<br> <br> <br> --<br> Kind Regards,<br> <br> Gavin Henry.<br> OpenLDAP Engineering Team.<br> <br> E ghenry(a)OpenLDAP.org<br> <br> Community developed LDAP software.<br> <br> <a href="http://www.openldap.org/project/" target="_blank">http://www.openldap.org/project/</a><br> <br> <br> </div></div></blockquote></div><br></div> ------=_Part_7335_29974836.1220369704755--

1 0

Re: (ITS#5661) contextCSN gets corrupted on the stand by mirror
by ali.pouya＠free.fr 02 Sep '08

02 Sep '08

Pierangelo mazarati wrote : > Can you try a fresh reload of the database(s) stripping out the entryCSN > and letting slapadd generate them, using the -S <SID> switch (along with > the -w switch), in order to enforce a SID of 001 (or 002, as you like)? Hi Pierangelo, I made a new directory with only one contextCSN (SID=002) as you recommended, and reproduced the contextCSN corruption problem several times. Example1 : contextCSN:: 0L0NojA5MDIxMjU5NDkuNzMwMjg1WiMwMDAwMDAjMDAyIzAwMDAwMA== The four corrupted bytes at the beginning are : D0 BD 02 A2 (hex) Example2 : contextCSN:: 4I54oTA5MDIxNTE5MTYuMjYzNDIxWiMwMDAwMDAjMDAyIzAwMDAwMA== The four corrupted bytes at the beginning are : E0 8E 78 A1 (hex) I insist on the fact that the problem heppens ONLY if I use TWO syncrepl directives as recommended in the Admin Guide. If I use only ONE syncrepl directive, I don't reproduce the problem and the mirrors get synchronized correctly (whichever mirror side I use for writing). Also the problem happens on the stand by mirror only when therer are write operations on the active mirror (> 1000 writes per minute). I do not understand the interest of using TWO syncrepl directives for mirrormode. Thanks for your help Best Regards Ali

1 0

Re: (ITS#5545) Inconsistency in multi-master setup
by ST＠itsc.cuhk.edu.hk 02 Sep '08

02 Sep '08

Oops, I tested on both nodes, which is not supposed to do so in mirrormode, right? Sorry for the trouble caused. Thanks.

1 0

Re: (ITS#5672) Build of back-ndb fails - back-ndb.h:26:22: error: NdbApi.hpp: No such file or directory
by hyc＠symas.com 01 Sep '08

01 Sep '08

Pierangelo Masarati wrote: > hyc(a)symas.com wrote: >> hyc(a)symas.com wrote: >>> ando(a)sys-net.it wrote: >>>> michael(a)stroeder.com wrote: >>>>> Full_Name: Michael Ströder >>>>> Version: HEAD >>>>> OS: Linux >>>>> URL: ftp://ftp.openldap.org/incoming/ >>>>> Submission from: (NULL) (84.163.97.73) >>>>> I think this says it all: >>>>> >>>>> back-ndb.h:26:22: error: NdbApi.hpp: No such file or directory >>>>> >>>>> I could provide the complete build output if needed. >>> As such, this section Works As Designed. ITS will be closed... > > Well, the reliability of expecting things to be where people document > them is the reason autotools were created in the first place... > however, I'm having other problems right now: NdbInterpretedCode.hpp > isn't included, so all occurrences of NdbInterpretedCode cause an error; > but when I explicitly include that header, it requires some other header > that is not available... Apparently, I'm using the wrong version, but > this seems to indicate that the whole Ndb API needs to stabilize. Ah yes, that's true. We had to use a number of APIs that weren't documented when development began, and also a number of APIs were added for us. So yes, you definitely need a very recent release. -- -- 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: N-way multimaster replication problem (ITS#5617)
by ghenry＠OpenLDAP.org 01 Sep '08

01 Sep '08

Miguel Jinez wrote: > I have SERVERID 1, SERVERID 2 and SERVERID 3 for each master configuration > On list please. See below. > > > can you try Multi-Master and not MirrorMode: > > http://www.openldap.org/doc/admin24/replication.html#N-Way%20Multi-Master > > You don't have: > > serverID 1 LDAP_URI > -- Kind Regards, Gavin Henry. OpenLDAP Engineering Team. E ghenry(a)OpenLDAP.org Community developed LDAP software. http://www.openldap.org/project/

1 0

Re: (ITS#5673) Replication delete problem
by ghenry＠OpenLDAP.org 01 Sep '08

01 Sep '08

Miguel Jinez wrote: > It is no the same problem, is a new problem On list please. They are related. -- Kind Regards, Gavin Henry. OpenLDAP Engineering Team. E ghenry(a)OpenLDAP.org Community developed LDAP software. http://www.openldap.org/project/

1 0

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

openldap-bugs September 2008