openldap-bugs

openldap-bugs@openldap.org

1 participants
21001 discussions

Re: (ITS#6907) openldap 2.4.25 segfaults when using apache directory studio to rename objects
by catchall＠bromberger.com 13 Apr '11

13 Apr '11

I don't know whether I did it correctly, but here's my first attempt at a gdb backtrace. Note that it keeps going - I went to #7500 before I gave up - it was just repeating, e.g., lines 5-7. Does this help? What else should I be providing? Seth. Starting program: /usr/local/libexec/slapd -d 256 -h "ldaps:///" -u openldap -g openldap -f /usr/local/etc/openldap/slapd.conf [Thread debugging using libthread_db enabled] ... conn=1000 op=4 MODRDN dn="uid=myuid,ou=Testing,dc=example,dc=org" Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x7ffff4901700 (LWP 12020)] 0x000000000049af05 in ?? () (gdb) backtrace #0 0x000000000049af05 in ?? () #1 0x00000000004421e3 in backend_check_restrictions () #2 0x0000000000437220 in fe_op_search () #3 0x000000000049a577 in overlay_op_walk () #4 0x000000000049afb8 in ?? () #5 0x00000000004373f1 in fe_op_search () #6 0x000000000049a577 in overlay_op_walk () #7 0x000000000049afb8 in ?? () #8 0x00000000004373f1 in fe_op_search () #9 0x000000000049a577 in overlay_op_walk () #10 0x000000000049afb8 in ?? () #11 0x00000000004373f1 in fe_op_search () #12 0x000000000049a577 in overlay_op_walk () #13 0x000000000049afb8 in ?? () #14 0x00000000004373f1 in fe_op_search () #15 0x000000000049a577 in overlay_op_walk () #16 0x000000000049afb8 in ?? () #17 0x00000000004373f1 in fe_op_search () #18 0x000000000049a577 in overlay_op_walk () #19 0x000000000049afb8 in ?? () #20 0x00000000004373f1 in fe_op_search () #21 0x000000000049a577 in overlay_op_walk () #22 0x000000000049afb8 in ?? () #23 0x00000000004373f1 in fe_op_search () #24 0x000000000049a577 in overlay_op_walk () #25 0x000000000049afb8 in ?? () #26 0x00000000004373f1 in fe_op_search () #27 0x000000000049a577 in overlay_op_walk () #28 0x000000000049afb8 in ?? () #29 0x00000000004373f1 in fe_op_search () #30 0x000000000049a577 in overlay_op_walk () #31 0x000000000049afb8 in ?? () #32 0x00000000004373f1 in fe_op_search () #33 0x000000000049a577 in overlay_op_walk () #34 0x000000000049afb8 in ?? () #35 0x00000000004373f1 in fe_op_search () #36 0x000000000049a577 in overlay_op_walk () #37 0x000000000049afb8 in ?? () #38 0x00000000004373f1 in fe_op_search () #39 0x000000000049a577 in overlay_op_walk () #40 0x000000000049afb8 in ?? () #41 0x00000000004373f1 in fe_op_search ()

1 0

Re: (ITS#6907) openldap 2.4.25 segfaults when using apache directory studio to rename objects
by quanah＠zimbra.com 13 Apr '11

13 Apr '11

--On Wednesday, April 13, 2011 12:26 AM +0000 openldap(a)bromberger.com wrote: > Full_Name: Seth > Version: 2.4.25 > OS: Linux (Ubuntu) 2.6.35.4 > URL: ftp://ftp.openldap.org/incoming/ > Submission from: (NULL) (173.13.140.242) > > > Using Apache Directory Studio to interface with hdb-based openldap server > via TLS. Moving a DN from one ou to another crashes slapd. The move (the > first one if there are multiples) appears to work before slapd segfaults. > An example in the logfile: > > > Apr 12 17:17:52 ldap slapd[11278]: conn=1000 op=13 MODRDN > dn="uid=someuserid,ou=Organizations,dc=example,dc=org" > Apr 12 17:17:52 ldap kernel: [974579.184525] slapd[11281]: segfault at > 7f0bdfab7ff8 ip 000000000049af05 sp 00007f0bdfab8000 error 6 in > slapd[400000+19e000] > > This is 100% reproducible. Also affects 2.4.23-0ubuntu3.5 (ubuntu > package) - I upgraded from source to try to work around this bug. > Please provide a gdb backtrace. --Quanah -- Quanah Gibson-Mount Sr. Member of Technical Staff Zimbra, Inc A Division of VMware, Inc. -------------------- Zimbra :: the leader in open source messaging and collaboration

1 0

Re: (ITS#6789)
by sgallagh＠redhat.com 13 Apr '11

13 Apr '11

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Has there been any progress on this issue? It causes a lot of headaches trying to debug connection issues in SSSD. - -- Stephen Gallagher RHCE 804006346421761 Delivering value year after year. Red Hat ranks #1 in value among software vendors. http://www.redhat.com/promo/vendor/ -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/ iEYEARECAAYFAk2lsTIACgkQeiVVYja6o6Nh7wCdF5Sn/IsMpNH9gB8WA9A7tEgB ctUAoKPTUIL1rfZPnArrNBmAxQ/VuEq7 =DxDR -----END PGP SIGNATURE-----

1 0

(ITS#6908) Query
by preeti＠omnimessaging.com 13 Apr '11

13 Apr '11

Full_Name: Preeti Version: 2.3.32 OS: Windows XP URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (114.143.122.119) My OpenLdap Client file "ldapmodify" is missing. Can you please provide me with the "openldap-client-2.3.32.tgz" file.

1 0

(ITS#6907) openldap 2.4.25 segfaults when using apache directory studio to rename objects
by openldap＠bromberger.com 12 Apr '11

12 Apr '11

Full_Name: Seth Version: 2.4.25 OS: Linux (Ubuntu) 2.6.35.4 URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (173.13.140.242) Using Apache Directory Studio to interface with hdb-based openldap server via TLS. Moving a DN from one ou to another crashes slapd. The move (the first one if there are multiples) appears to work before slapd segfaults. An example in the logfile: Apr 12 17:17:52 ldap slapd[11278]: conn=1000 op=13 MODRDN dn="uid=someuserid,ou=Organizations,dc=example,dc=org" Apr 12 17:17:52 ldap kernel: [974579.184525] slapd[11281]: segfault at 7f0bdfab7ff8 ip 000000000049af05 sp 00007f0bdfab8000 error 6 in slapd[400000+19e000] This is 100% reproducible. Also affects 2.4.23-0ubuntu3.5 (ubuntu package) - I upgraded from source to try to work around this bug.

1 0

(ITS#6906) fix Admin guide cachesize calculations to remove refence to Hash-format
by Tim.Mooney＠ndsu.edu 12 Apr '11

12 Apr '11

Full_Name: Tim Mooney Version: 2.4.25 OS: Linux, Solaris URL: ftp://ftp.openldap.org/incoming/TimMooney-110412-04.patch Submission from: (NULL) (2001:4930:106:0:18bb:1140:fa3d:f713) As I mentioned here: http://www.openldap.org/lists/openldap-technical/201104/msg00088.html and Quanah clarified here: http://www.openldap.org/lists/openldap-technical/201104/msg00123.html The admin guide chapter on tuning still mentions that indexes are in Hash format, though that's no longer true. I've uploaded patch TimMooney-110412-04.patch against HEAD that clarifies that indexes are now also in B-tree format and so the same calculations for cache on dn2id can be used with each index. I also corrected one other typo, changing To avoid thrashing the, your cache to To avoid thrashing, your cache

1 0

(ITS#6904) correct the recommended backend verbiage in slapd.backends(5)
by Tim.Mooney＠ndsu.edu 12 Apr '11

12 Apr '11

I've uploaded an additional patch, TimMooney-110412-03.patch, for slapd-bdb.5/slapd-hdb.5, that switches the "recommended" status from bdb to hdb. This should likely be applied in addition to the patch mentioned in the initial report.

1 0

(ITS#6905) fix up intros to chapters 5 and 6
by Tim.Mooney＠ndsu.edu 12 Apr '11

12 Apr '11

Full_Name: Tim Mooney Version: 2.4.25 OS: Linux, Solaris URL: ftp://ftp.openldap.org/incoming/TimMooney-110412-02.patch Submission from: (NULL) (2001:4930:106:0:18bb:1140:fa3d:f713) As I mentioned here http://www.openldap.org/lists/openldap-technical/201104/msg00088.html the LDAP admin guide is very good, but I think that the intros to chapters 5 (configuring slapd via slapd-config) and chapter 6 (configuring slapd via slapd.conf) needed a bit of updating. Specifically - it's not very explicit that slapd-config is now preferred and slapd.conf should only be used as a last resort. - the Note that some backends don't support slapd-config was vague enough that it would prevent many people from choosing that route, because they wouldn't know whether a backend they needed would support the new-style config system. By specifically listing which backends don't support slapd-config, hopefully more people will be willing to begin with (or convert to) slapd-config. I also tried to - make it clear that chapters 5 and 6 were essentially explaining the same thing, but the slapd-config way is preferred. - Although slapd-config stores its configuration in text files, they shouldn't be edited manually. The patch against head has been uploaded. Any and all comments and feedback welcome.

1 0

Re: (ITS#6891)
by tgates81＠gmail.com 12 Apr '11

12 Apr '11

Hi Quanah, slapd does have access to the files because it re-created after I deleted them on start. root@directory-proxy2:/var/lib/ldap# ls -la total 11752 drwxr-x--- 2 openldap openldap 4096 2011-04-12 16:01 . drwxr-xr-x 44 root root 4096 2011-03-25 12:47 .. -rw-r--r-- 1 openldap openldap 4096 2011-04-12 16:01 alock -rw------- 1 openldap openldap 24576 2011-04-12 16:01 __db.001 -rw------- 1 openldap openldap 843776 2011-04-12 16:01 __db.002 -rw------- 1 openldap openldap 13115392 2011-04-12 16:01 __db.003 -rw------- 1 openldap openldap 2359296 2011-04-12 16:01 __db.004 -rw------- 1 openldap openldap 540672 2011-04-12 16:01 __db.005 -rw------- 1 openldap openldap 32768 2011-04-12 16:01 __db.006 -rw-r--r-- 1 openldap openldap 120 2011-04-12 14:29 DB_CONFIG -rw------- 1 openldap openldap 8192 2011-04-12 14:29 dn2id.bdb -rw------- 1 openldap openldap 32768 2011-04-12 14:29 id2entry.bdb -rw------- 1 openldap openldap 10485760 2011-04-12 16:01 log.0000000001 root@directory-proxy2:/var/lib/ldap# ps aux |grep slapd openldap 5519 0.0 0.6 56244 6316 ? Ssl 16:01 0:00 /usr/sbin/slapd -h ldap:/// ldapi:/// ldaps:/// -g openldap -u openldap -F /etc/ldap/slapd.d I've disabled apparmor as well.. On Tue, Apr 12, 2011 at 3:42 PM, Quanah Gibson-Mount <quanah(a)zimbra.com> wrote: > --On Tuesday, April 12, 2011 3:04 PM -0400 Tyler Gates <tgates81(a)gmail.com> > wrote: > >> I can't get a back trace because it doesn't crash under gdb, slapd >> only responds to queries with error: >> >> client: >> result: 80 Other (e.g., implementation specific) error >> text: internal error >> server: >> bdb(dc=example,dc=com): PANIC: fatal region error detected; run recovery >> >> Apr 12 14:25:34 directory-proxy2 slapd[2526]: backend_startup_one >> (type=hdb, suffix="(null)"): bi_db_open failed! (-1) > > These all indicate permissions issues. It sounds like the slapd instance > does not have permissions to access the database. > > --Quanah > > > -- > > Quanah Gibson-Mount > Sr. Member of Technical Staff > Zimbra, Inc > A Division of VMware, Inc. > -------------------- > Zimbra :: the leader in open source messaging and collaboration >

1 0

Re: (ITS#6891)
by quanah＠zimbra.com 12 Apr '11

12 Apr '11

--On Tuesday, April 12, 2011 3:04 PM -0400 Tyler Gates <tgates81(a)gmail.com> wrote: > I can't get a back trace because it doesn't crash under gdb, slapd > only responds to queries with error: > > client: > result: 80 Other (e.g., implementation specific) error > text: internal error > server: > bdb(dc=example,dc=com): PANIC: fatal region error detected; run recovery > > Apr 12 14:25:34 directory-proxy2 slapd[2526]: backend_startup_one > (type=hdb, suffix="(null)"): bi_db_open failed! (-1) These all indicate permissions issues. It sounds like the slapd instance does not have permissions to access the database. --Quanah -- Quanah Gibson-Mount Sr. Member of Technical Staff Zimbra, Inc A Division of VMware, Inc. -------------------- Zimbra :: the leader in open source messaging and collaboration

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