openldap-bugs December 2009

openldap-bugs@openldap.org

28 participants
133 discussions

Re: (ITS#6389) Allow setting syncrepl keepalive
by masarati＠aero.polimi.it 09 Dec '09

09 Dec '09

quanah(a)zimbra.com wrote: > When configuring a replica using syncrepl, it should be possible to have an > option that sets the keepalive used for the tcp connection between the replica > and master. Support for this was added to libldap, but no corresponding support > was added to the syncrepl stanza. I'd suggest a keyword like "keepalive" for > configuring this, based in seconds. A fix is in HEAD (it's been there for a while); please test. Thanks, p.

1 0

Re: (ITS#6037) errno in signal handlers
by hyc＠symas.com 08 Dec '09

08 Dec '09

h.b.furuseth(a)usit.uio.no wrote: > Full_Name: Hallvard B Furuseth > Version: HEAD, RE24 > OS: > URL: > Submission from: (NULL) (129.240.6.233) > Submitted by: hallvard > > > Signal handlers slapd/daemon.c:slap_sig_shutdown()/slap_sig_wake() need > to save/restore errno, since they call functions that can modify errno. > > And/or possibly they need sock_errno() - sock_errset() from ac/socket.h, > since they use tcp_write()? I don't know if that's safe inside signal > handlers, or if Windows itself saves/restores this code around signal > handlers. Windows doesn't really have signals in the first place. SIGTERM/SIGINT are simulated by the C runtime library, none of the other signals are really supported... All of the "support" is synchronous only - can only be invoked by raise(). I don't think there's any issue here. (An exception - I wrote a patch for the MSYS 1.11 runtime to propagate SIGINT/SIGTERM to other processes. The patch only works on win32, not win64 because I never got a decent win64 build environment working... But if you're using MSYS, it is possible to send these signals to a running slapd and have it shutdown cleanly. Otherwise, signals don't really exist for native Windows apps...) > slapd/main.c:wait4child() can loop forever after EINTR, since it checks > for errno even when the waitpid returns 0. I don't know if there is > a good reason it does that. Some of the commits have my name on them > but I think I copied it from elsewhere. > > > Anyway, I'm doing the minimal changes now: save/restore errno, and > set errno=0 in each loop iteration instead of before the loop. > Anyone with better ideas can take it from there. Good enough. Closing this ITS. -- -- 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#6423 slapo-rwm and dynlist overlay
by hyc＠symas.com 08 Dec '09

08 Dec '09

masarati(a)aero.polimi.it wrote: > The problem is that back-relay presents slapo-rwm(3) an entry with both > REP_ENTRY_MODIFIABLE and REP_ENTRY_MUSTRELEASE flags set. As a > consequence, the entry is released (namely, freed) before it is returned > by the frontend. As far as I understand, the two flags should be mutually > exclusive. As soon as one entry is marked as REP_ENTRY_MODIFIABLE, it is > supposed to be a copy of an entry that may need to be released, or an > entry created from scratch and thus not stored anywhere. I have a quick > fix to your problem, that consists in ignoring REP_ENTRY_MUSTRELEASE when > REP_ENTRY_MODIFIABLE is also set. However, a better fix would consist in > tracking where the issue occurs. Unless I track it, I'll commit the quick > fix. We already have ITS#5340 for issues with these flags... -- -- 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

ldapsearch with delay (on the same machine over TLS/SSL only)
by Pavel Lisy 08 Dec '09

08 Dec '09

Hello I've found one problem I have latest openldap release 2.4.3 (but it is similar on former versions too). There is small delay (1-2s) on beginning when I query ldap server from the same machine over TLS or SSL. Without SSL is no delay. Is it bug or is it normal? How can I debug it more? QUERY with delay: from ldap1 to ldap1 (the same machine) [root@ldap1 ~]# ldapsearch24 -Z -h ldap1.ldapnet.tmapy.cz -x -s sub -D "cn=Manager,dc=tmapy,dc=cz" -b 'dc=tmapy,dc=cz' -w p*****d QUERY without delay: from ldap1 to ldap1 (the same machine without TLS) [root@ldap1 ~]# ldapsearch24 -h ldap1.ldapnet.tmapy.cz -x -s sub -D "cn=Manager,dc=tmapy,dc=cz" -b 'dc=tmapy,dc=cz' -w p*****d from ldap1 to ldap2 [root@ldap1 ~]# ldapsearch24 -Z -h ldap2.ldapnet.tmapy.cz -x -s sub -D "cn=Manager,dc=tmapy,dc=cz" -b 'dc=tmapy,dc=cz' -w p*****d from ldap2 to ldap1 [root@ldap2 ~]# ldapsearch24 -Z -h ldap1.ldapnet.tmapy.cz -x -s sub -D "cn=Manager,dc=tmapy,dc=cz" -b 'dc=tmapy,dc=cz' -w p*****d I think my DNS/hosts settings are correct My testing environment: servers: ldap1.ldapnet.tmapy.cz ldap2.ldapnet.tmapy.cz ldap3.ldapnet.tmapy.cz from ldap1 config: slapd.conf TLSCACertificateFile /etc/pki/tls/cacert.pem TLSCertificateFile /etc/pki/tls/certs/ldap1.ldapnet.tmapy.cz-cert.pem TLSCertificateKeyFile /etc/pki/tls/private/ldap1.ldapnet.tmapy.cz-key.pem [root@ldap1 ~]# hostname ldap1.ldapnet.tmapy.cz [root@ldap1 ~]# grep ldap1 /etc/hosts 192.168.241.10 ldap1.ldapnet.tmapy.cz [root@ldap1 ~]# ifconfig eth0 eth0 inet adr:192.168.241.10 Všesměr:192.168.241.255 Maska:255.255.255.0 [root@ldap1 ~]# grep ldap1 /etc/pki/tls/certs/ldap1.ldapnet.tmapy.cz-cert.pem Subject: C=CS, ST=Kraj kralovehradecky, O=T-MAPY spol. s r.o.,OU=ldapnet, CN=ldap1.ldapnet.tmapy.cz Pavel Lisy -- Pavel Lisy <pali(a)tmapy.cz>

3 4

Re: (ITS#6425) Uninitialized var in syncrepl
by h.b.furuseth＠usit.uio.no 08 Dec '09

08 Dec '09

h.b.furuseth(a)usit.uio.no writes: > In syncrepl.c lines 1001-1010: Also in equivalent code around line 820. -- Hallvard

1 0

(ITS#6425) Uninitialized var in syncrepl
by h.b.furuseth＠usit.uio.no 08 Dec '09

08 Dec '09

Full_Name: Hallvard B Furuseth Version: HEAD, 2.4.20 OS: Linux URL: Submission from: (NULL) (129.240.6.233) Submitted by: hallvard In syncrepl.c lines 1001-1010: LDAPControl **next; rctrlp = ldap_control_find( LDAP_CONTROL_SYNC_DONE, rctrls, &next ); if ( next && ldap_control_find( LDAP_CONTROL_SYNC_DONE, next, NULL ) ) 'next' in the last line is uninitalized if 2nd line failed. Seen in testrun/srv2/valgrind.* with './run -b null test050' and and slapd under valgrind.

1 0

ITS#6423 slapo-collect & slapo-valsort could suffer from the same issue
by masarati＠aero.polimi.it 08 Dec '09

08 Dec '09

This should now be fixed in HEAD, collect.c 1.17 <- 1.16 valsort.c 1.27 <- 1.26 please review and test. p.

1 0

Re: ITS#6423 slapo-rwm and dynlist overlay
by masarati＠aero.polimi.it 08 Dec '09

08 Dec '09

Fixed in HEAD; the fix is in overlays/dynlist.c 1.68->1.69 Please test and report. Thanks, p.

1 0

(ITS#6424) slapd-meta pseudoroot* directives deprecated
by masarati＠aero.polimi.it 08 Dec '09

08 Dec '09

Full_Name: Pierangelo Masarati Version: HEAD/re24 OS: irrelevant URL: ftp://ftp.openldap.org/incoming/ Submission from: (NULL) (93.149.38.247) Submitted by: ando The man page erroneously reports them. A fix is coming. See also ITS#6418. p.

1 0

ITS#6423 slapo-rwm and dynlist overlay
by masarati＠aero.polimi.it 08 Dec '09

08 Dec '09

The problem is that back-relay presents slapo-rwm(3) an entry with both REP_ENTRY_MODIFIABLE and REP_ENTRY_MUSTRELEASE flags set. As a consequence, the entry is released (namely, freed) before it is returned by the frontend. As far as I understand, the two flags should be mutually exclusive. As soon as one entry is marked as REP_ENTRY_MODIFIABLE, it is supposed to be a copy of an entry that may need to be released, or an entry created from scratch and thus not stored anywhere. I have a quick fix to your problem, that consists in ignoring REP_ENTRY_MUSTRELEASE when REP_ENTRY_MODIFIABLE is also set. However, a better fix would consist in tracking where the issue occurs. Unless I track it, I'll commit the quick fix. p.

1 0

← Newer
1
...
5
6
7
8
9
10
11
...
14
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

openldap-bugs December 2009