openldap-devel

openldap-devel@openldap.org

1418 discussions

ITS#6152 pcache enhancements
by Howard Chu 18 Aug '09

18 Aug '09

Just starting to look at implementing Time-To-Refresh, and wondering how it impacts other parameters. proxysavequeries allows queries to be saved and reloaded across restarts. Currently it will only reload non-expired queries. The implication of setting the offline mode is that expired entries may remain in the cache. Should they in fact be reloadable too? The proxytemplate directive has gotten a bit cumbersome, with optional negative TTL and sizelimit TTL added on. This would be the natural place to add the TTR config as well. Should there be independent TTR for negative and sizelimit too? (And it seems we should move from positional arguments to named arguments here...) Given the existence of a TTR feature, I'm less concerned about the proxysavequeries behavior. One should simply set a long TTL and set TTR as needed to keep the cache fresh. As a side point, I'd prefer more consistency in the config keyword names, we should have used "pcache" as the prefix for all the keywords... pcache <database> ... pcacheattrset <index> ... pcachemaxqueries <queries> pcachevalidate <bool> pcachepersist <bool> pcacheset template=<string> attrset=<index> ttl=<secs> [negttl=<secs>] ... pcacheposition (head|tail) -- -- 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 2

Re: commit: ldap/libraries/libldap ldap-int.h util-int.c
by Hallvard B Furuseth 18 Aug '09

18 Aug '09

ando(a)OpenLDAP.org writes: > util-int.c 1.61 -> 1.62 You are returning the static pointer from gmtime(), not the provided result pointer or NULL. -- Hallvard

1 0

Re: HAVE_GMTIME_R?
by masarati＠aero.polimi.it 18 Aug '09

18 Aug '09

As far as I can tell (I cannot tell), the only issue should be with the WIN32 version of lutil_gettime(); the modifications I'm finalizing get rid of all mutexes except for directly 'round gmtime(3), but I don't know whether the WIN32 function GetSystemTime( &st ); is reentrant or not. I bet it is, since it works on a buffer passed by the caller. p.

1 0

Re: commit: ldap/servers/slapd proto-slap.h syncrepl.c
by masarati＠aero.polimi.it 17 Aug '09

17 Aug '09

> Update of /repo/OpenLDAP/pkg/ldap/servers/slapd > > Modified Files: > proto-slap.h 1.789 -> 1.790 > syncrepl.c 1.474 -> 1.475 > > Log Message: > For ITS#6152 add slapd_str2scope(), slapd_scope2bv() These were already available as ldap_pvt_* functions. p.

2 1

Re: commit: ldap/servers/slapd/overlays pcache.c
by Howard Chu 17 Aug '09

17 Aug '09

ando(a)OpenLDAP.org wrote: > Update of /repo/OpenLDAP/pkg/ldap/servers/slapd/overlays > > Modified Files: > pcache.c 1.184 -> 1.185 > > Log Message: > expose cachedQueryURL in overlay's monitor entry (partial fulfilment of ITS#5178; needs new register_overlay() code in back-monitor) Hmmm, not sure exposing the actual URLs is a good idea. Why not just the counts? Also, if there are thousands of cached queries, this could get to be an expensive lookup. -- -- Howard Chu CTO, Symas Corp. http://www.symas.com Director, Highland Sun http://highlandsun.com/hyc/ Chief Architect, OpenLDAP http://www.openldap.org/project/

2 1

Re: commit: ldap/servers/slapd/overlays pcache.c
by masarati＠aero.polimi.it 17 Aug '09

17 Aug '09

> Update of /repo/OpenLDAP/pkg/ldap/servers/slapd/overlays > > Modified Files: > pcache.c 1.183 -> 1.184 > > Log Message: > For refresh_purge, only delete entry if ours is the last queryid on it. > Otherwise just remove our queryid. Howard, I have another round of monitor-related mods to pcache. I could successfully merge your mods with mine. Do you mind if I commit, or do you have more to sync? p.

2 1

Re: commit: ldap/servers/slapd/overlays pcache.c
by Howard Chu 17 Aug '09

17 Aug '09

hyc(a)OpenLDAP.org wrote: > Update of /repo/OpenLDAP/pkg/ldap/servers/slapd/overlays > > Modified Files: > pcache.c 1.177 -> 1.178 > > Log Message: > ITS#6251 framework for TTR. refresh not yet implemented. At the moment the idea for implementing refresh is to refactor syncrepl_entry() so that we can reuse the bulk of it. (Mainly the code that generates an appropriate Modify request based on the difference between the local entry and the remote entry.) Detecting renames will be trickier since the cache typically won't have any entryUUIDs. I think we'll simply have to treat it as a Delete / Add... -- -- Howard Chu CTO, Symas Corp. http://www.symas.com Director, Highland Sun http://highlandsun.com/hyc/ Chief Architect, OpenLDAP http://www.openldap.org/project/

3 7

Re: RE24 testing for 2.4.18 (round 2)
by William Jojo 17 Aug '09

17 Aug '09

---- Original message ---- >Date: Fri, 14 Aug 2009 14:08:23 -0700 >From: Quanah Gibson-Mount <quanah(a)zimbra.com> >Subject: Re: RE24 testing for 2.4.18 (round 2) >To: openldap-devel(a)openldap.org > >--On Wednesday, August 12, 2009 5:59 PM -0700 Quanah Gibson-Mount ><quanah(a)zimbra.com> wrote: > >> Please test RE24. > > >A few more fixes, mainly back-meta, back-monitor, and slapo-translucent. > I have the following in my libraries/libldap/os-ip.c: #ifdef TCP_KEEPCNT <<<<<<< os-ip.c if ( setsockopt( s, SOL_SOCKET, TCP_KEEPCNT, ======= if ( setsockopt( s, IPPROTO_TCP, TCP_KEEPCNT, >>>>>>> 1.118.2.19 I have no idea how others are posting successes... Bill >--Quanah > >-- > >Quanah Gibson-Mount >Principal Software Engineer >Zimbra, Inc >-------------------- >Zimbra :: the leader in open source messaging and collaboration

3 2

FAQ and ancient entries
by Quanah Gibson-Mount 15 Aug '09

15 Aug '09

In looking at our FAQ, many entries detail ancient (2.1 and earlier) information that is completely obsolete. It would be nice to be able to archive off the cruft so that the site is actually of more use. Thoughts on how best to achieve this? Some examples: SLAPD (with LDBM database) prerequisites include suitable database support. One of following packages can fulfill the prerequisite. Berkeley Database from Sleepycat (http://www.sleepycat.com/) GNU Database Manager (GDBM) from FSF (http://www.gnu.org/) For OpenLDAP 2.1 (or later), Berkeley DB 4.1 is recommended. SLURPD prerequisites include: Thread support (Posix Threads and others) SSL/TLS prerequisites: OpenSSL 0.9.6 from http://www.openssl.org The OpenSSL library usually needs to be patched before it will work correctly. See this FAQ article for details: http://www.openldap.org/faq/index.cgi?file=185 (Note about OpenSSL and crypt()) (Answer) How do I select a suitable LDBM backend? (Answer) How do I force LDBM to use of BerkeleyDB? (Answer) How do I force LDBM to use of GDBM? (Answer) Can LDBM use NDBM? etc. --Quanah -- Quanah Gibson-Mount Principal Software Engineer Zimbra, Inc -------------------- Zimbra :: the leader in open source messaging and collaboration

5 7

Normalize() API: syntax of attribute value vs. syntax of asserted value?
by masarati＠aero.polimi.it 14 Aug '09

14 Aug '09

The Normalize() API (slap_mr_normalize_func) receives a Syntax* pointing to the syntax per which it is normalizing. However, when used in conjunction with MRA filters, it might be useful to know what syntax the value that is being normalized conforms to, rather than for what syntax normalization is taking place. Note that what syntax the normalizer is normalizing for is somehow already determined by the specific normalization function that is in use. Also note that when the normalization of the asserted value occurs inside an MRA filter, the normalizer receives a pointer to the original syntax of the attribute rather than of the syntax it is normalizing for! I think an easy solution would be to swap the behavior: pass the asserted syntax when normalizing the asserted value, and the attribute's syntax when normalizing the attribute value. A better solution would be to extend the API in order to pass both. p.

6 6

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

openldap-devel