Re: (ITS#6357) Problem configuring openldap
by quanah＠zimbra.com
--On Friday, October 30, 2009 11:34 PM +0000 pierre42d(a)9online.fr wrote:
I suggest you read the README file shipped with the OpenLDAP release.
--Quanah
--
Quanah Gibson-Mount
Principal Software Engineer
Zimbra, Inc
--------------------
Zimbra :: the leader in open source messaging and collaboration
13 years, 11 months
Re: (ITS#5810) $BACKEND selections in make test
by h.b.furuseth＠usit.uio.no
Maybe it's about time to close this one...
Could someone test if NDB likes this acfilter.sh instead of the current
one? In the tests where NDB works, anyway. I don't know which ones.
The current acfilter depends on awk being GNU awk, which can be wrong
even on Linux. Ubuntu has awk = mawk. This variant uses perl instead.
back-ldif passes most tests with this acfilter and a few additions of
BACKEND2SORT="ldif=e".
#! /bin/sh
# $OpenLDAP: pkg/ldap/tests/scripts/acfilter.sh,v 1.15 2009/01/21 23:40:44 kurt Exp $
## This work is part of OpenLDAP Software <http://www.openldap.org/>.
##
## Copyright 1998-2009 The OpenLDAP Foundation.
## All rights reserved.
##
## Redistribution and use in source and binary forms, with or without
## modification, are permitted only as authorized by the OpenLDAP
## Public License.
##
## A copy of this license is available in the file LICENSE in the
## top-level directory of the distribution or, alternatively, at
## <http://www.OpenLDAP.org/license.html>.
# Strip LDIF comments, or strip everything when $BACKEND=null.
# Optionally sort attrs/entries according to $BACKEND2SORT:
# BACKEND2SORT="<backend>=[a][e],<backend>=[a][e],..." says whether
# to sort attributes in an entry, and whether to sort entries.
# 2+ empty input lines separate each list of entries to be sorted.
# This feature needs Perl, otherwise any sorting is just by lines.
# Defaults, used when $BACKEND2SORT does not mention $BACKEND
acfilter_b2s="ndb=a"
case ,$BACKEND2SORT in *,$BACKEND=*) acfilter_b2s=$BACKEND2SORT ;; esac
case $BACKEND,$acfilter_b2s in
null,*) : ;;
*,$BACKEND=[ae]*)
if (perl -e 0) 2>/dev/null; then perl -we '
($end, $sort, @e, @out) = ("", $ARGV[0] =~ /,\Q$ARGV[1]\E=(\w*)/);
sub lsort { shift(@_) =~ /[$sort]/o ? sort(@_) : @_; }
foreach (<STDIN>, "", "") {
if (/^ /s) { $e[-1] .= $_; next; }
if (/./m) { push @e, $_ unless /^\#/s; next; }
if (!@out) { print $end; $end = ""; }
if (@e) { push @out, join "", shift(@e), &lsort("a", splice @e); }
else { print join("\n", &lsort("e", splice @out)), $end; }
$end = $_;
}' ",$acfilter_b2s" "$BACKEND"
else grep -v '^#' | sort ; fi ;;
*) grep -v '^#' ;;
esac
13 years, 11 months
Re: (ITS#6337) Combining back-relay/slapo-rwm and back-hdb/slapo-accesslog seg faults
by h.b.furuseth＠usit.uio.no
Committed a new fix for Bind and Extended operations in HEAD.
This one has no impact on cases which already worked before this ITS:
It just searches for the callback and removes it.
Possibly something like the previous fix would actually make back-relay
better behaved, but such behavior change and my chatter in previous
message should go in a separate ITS if someone investigates it.
--
Hallvard
13 years, 11 months
