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