Date: Sat, 05 May 2012 21:11:22 +0200 From: "Hartmann, O." <ohartman@zedat.fu-berlin.de> To: "Bjoern A. Zeeb" <bzeeb-lists@lists.zabbadoz.net>, freebsd-current@freebsd.org Subject: Re: OpenLDAP 2.4.31 on FreeBSD 10.0-CURRENT/amd64 broken! Message-ID: <4FA57B5A.8030203@zedat.fu-berlin.de> In-Reply-To: <EFEED525-6FDC-42CC-999F-93D700E315B4@lists.zabbadoz.net> References: <4FA54D4A.4050703@zedat.fu-berlin.de> <EFEED525-6FDC-42CC-999F-93D700E315B4@lists.zabbadoz.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 05/05/12 20:36, Bjoern A. Zeeb wrote: > On 5. May 2012, at 15:54 , Hartmann, O. wrote: > >> I'm floating like a dead man in the water and I was wondering if someone >> else doesn't face this problem. FreeBSD is said to be run in large >> environments, so at least one should have OpenLDAP as user backend >> running ... > > Ok, I am running the same openldap version on HEAD in production on amd64 and i386. > My HEAD is from about 20120426 1130 UTC and my packages are cleanly build on that > wprld. Here: FreeBSD 10.0-CURRENT #0 r235062: Sat May 5 20:02:26 CEST 2012 I can not say what the last working SVN release tag was, maybe around r234900, I can't say. As I wrote, Thursday last week, ~20120503, a buildworld from 20120502 worked fine. On Thursday 03.05.2012 I did the last source update and buildworld, shutdown the box at home and switched it on on Friday morning before heading to the laboratory - and LDAP failed. At the lab, the box was still working, but I was highly risky in my motivations and recompiled that box with a freshly updates source tree - and it failed, too. I did not follow the tags, sorry. Last I realized flushing into the sources was OpenSSL stuff due to security issues. I use OpenSSL certificates to connect via TLS to the server. There is also usually no core dumped. > > > The differences to what I see with your setups are: > > 1) I am still on db48. > > 2) ? Contrary to you I probably have MALLOC_PRODUCTION set for my builds; > not sure if that makes a difference but if this problem started recently > the jemalloc import could be the cause? You didn't say from when your > last and your latest HEAD was. MALLOC_PRODUCTION is set in /etc/make.conf as well as in /etc/src.conf. /etc/src.conf is this on both failing FBSD 10.0-CUR boxes: WITH_CLANG= YES WITH_CLANG_EXTRAS= YES # WITH_BIND_LARGE_FILE= YES WITH_BIND_SIGCHASE= YES WITH_BIND_LIBS= YES # WITH_IDEA= YES WITH_HESIOD= YES # #WITH_BSD_GREP= YES #WITH_ICONV= YES # WITH_LIBCPLUSPLUS= YES # #WITH_OFED= YES # PORTS_MODULES= x11/nvidia-driver # MALLOC_PRODUCTION= YES Grüße, Oliver > > Gruesse > /bz >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4FA57B5A.8030203>