[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: leaks (was: releng plans)

To: openldap-devel@OpenLDAP.org
Subject: Re: leaks (was: releng plans)
From: Quanah Gibson-Mount <quanah@stanford.edu>
Date: Wed, 01 Sep 2004 15:12:42 -0700
Content-disposition: inline
In-reply-to: <2477CD7EAC6E00CA4C11FD1E@cadabra-dsl.stanford.edu>
References: <6.1.2.0.0.20040828163956.041c06f8@127.0.0.1> <EC99A027429E3B7722FCAB7F@cadabra.stanford.edu> <6.1.2.0.0.20040830123911.04a4bab8@127.0.0.1> <27F8117D836BBC534DF30C4B@cadabra.stanford.edu> <2477CD7EAC6E00CA4C11FD1E@cadabra-dsl.stanford.edu>

--On Tuesday, August 31, 2004 9:45 AM -0700 Quanah Gibson-Mount <quanah@stanford.edu> wrote:

--On Monday, August 30, 2004 12:45 PM -0700 Quanah Gibson-Mount
<quanah@stanford.edu> wrote:

--On Monday, August 30, 2004 12:40 PM -0700 "Kurt D. Zeilenga"
<Kurt@OpenLDAP.org> wrote:

With or without syncrepl?  I note that Jong switched syncrepl
back to malloc(3).  That might help.

Without. :( My production servers are all still using slurpd.


Just to follow up, I have deployed a profiling version of slapd to the
affected replicas, so hopefully one will be needing help soon, so I can
see what its issues are. ;)

Follow-up: I was able to get profiling data today from 3 servers. Howard examined the output, and has committed a patch. I'm currently building 2.2.15 with that patch, and will deploy it onto my servers, and see if they continue to run into this issue or not.

--Quanah

--
Quanah Gibson-Mount
Principal Software Developer
ITSS/Shared Services
Stanford University
GnuPG Public Key: http://www.stanford.edu/~quanah/pgp.html

Follow-Ups:
- Re: leaks (was: releng plans)
  - From: Quanah Gibson-Mount <quanah@stanford.edu>

Prev by Date: Re: syncrepl ramblings again
Next by Date: Overlays: freeing resources on a hijacked Operation
Index(es):
- Chronological
- Thread