[Pkg-openldap-devel] Bug#598361: Bug#598361: Bug#598361: Bug#598361: slapd: slapcat gives "unclean shutdown detected; attempting recovery" after squeeze upgrade

[Quanah Gibson-Mount]

--On Tuesday, September 28, 2010 10:40 AM -0500 "Walton, Bryan K" 
<bryan-walton at uiowa.edu> wrote:

> On Tue, 2010-09-28 at 08:29 -0700, Quanah Gibson-Mount wrote:
>
>> Finally, I would note I receive no such error while using slapcat with
>> my  own build of OpenLDAP 2.4.23:
>>
>> zimbra at zre-ldap002:~$ /opt/zimbra/openldap/sbin/slapcat -F
>> /opt/zimbra/data/ldap/config -l /tmp/test.out -b ''
>> zimbra at zre-ldap002:~$
>
> Hi Quanah, I'll acknowledge that I this doesn't seem universal.  This is
> only happening on our master ldap server.  On the slave, we are get no
> such error.  Both are running the current slapd that exists in Squeeze.
>
> Still, the master does give the error.  Running db4.8_recover doesn't
> fix whatever the problem is (if there is really a problem).  Any ideas
> how this can be fixed?  Is this possibly a bug in BDB 4.8 rather than
> slapd?

Well, the server I ran slapcat on is a master server, so I don't think 
that's relevant. ;)  Also, numerous people run OpenLDAP with bdb 4.8 and 
haven't reported such an error.  So it's unlikely it is a 4.8 issue, 
although it could be possible.

Is there any difference in file system between the master and replicas? 
I.e., nfs or ext4 vs ext3 etc?

--Quanah


--

Quanah Gibson-Mount
Principal Software Engineer
Zimbra, Inc
--------------------
Zimbra ::  the leader in open source messaging and collaboration