[Date Prev][Date Next]
[Chronological]
[Thread]
[Top]
syncrepl, version compatibility, partial sync
- To: openldap-technical@openldap.org
- Subject: syncrepl, version compatibility, partial sync
- From: Ryan Lovett <relovett@gmail.com>
- Date: Thu, 14 Jun 2012 10:31:12 -0700
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:content-type; bh=fw0G+zCkPd2LHLDEo92fv74XaK00DZxznH81uMMnQJQ=; b=oF/8e1tyghL+12c/kujY/ksnHNkYBF0hxaWU2pqSBrRzTPZu8hb0LzuZWNRaG/+aDs vI5EmvaNpYh70LwJNIq6lxwgx33EbClDvokf6Nzz2K+f3ofPHRN1aBk9t+rEBtbHiAdv yO98VJA/X94g2SGivhA5V6E/KdlBV1I+W9BrF5AQ9+1cpxmoIGGLNS+D6UqxiW99CdCU LXXJjEPfbyxYh8sRs0/oIbFHPssvw4p11VwcdMFUsfSwRLV5GdXirqqGG/viW6dWZrMc eog3jHDC2uvEXf4ugqqMe9j7CT4wCxNZVQT4BntA4VxX9v+pwpDL0ny2FVq42oZUW2Et 0IuQ==
Are there any known syncrepl compatibility issues between different 2.4.x versions? We have a master slapd running 2.4.9 and a consumer running 2.4.21 and are having a syncrepl problem and only some of the data is being synced. The consumer slapd logs messsages of the form:
entries have identical CSN ou=something,dc=something 20120614015145.546743Z#000000#000#000000
and
syncrepl_entry: rid=200 entry unchanged, ignored (ou=other,dc=something)
syncrepl_message_to_entry: rid=200 mods check (objectClass: value #8 invalid per syntax)
do_syncrepl: rid=200 rc 21 retrying (9 retries left)
In searching for similar reports we came across various solutions such as making sure the provider URI in the consumer's slapd.conf matches the server's URI at launch. The server was listening on ldap:/// and the consumer was set to connect to ldap://
server.name/ though we modified the server to only listen on ldap://
server.name/.
We also tried stopping both consumer and master, flushing the databases on both, slapadding data back into the master, then starting both instances. The log messages return however just with updated CSN stamps. The consumer is never able to pull down the full data even with refreshAndPersist with sensible retry intervals.
Thanks for your time,
Ryan