On 9/19/09, Quanah Gibson-Mount quanah@zimbra.com wrote:
--On Saturday, September 19, 2009 9:07 AM +0100 Tony Smith tony.smith.124@googlemail.com wrote:
thank you Quanah, very much appreciated! I copied the config from some howto and didn't really understand how each option works.
Update: the deletion still happens after I commented out the attrs line, or changed it to attrs="*,+". I tried to comment out other options as well, and it seems that after leaving out these options the problem is gone:
Hi Tony,
Did you reload your replica after fixing the syncrepl stanza? If you hadn't, it meant that all of your entries in the replica continued to be broken, which is why deletions continued.
#filter="(objectClass=*)" #scope=sub #schemachecking=offI don't really understand the impact of each; I post it here in case it might be helpful for someone else.
Those are already the default values. Commenting them out would have had no effect. See above.
Hi Quanah,
yes I did resync after each change by running:
/opt/openldap-2.3.43/lib/slapd -u openldap -g openldap -d 1 -c rid=001,csn=0
I re-checked again: - stop slapd on consumer - resync using the above command (Ctrl-C to stop when the resync seems done, no more messages) - add an entry on provider - uncomment these options - start slapd on consumer
and the deletions re-appear again.
Regards, Tony