https://fedorahosted.org/389/attachment/ticket/47516/0001-Ticket-47516-replication-stops-with-excessive-clock-.patch
With the previous patch, the CSN generator was not updated. The server
with the greater timestamp could issue update operations that the
consumer could not "overrule".
With the new patch, the csn generator on the consumer is immediately
updated.
Note that with this patch, if there is some sort of ntp or vm problem
that causes a large system time difference, the large time skews will be
immediately propagated everywhere.
We need to figure out some way to safely "repair" or "reset" the CSN
time generator, or, figure out some way to avoid large time skews
altogether.
--
389-devel mailing list
389-devel@xxxxxxxxxxxxxxxxxxxxxxx
https://admin.fedoraproject.org/mailman/listinfo/389-devel
