The requirement is to minimize the replication retry time to minimum so that even if one of the masters in the community goes down and comes up after one hour, replication happens immediately without manual intervention.
On 11/16/2012 05:39 AM, Divya Ravikumar wrote:
converted from rtf
I need to reduce the max Replication retry time for my customer requirement
What is the requirement?
I have raised a enhancement request with fedora(Ticket 525).
As an immediate fix I have reduced the value of PROTOCOL_BACKOFF_MAXIMUM variable in repl5_prot_private.h from 300 seconds to 15 seconds.
I find no impact other the max retry time limit getting reduced to 15 seconds(which is my actual requirement) .
Does anyone see any potential flaw happening in replication due to this change.
Thanks in advance.
--389-devel mailing list