Replication from 1.2.5 to 1.2.6 failed
On 10/16/2010 11:42 AM, Edward Z. Yang wrote:
> I've got a failure, and I'm able to gdb it. However, I don't
> know what to look for. What kind of tracing would you like to
> see? I was going to wireshark but decrypting the Kerberos would
> be annoying.
If you can break in acquire_replica() on the master when it attempts to
begin replication with the older instance, I would like to see what
I am guessing that this will have some garbage value since it is not
being initialized to 0 as discussed earlier today on this list.
> [16/Oct/2010:14:17:16 -0400] NSMMReplicationPlugin - agmt="cn=GSSAPI Replication to busy-beaver.mit.edu" (busy-beaver:389): Unable to parse the response to the startReplication extended operation. Replication is aborting.
> 389 users mailing list
389 users mailing list