Segmentation fault - on replication - freebsd

What steps will reproduce the problem?
1. memcached -p 11211 -m 64 -x 127.0.0.1 -vvv -u nobody
2. memcached -p 11222 -m 64 -x 127.0.0.1 -v -u nobody

What is the expected output? What do you see instead?

On master:
replication: start
REP>16 OK
16: going from conn_parse_cmd to conn_new_cmd
16: going from conn_new_cmd to conn_waiting
16: going from conn_waiting to conn_read
<20 server listening (udp)
<20 server listening (udp)
<20 server listening (udp)
<20 server listening (udp)
<21 new auto-negotiating client connection
21: going from conn_new_cmd to conn_waiting
21: going from conn_waiting to conn_read
21: going from conn_read to conn_parse_cmd
21: Client using the ascii protocol
<21 get 2e948d53e053a7a36ba02f2c7effaa0e
> NOT FOUND 2e948d53e053a7a36ba02f2c7effaa0e
>21 END
21: going from conn_parse_cmd to conn_mwrite
21: going from conn_mwrite to conn_new_cmd
21: going from conn_new_cmd to conn_waiting
21: going from conn_waiting to conn_read
21: going from conn_read to conn_parse_cmd
<21 set 2e948d53e053a7a36ba02f2c7effaa0e 3 180 1765
21: going from conn_parse_cmd to conn_nread
> NOT FOUND 2e948d53e053a7a36ba02f2c7effaa0e
replication: alloc c=1
>21 STORED
21: going from conn_nread to conn_write
replication: pop
21: going from conn_write to conn_new_cmd
21: going from conn_new_cmd to conn_waiting
21: going from conn_waiting to conn_read
Segmentation fault

Slave is working well

What version of the product are you using? On what operating system?
libevent-1.4.14b_2, memcached-1.4.15 , freebsd 9.1-p6 , installed from ports.

Original issue reported on code.google.com by rabcz...@googlemail.com on 23 Aug 2013 at 1:48

JumpingYang / memcached

Segmentation fault - on replication - freebsd #339