After clustering 3 machines, 1st machine is unable to accept memcache operations

Created an attachment (id=272)
membase logs

Image: CentOS5_2_X86_64_V4_1_10
Build: northscale_server_rhel_5.4_x86_64_1.6.0a-58-g30592d4.tar.gz

Deployed 3 images. Installed the build on each. Using the CLI I added the 2nd 2 nodes to the cluster:

membase server-add -c domU-12-31-39-13-D9-11.compute-1.internal:8080 --server-add domU-12-31-39-13-DC-31.compute-1.internal
membase server-add -c domU-12-31-39-13-D9-11.compute-1.internal:8080 --server-add domU-12-31-39-0A-61-21.compute-1.internal

Things look good:

membase server-list -c domU-12-31-39-13-D9-11.compute-1.internal:8080
ns_1@10.201.218.223 healthy
ns_1@10.201.223.191 healthy
ns_1@10.211.98.207 healthy

However, memslap to domU-12-31-39-13-D9-11.compute-1.internal fails, unable to establish a connection. Memslap to the other two hosts fail. Additionally, telnet localhost 11211 works for the other two hosts in the cluster, but not the initial host.

Moxi seems to be running on 11211:

[root@domU-12-31-39-13-D9-11 ~]# ps awxx |grep moxi
2549 ? Ssl 0:00 ./bin/moxi/moxi -Z port_listen=11211,downstream_max=1 -z auth=,url=http://127.0.0.1:8080/pools/default/bucketsStreamingConfig/default,#@ -p 0

But it is unreachable:

[root@domU-12-31-39-13-D9-11 ~]# telnet localhost 11211Trying 127.0.0.1...
telnet: connect to address 127.0.0.1: Connection refused
telnet: Unable to connect to remote host: Connection refused

This should not be a firewall issue. All three hosts were deployed simultaneously with the same firewall configurations and 2/3 work.

Logs attached.