mariadb/galera doesn't start after cluster reboot

[Yinon]

I have red hat openstack with 3 controllers. mariadb is set with galera.
I am testing reboot scenarios. I reboot all 3 controllers, and after they come up, mariadb is down. Examining the log shows some problems with Galera:

161209 4:02:29 [ERROR] WSREP: failed to open gcomm backend connection: 110: failed to reach primary view: 110 (Connection timed out)
         at gcomm/src/pc.cpp:connect():141
161209 4:02:29 [ERROR] WSREP: gcs/src/gcs_core.c:gcs_core_open():202: Failed to open backend connection: -110 (Connection timed out)
161209 4:02:29 [ERROR] WSREP: gcs/src/gcs.c:gcs_open():1291: Failed to open channel 'galera_cluster' at 'gcomm://overcloud-controller-0,overcloud-controller-1,overcloud-controller-2': -110 (Connection timed out)
161209 4:02:29 [ERROR] WSREP: gcs connect failed: Connection timed out
161209 4:02:29 [ERROR] WSREP: wsrep::connect() failed: 7
161209 4:02:29 [ERROR] Aborting

It was working prior to reboot.
I disabled firewall and even set selinux to permissive (setenforce 0).

What is the problem?

I saw some posts saying I should run galera_new_cluster, but:
a. this script doesn't exist
b. the cluster was already working, so setting a new cluster doesn't make sense.

If any other info is needed I am happy to add.

[Yinon (yinonby)]

Was probably a problem with firewall blocking ports