Anything wrong with cache=local for KC18 when not running a cluster?

Configuring the server
1

I’m running Keycloak 18 using JDK16 on OpenBSD using kc.sh start and the terminal prints the following messages when it boots:

2022-07-12 06:03:45,794 WARN  [org.jgroups.protocols.UDP] (keycloak-cache-init) JGRP000015: the receive buffer of socket MulticastSocket was set to 20.00MB, but the OS only allocated 2.10MB
2022-07-12 06:03:45,795 WARN  [org.jgroups.protocols.UDP] (keycloak-cache-init) JGRP000015: the receive buffer of socket MulticastSocket was set to 25.00MB, but the OS only allocated 2.10MB

I tried to figure out what MulticastSocket is and didn’t find something that helped me out there, so in keycloak.conf I set cache=local and I now only receive WARN for:

2022-07-12 08:20:28,376 WARN  [org.infinispan.CONFIG] (keycloak-cache-init) ISPN000569: Unable to persist Infinispan internal caches as no global state enabled
2022-07-12 08:20:28,635 WARN  [org.infinispan.PERSISTENCE] (keycloak-cache-init) ISPN000554: jboss-marshalling is deprecated and planned for removal

These were showing up before and everything is otherwise working fine.

Since I’m not currently planned to run Keycloak as a cluster, is there anything about setting cache=local that might cause problems at any point?

2

If you don‘t want to run Keycloak in a cluster, having cache=local is ok.
But you will have to live with all the cons which come with not having a cluster. So, if your KC node fails, your complete auth system is down and nobody is able to use your systems. :man_shrugging: