wildfly 12.0.0 domain mode clustering

[glt]

Hi,

starting Wildfly 12.0.0 in domain mode I've got following error:

[Server:pk-fe1-one] 19:47:23,067 ERROR [org.jboss.msc.service.fail] (ServerService Thread Pool -- 73) MSC000001: Failed to start service org.wildfly.clustering.jgroups.channel.ee-web: 
org.jboss.msc.service.StartException in service org.wildfly.clustering.jgroups.channel.ee-web: java.lang.Exception: failed to open a port in range 55200-55200
[Server:pk-fe1-one]     at org.jboss.as.clustering.jgroups.subsystem.ChannelBuilder.start(ChannelBuilder.java:124)
[Server:pk-fe1-one]     at org.wildfly.clustering.service.AsynchronousServiceBuilder.lambda$start$0(AsynchronousServiceBuilder.java:99) [wildfly-clustering-service-12.0.0.Final.jar:12.0.0.Final]

and:

[Server:pk-fe1-two] 19:47:31,198 ERROR [org.jboss.msc.service.fail] (ServerService Thread Pool -- 75) MSC000001: Failed to start service org.wildfly.clustering.jgroups.channel.ee-web:
org.jboss.msc.service.StartException in service org.wildfly.clustering.jgroups.channel.ee-web: java.lang.Exception: failed to open a port in range 55350-55350
[Server:pk-fe1-two] at org.jboss.as.clustering.jgroups.subsystem.ChannelBuilder.start(ChannelBuilder.java:124)

After wildfly starts, netstat command shows two processes running on that ports:

udp        0      0 192.168.88.71:55200         0.0.0.0:*                               11262/java
udp        0      0 192.168.88.71:55350         0.0.0.0:*                               11302/java

Socket binding section is like default:

<socket-binding name="jgroups-tcp" interface="private" port="7600"/>

<socket-binding name="jgroups-tcp-fd" interface="private" port="57600"/>

<socket-binding name="jgroups-udp" interface="private" port="55200" multicast-address="${jboss.default.multicast.address:230.0.0.7}" multicast-port="45688"/>

<socket-binding name="jgroups-udp-fd" interface="private" port="54200"/>

<socket-binding name="messaging-group" interface="multicast" port="0" multicast-address="${jboss.messaging.group.address:231.7.7.10}" multicast-port="${jboss.messaging.group.port:9876}"/>

[pferraro]

Please post the configuration for your jgroups and infinispan subsystems.  I suspect you're trying to create multiple jgroups channels using the same bind address/port.

[glt]

I didn't changed it after migration from wildfly 8.2. Only added -bprivate IP address binding.

<subsystem xmlns="urn:jboss:domain:jgroups:6.0">

                <channels>

                    <channel name="ee-server" stack="udp"/>

                    <channel name="ee-web" stack="udp"/>

                    <channel name="ee-ejb" stack="udp"/>

                    <channel name="ee-hibernate" stack="udp"/>

                </channels>

                <stacks default="udp">

                    <stack name="udp">

                        <transport type="UDP" socket-binding="jgroups-udp"/>

                        <protocol type="PING"/>

                        <protocol type="MERGE3"/>

                        <protocol type="FD_SOCK" socket-binding="jgroups-udp-fd"/>

                        <protocol type="FD_ALL"/>

                        <protocol type="VERIFY_SUSPECT"/>

                        <protocol type="pbcast.NAKACK2"/>

                        <protocol type="UNICAST3"/>

                        <protocol type="pbcast.STABLE"/>

                        <protocol type="pbcast.GMS"/>

                        <protocol type="UFC"/>

                        <protocol type="MFC"/>

                        <protocol type="FRAG2"/>

                        <protocol type="RSVP"/>

                    </stack>

                    <stack name="tcp">

                        <transport type="TCP" socket-binding="jgroups-tcp"/>

                        <socket-protocol type="MPING" socket-binding="jgroups-mping"/>

                        <protocol type="MERGE2"/>

                        <protocol type="FD_SOCK" socket-binding="jgroups-tcp-fd"/>

                        <protocol type="FD"/>

                        <protocol type="VERIFY_SUSPECT"/>

                        <protocol type="pbcast.NAKACK2"/>

                        <protocol type="UNICAST3"/>

                        <protocol type="pbcast.STABLE"/>

                        <protocol type="pbcast.GMS"/>

                        <protocol type="MFC"/>

                        <protocol type="FRAG2"/>

                        <protocol type="RSVP"/>

                    </stack>

                </stacks>

            </subsystem>

<subsystem xmlns="urn:jboss:domain:infinispan:5.0">

                <cache-container name="web" default-cache="passivation" module="org.wildfly.clustering.web.infinispan">

                    <local-cache name="passivation">

                        <transaction mode="BATCH"/>

                        <file-store passivation="true" purge="false"/>

                    </local-cache>

                    <local-cache name="persistent">

                        <transaction mode="BATCH"/>

                        <file-store passivation="false" purge="false"/>

                    </local-cache>

                    <local-cache name="concurrent">

                        <file-store passivation="true" purge="false"/>

                    </local-cache>

                </cache-container>

                <cache-container name="ejb" aliases="sfsb" default-cache="passivation" module="org.wildfly.clustering.ejb.infinispan">

                    <local-cache name="passivation">

                        <transaction mode="BATCH"/>

                        <file-store passivation="true" purge="false"/>

                    </local-cache>

                    <local-cache name="persistent">

                        <transaction mode="BATCH"/>

                        <file-store passivation="false" purge="false"/>

                    </local-cache>

                </cache-container>

                <cache-container name="hibernate" default-cache="local-query" module="org.infinispan.hibernate-cache">

                    <local-cache name="entity">

                        <transaction mode="NON_XA"/>

                        <object-memory size="10000"/>

                        <expiration max-idle="100000"/>

                    </local-cache>

                    <local-cache name="local-query">

                        <transaction mode="NONE"/>

                        <object-memory size="10000"/>

                        <expiration max-idle="100000"/>

                    </local-cache>

                    <local-cache name="timestamps">

                        <transaction mode="NONE"/>

                    </local-cache>

                </cache-container>

                <cache-container name="server" default-cache="default" module="org.wildfly.clustering.server">

                    <local-cache name="default">

                        <transaction mode="BATCH"/>

                    </local-cache>

                </cache-container>

            </subsystem>

Also there are 2 server groups:

<server name="pk-fe1-one" group="main-server-group"/>

        <server name="pk-fe1-two" group="main-server-group" auto-start="true">

            <socket-bindings port-offset="150"/>

        </server>

[pferraro]

It looks like these channels might have been auto-created following a configuration upgrade?

                    <channel name="ee-server" stack="udp"/>

                    <channel name="ee-web" stack="udp"/>

                    <channel name="ee-ejb" stack="udp"/>

                    <channel name="ee-hibernate" stack="udp"/>

Only one of these channels will ever be able to start at the same time, since they use the same protocol stack, and thus will try to listen on the same bind address/port.

The default configuration for WF12 defines a single channel (named "ee"), over which all clustering services multiplex.

However, looking at your Infinispan subsystem configuration, I do not see any clustered Infinispan caches that would attempt to start a channel - so I'm a bit confused as to which service is forcing the channel to start.  Messaging perhaps?  Can you post that subsystem configuration?

[glt]

Sorry, it seems I posted infinispan subsystem section from before migration.

After migration there is a <transport/> tag with "channel" attribute in every cache tag. something like <transport channel="ee-server" lock-timeout="60000"/>.

When I removed the channel attribute and restored jgroups channels to default, servers starts without errors.

Thank you.

Btw. welcome screen in WF12 shows "Welcome to WildFly 11"

[pferraro]

glt Glad it's working.  Thanks for pointing out the version mismatch.  We ended up removing the version info from the welcome page altogether.

Remove version info from server welcome page · wildfly/wildfly@31db6a5 · GitHub