Okay, so I gave up using that way and now using S3 instead, that way I can see what they send to each other.

First thing I noticed is the file that they modify looks like this:
74594d4ef42f e0172d5d-8392-5ee7-bba3-03d997c2ef6e 127.0.0.1:55200 T

When another instance connects, it tries to connect to 127.0.0.1:55200 which obviously doesn’t work, and that logs out like this:

04:26:15,908 WARN  [org.jgroups.protocols.pbcast.GMS] (MSC service thread 1-2) 74594d4ef42f: JOIN(74594d4ef42f) sent to e8a65867f61d timed out (after 3000 ms), on try 1


Again I looked reverse and found out that jgroups-udp uses this port number. So I found that docker uses 172.17.0.x and assigned 2 IP addresses. Like this:

    <interface name="dockerIp">
        <inet-address value="${jboss.bind.address:172.17.0.3}"/>    </interface>
    <socket-binding name="jgroups-udp" interface="dockerIp" port="55200" multicast-address="${jboss.default.multicast.address:230.0.0.4}" multicast-port="45688"/>
However right now I still get the previous problem:

04:28:31,418 ERROR [org.jboss.msc.service.fail] (MSC service thread 1-2) MSC000001: Failed to start service jboss.jgroups.channel.ee: org.jboss.msc.service.StartException in service jboss.jgroups.channel.ee: java.security.PrivilegedActionException: java.net.BindException: [UDP] /0.0.0.0 is not a valid address on any local network interface

        at org.wildfly.clustering.jgroups.spi.service.ChannelBuilder.start(ChannelBuilder.java:80)

        at org.jboss.msc.service.ServiceControllerImpl$StartTask.startService(ServiceControllerImpl.java:1948)

        at org.jboss.msc.service.ServiceControllerImpl$StartTask.run(ServiceControllerImpl.java:1881)

        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)

        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)

        at java.lang.Thread.run(Thread.java:745)

Caused by: java.security.PrivilegedActionException: java.net.BindException: [UDP] /0.0.0.0 is not a valid address on any local network interface

        at org.wildfly.security.manager.WildFlySecurityManager.doChecked(WildFlySecurityManager.java:640)

        at org.jboss.as.clustering.jgroups.JChannelFactory.createChannel(JChannelFactory.java:98)

        at org.wildfly.clustering.jgroups.spi.service.ChannelBuilder.start(ChannelBuilder.java:78)

        ... 5 more

Why does it keep showing that my IP address is 0.0.0.0? Despite I specifically put the right IP address?

PS: I’m well aware this is a bad hack to put IP addresses specifically, but at this point I’m trying to get it working.


From: John Bartko <john.bartko@drillinginfo.com>
Date: Thursday, March 31, 2016 at 12:53 PM
To: Abdullah Sarp Kaya <akaya@expedia.com>
Cc: Marek Posolda <mposolda@redhat.com>, "keycloak-user@lists.jboss.org" <keycloak-user@lists.jboss.org>
Subject: Re: [keycloak-user] Keycloak Clustering, other instance logs me out

You're on the right track. Taking a look at my notes, the following may be required:
  • docker container with host mode networking and NET_ADMIN capabilities if clustering across container hosts
  • entrypoint parameters:
    -Djgroups.bind_addr=HOST_IP -b HOST_IP --server-config standalone-ha.xml
    note the host default interface IP must be used and not a wildcard of 0.0.0.0
  • the socket-binding changed

On Wed, Mar 30, 2016 at 7:49 PM, Sarp Kaya <akaya@expedia.com> wrote:
Sorry to send an e-mail again what I did is changed this:
<socket-binding name="jgroups-udp" interface="public" port="55200" multicast-address="${jboss.default.multicast.address:230.0.0.4}" multicast-port="45688"/>
Also set the public interface as:
<interface name="public">
    <inet-address value="${jboss.bind.address:0.0.0.0}"/>
</interface>
But now I’m getting this error:

00:45:40,146 ERROR [org.jboss.msc.service.fail] (MSC service thread 1-2) MSC000001: Failed to start service jboss.jgroups.channel.ee: org.jboss.msc.service.StartException in service jboss.jgroups.channel.ee: java.security.PrivilegedActionException: java.net.BindException: [UDP] /0.0.0.0 is not a valid address on any local network interface

        at org.wildfly.clustering.jgroups.spi.service.ChannelBuilder.start(ChannelBuilder.java:80)

        at org.jboss.msc.service.ServiceControllerImpl$StartTask.startService(ServiceControllerImpl.java:1948)

        at org.jboss.msc.service.ServiceControllerImpl$StartTask.run(ServiceControllerImpl.java:1881)

        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)

        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)

        at java.lang.Thread.run(Thread.java:745)

Caused by: java.security.PrivilegedActionException: java.net.BindException: [UDP] /0.0.0.0 is not a valid address on any local network interface

        at org.wildfly.security.manager.WildFlySecurityManager.doChecked(WildFlySecurityManager.java:640)

        at org.jboss.as.clustering.jgroups.JChannelFactory.createChannel(JChannelFactory.java:98)

        at org.wildfly.clustering.jgroups.spi.service.ChannelBuilder.start(ChannelBuilder.java:78)

        ... 5 more

Caused by: java.net.BindException: [UDP] /0.0.0.0 is not a valid address on any local network interface

        at org.jgroups.util.Util.checkIfValidAddress(Util.java:3522)

        at org.jgroups.stack.Configurator.ensureValidBindAddresses(Configurator.java:903)

        at org.jgroups.stack.Configurator.setupProtocolStack(Configurator.java:118)

        at org.jgroups.stack.Configurator.setupProtocolStack(Configurator.java:57)

        at org.jgroups.stack.ProtocolStack.setup(ProtocolStack.java:477)

        at org.jgroups.JChannel.init(JChannel.java:853)

        at org.jgroups.JChannel.<init>(JChannel.java:159)

        at org.jboss.as.clustering.jgroups.JChannelFactory$1.run(JChannelFactory.java:95)

        at org.jboss.as.clustering.jgroups.JChannelFactory$1.run(JChannelFactory.java:92)

        at org.wildfly.security.manager.WildFlySecurityManager.doChecked(WildFlySecurityManager.java:636)

        ... 7 more


If I put a different IP address (say the docker machine IP address) I get the same error as well.

From: John Bartko <john.bartko@drillinginfo.com>
Date: Thursday, March 31, 2016 at 1:23 AM
To: Marek Posolda <mposolda@redhat.com>
Cc: Abdullah Sarp Kaya <akaya@expedia.com>, "keycloak-user@lists.jboss.org" <keycloak-user@lists.jboss.org>
Subject: Re: [keycloak-user] Keycloak Clustering, other instance logs me out
When clustering across separate hosts, I had to change the jgroups-udp socket binding to listen on the public interface (binds to loopback by default).

On Wed, Mar 30, 2016 at 5:52 AM, Marek Posolda <mposolda@redhat.com> wrote:

When you start the second instance, are you seeing something like this in log of both servers?

INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (Incoming-10,shared=udp)
ISPN000094: Received new cluster view: [node1/keycloak|1] (2) [node1/keycloak, node2/keycloak]

If not, then clustering doesn't work properly and the servers doesn't form a cluster with each other. From the log you sent, there is just startup of one server, which indicates that clustering may not work.

Maybe multicast doesn't work in your network. Either disable firewall/selinux/whatever to have multicast working or switch to TCP JGroups channel instead of UDP. See the Wildfly and JGroups documentation for more details.

Also I personally use the virtual hosts to test clustering of 2 servers on same machine (Ie. have virtual servers like kc1:8080 and kc2:8080) . Using same host but differ just in port number ( host:8080 and host:8081 ) may causing mess with cookies, so
I am personally not using the setup like this.

Marek


On 30/03/16 08:38, Sarp Kaya wrote:
I have tried using standalone-ha.xml with shared database. I thought that would be enough but it seems like it’s not. The problem is

I log into kc1 instance, and subsequent requests are authenticated.
Then I try viewing
host:8080/auth/realms/master/account
Which is also authenticated.

Then I try to view this on kc1 by changing port like:
host:8081/auth/realms/master/account

At this point I expect to see same page. However I get prompted for login for both kc1 and kc2. I see no logs at this point.

So now I have switched to using keycloak-ha-postgres because it seemed to me that it comes clustering enabled out of box. So I nearly did exactly what this page:
 told me to so. The only difference that I have done is adding ports (with –p 8080:8080 to one instance and –p 8081:8080 to the another one) and adding a new user.

Once I start the I get this log:

05:28:49,888 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-1) ISPN000078: Starting JGroups channel keycloak

05:28:49,893 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-2) ISPN000078: Starting JGroups channel server

05:28:49,902 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-1) ISPN000094: Received new cluster view for channel keycloak: [a05014a5dc24|0] (1) [a05014a5dc24]

05:28:49,907 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-1) ISPN000079: Channel keycloak local address is a05014a5dc24, physical addresses are [127.0.0.1:55200]

05:28:49,902 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-2) ISPN000094: Received new cluster view for channel server: [a05014a5dc24|0] (1) [a05014a5dc24]

05:28:49,914 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-2) ISPN000079: Channel server local address is a05014a5dc24, physical addresses are [127.0.0.1:55200]

05:28:49,925 INFO  [org.infinispan.factories.GlobalComponentRegistry] (MSC service thread 1-2) ISPN000128: Infinispan version: Infinispan 'Mahou' 8.1.0.Final

05:28:49,926 INFO  [org.infinispan.factories.GlobalComponentRegistry] (MSC service thread 1-1) ISPN000128: Infinispan version: Infinispan 'Mahou' 8.1.0.Final

05:28:49,978 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-2) ISPN000078: Starting JGroups channel web

05:28:49,982 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-2) ISPN000094: Received new cluster view for channel web: [a05014a5dc24|0] (1) [a05014a5dc24]

05:28:49,984 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-2) ISPN000079: Channel web local address is a05014a5dc24, physical addresses are [127.0.0.1:55200]

05:28:49,985 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-1) ISPN000078: Starting JGroups channel hibernate

05:28:49,986 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-1) ISPN000094: Received new cluster view for channel hibernate: [a05014a5dc24|0] (1) [a05014a5dc24]

05:28:49,987 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-1) ISPN000079: Channel hibernate local address is a05014a5dc24, physical addresses are [127.0.0.1:55200]

05:28:50,028 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-1) ISPN000078: Starting JGroups channel ejb

05:28:50,030 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-1) ISPN000094: Received new cluster view for channel ejb: [a05014a5dc24|0] (1) [a05014a5dc24]

05:28:50,031 INFO  [org.infinispan.remoting.transport.jgroups.JGroupsTransport] (MSC service thread 1-1) ISPN000079: Channel ejb local address is a05014a5dc24, physical addresses are [127.0.0.1:55200]

05:28:50,357 INFO  [org.jboss.as.clustering.infinispan] (ServerService Thread Pool -- 56) WFLYCLINF0002: Started realmVersions cache from keycloak container

05:28:50,391 INFO  [org.jboss.as.clustering.infinispan] (ServerService Thread Pool -- 54) WFLYCLINF0002: Started offlineSessions cache from keycloak container

05:28:50,397 INFO  [org.jboss.as.clustering.infinispan] (ServerService Thread Pool -- 55) WFLYCLINF0002: Started loginFailures cache from keycloak container

05:28:50,396 INFO  [org.jboss.as.clustering.infinispan] (ServerService Thread Pool -- 52) WFLYCLINF0002: Started sessions cache from keycloak container

05:28:50,392 INFO  [org.jboss.as.clustering.infinispan] (ServerService Thread Pool -- 53) WFLYCLINF0002: Started realms cache from keycloak container

05:28:50,399 INFO  [org.jboss.as.clustering.infinispan] (ServerService Thread Pool -- 51) WFLYCLINF0002: Started users cache from keycloak container

05:28:50,402 INFO  [org.jboss.as.clustering.infinispan] (ServerService Thread Pool -- 50) WFLYCLINF0002: Started work cache from keycloak container



However I still have the same issue as above (I get logged out). Also I don’t get any new logs for the entire log-in, log-out processes.

Am I doing something wrong?
Thanks,
Sarp


_______________________________________________
keycloak-user mailing list
keycloak-user@lists.jboss.orghttps://lists.jboss.org/mailman/listinfo/keycloak-user


_______________________________________________
keycloak-user mailing list
keycloak-user@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/keycloak-user