libre.sh/keycloak-scim
2
0
Fork 0
Code Issues 14 Pull requests Releases Packages Activity Actions 6

Merge pull request #4425 from mposolda/cross-dc-wildfly

Browse source 
KEYCLOAK-5294 Updated README for cross-dc setup on Wildfly
This commit is contained in:
Marek Posolda 2017-08-25 21:27:51 +02:00 committed by GitHub
commit 34648c6f0c
2 changed files with 220 additions and 71 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

287
misc/CrossDataCenter.md
View file

@ -3,116 +3,265 @@ Test Cross-Data-Center scenario (test with external JDG server)
These are temporary notes. This docs should be removed once we have cross-DC support finished and properly documented. These are temporary notes. This docs should be removed once we have cross-DC support finished and properly documented.
Note that these steps are already automated, see Cross-DC tests section in [HOW-TO-RUN.md](../testsuite/integration-arquillian/HOW-TO-RUN.md) document. These steps are already automated for embedded Undertow, see Cross-DC tests section in [HOW-TO-RUN.md](../testsuite/integration-arquillian/HOW-TO-RUN.md) document. For Wildfly they are not yet automated.
Following instructions are related to Wildfly server.
What is working right now is: What is working right now is:
- Propagating of invalidation messages for "realms" and "users" caches - Propagating of invalidation messages for `realms`, `users` and `authorization` caches
- All the other things provided by ClusterProvider, which is: - sessions, offline sessions and login failures are propagated between datacenters
-- ClusterStartupTime (used for offlineSessions and revokeRefreshToken) is shared for all clusters in all datacenters
-- Periodic userStorage synchronization is always executed just on one node at a time. It won't be never executed concurrently on more nodes (Assuming "nodes" refer to all servers in all clusters in all datacenters)
What doesn't work right now:
- UserSessionProvider and offline sessions
Basic setup Basic setup
=========== ===========
This is setup with 2 keycloak nodes, which are NOT in cluster. They just share the same database and they will be configured with "work" infinispan cache with remoteStore, which will point This is the example setup simulating 2 datacenters `site1` and `site2` . Each datacenter consists of 1 infinispan server and 2 Keycloak servers.
to external JDG server. So 2 infinispan servers and 4 Keycloak servers are totally in the testing setup.
JDG Server setup * Site1 consists of infinispan server `jdg1` and 2 Keycloak servers `node11` and `node12` .
----------------
- Download JDG 7.0 server and unzip to some folder
- Add this into JDG_HOME/standalone/configuration/standalone.xml under cache-container named "local" : * Site2 consists of infinispan server `jdg2` and 2 Keycloak servers `node21` and `node22` .
``` * Infinispan servers `jdg1` and `jdg2` forms cluster with each other. The communication between them is the only communication between the 2 datacenters.
<local-cache name="work" start="EAGER" batching="false" />
* Keycloak servers `node11` and `node12` forms cluster with each other, but they don't communicate with any server in `site2` . They communicate with infinispan server `jdg1` through the HotRod protocol (Remote cache).
* Same applies for `node21` and `node22` . They have cluster with each other and communicate just with `jdg2` server through the HotRod protocol.
TODO: Picture on blog
* For example when some object (realm, client, role, user, ...) is updated on `node11`, the `node11` will send invalidation message. It does it by saving special cache entry to the remote cache `work` on `jdg1` .
The `jdg1` notifies client listeners in same DC (hence on `node12`) and propagate the message to it. But `jdg1` is in replicated cache with `jdg2` .
So the entry is saved on `jdg2` too and `jdg2` will notify client listeners on nodes `node21` and `node22`.
All the nodes know that they should invalidate the updated object from their caches. The caches with the actual data (`realms`, `users` and `authorization`) are infinispan local caches.
TODO: Picture and better explanation?
* For example when some userSession is created/updated/removed on `node11` it is saved in cluster on current DC, so the `node12` can see it. But it's saved also to remote cache on `jdg1` server.
The userSession is then automatically seen on `jdg2` server because there is replicated cache `sessions` between `jdg1` and `jdg2` . Server `jdg2` then notifies nodes `node21` and `node22` through
the client listeners (Feature of Remote Cache and HotRod protocol. See infinispan docs for details). The node, who is owner of the userSession (either `node21` or `node22`) will update userSession in the cluster
on `site2` . Hence any user requests coming to Keycloak nodes on `site2` will see latest updates.
TODO: Picture and better explanation?
Example setup assumes all 6 servers are bootstrapped on localhost, but each on different ports.
Infinispan Server setup
-----------------------
1) Download Infinispan 8.2.6 server and unzip to some folder
2) Add this into `JDG1_HOME/standalone/configuration/clustered.xml` under cache-container named `clustered` :
```xml
<cache-container name="clustered" default-cache="default" statistics="true">
...
<replicated-cache-configuration name="sessions-cfg" mode="ASYNC" start="EAGER" batching="false">
<transaction mode="NON_XA" locking="PESSIMISTIC"/>
</replicated-cache-configuration>
<replicated-cache name="work" configuration="sessions-cfg" />
<replicated-cache name="sessions" configuration="sessions-cfg" />
<replicated-cache name="offlineSessions" configuration="sessions-cfg" />
<replicated-cache name="actionTokens" configuration="sessions-cfg" />
<replicated-cache name="loginFailures" configuration="sessions-cfg" />
</cache-container>
``` ```
- Start server: 3) Copy the server into the second location referred later as `JDG2_HOME`
4) Start server `jdg1`:
``` ```
cd JDG_HOME/bin cd JDG1_HOME/bin
./standalone.sh -Djboss.socket.binding.port-offset=100 ./standalone.sh -c clustered.xml -Djava.net.preferIPv4Stack=true \
-Djboss.socket.binding.port-offset=1010 -Djboss.default.multicast.address=234.56.78.99 \
-Djboss.node.name=jdg1
```
5) Start server `jdg2`:
```
cd JDG2_HOME/bin
./standalone.sh -c clustered.xml -Djava.net.preferIPv4Stack=true \
-Djboss.socket.binding.port-offset=2010 -Djboss.default.multicast.address=234.56.78.99 \
-Djboss.node.name=jdg2
```
6) There should be message in the log that nodes are in cluster with each other:
```
Received new cluster view for channel clustered: [jdg1|1] (2) [jdg1, jdg2]
``` ```
Keycloak servers setup Keycloak servers setup
---------------------- ----------------------
You need to setup 2 Keycloak nodes in this way. 1) Download Keycloak 3.3.0.CR1 and unzip to some location referred later as `NODE11`
For now, it's recommended to test Keycloak overlay on EAP7 because of infinispan bug, which is fixed in EAP 7.0 (infinispan 8.1.2), but not 2) Configure shared database for KeycloakDS datasource. Recommended to use MySQL, MariaDB or PostgreSQL. See Keycloak docs for more details
yet on Wildfly 10 (infinispan 8.1.0). See below for details.
1) Configure shared database in KEYCLOAK_HOME/standalone/configuration/standalone.xml . For example MySQL 3) Edit `NODE11/standalone/configuration/standalone-ha.xml` :
2) Add `module` attribute to the infinispan keycloak container: 3.1) Add attribute `site` to the JGroups UDP protocol:
``` ```xml
<cache-container name="keycloak" jndi-name="infinispan/Keycloak" module="org.keycloak.keycloak-model-infinispan"> <stack name="udp">
<transport type="UDP" socket-binding="jgroups-udp" site="${jboss.site.name}"/>
``` ```
3) Configure `work` cache to use remoteStore. You should use this: 3.2) Add output-socket-binding for `remote-cache` under `socket-binding-group` element:
```xml
<socket-binding-group ...>
...
<outbound-socket-binding name="remote-cache">
<remote-destination host="localhost" port="${remote.cache.port}"/>
</outbound-socket-binding>
</socket-binding-group>
``` ```
<local-cache name="work">
3.3) Add this `module` attribute under `cache-container` element of name `keycloak` :
```xml
<cache-container name="keycloak" jndi-name="infinispan/Keycloak" module="org.keycloak.keycloak-model-infinispan">
```
3.4) Add the `remote-store` under `work` cache:
```xml
<replicated-cache name="work" mode="SYNC">
<remote-store passivation="false" fetch-state="false" purge="false" preload="false" shared="true" cache="work" remote-servers="remote-cache"> <remote-store passivation="false" fetch-state="false" purge="false" preload="false" shared="true" cache="work" remote-servers="remote-cache">
<property name="rawValues">true</property> <property name="rawValues">true</property>
<property name="marshaller">org.keycloak.cluster.infinispan.KeycloakHotRodMarshallerFactory</property> <property name="marshaller">org.keycloak.cluster.infinispan.KeycloakHotRodMarshallerFactory</property>
</remote-store> </remote-store>
</local-cache> </replicated-cache>
``` ```
4) Configure connection to the external JDG server. Because we used port offset 100 for JDG (see above), the HotRod endpoint is running on 11322 . 3.5) Add the `store` like this under `sessions` cache:
So add the config like this to the bottom of standalone.xml under `socket-binding-group` element:
``` ```xml
<outbound-socket-binding name="remote-cache"> <distributed-cache name="sessions" mode="SYNC" owners="1">
<remote-destination host="localhost" port="11322"/> <store class="org.keycloak.models.sessions.infinispan.remotestore.KeycloakRemoteStoreConfigurationBuilder" passivation="false" fetch-state="false" purge="false" preload="false" shared="true">
</outbound-socket-binding> <property name="remoteCacheName">sessions</property>
<property name="useConfigTemplateFromCache">work</property>
</store>
</distributed-cache>
``` ```
5) Optional: Configure logging in standalone.xml to see what invalidation events were send: 3.6) Same for `offlineSessions` and `loginFailures` caches:
````
```xml
<distributed-cache name="offlineSessions" mode="SYNC" owners="1">
<store class="org.keycloak.models.sessions.infinispan.remotestore.KeycloakRemoteStoreConfigurationBuilder" passivation="false" fetch-state="false" purge="false" preload="false" shared="true">
<property name="remoteCacheName">offlineSessions</property>
<property name="useConfigTemplateFromCache">work</property>
</store>
</distributed-cache>
<distributed-cache name="loginFailures" mode="SYNC" owners="1">
<store class="org.keycloak.models.sessions.infinispan.remotestore.KeycloakRemoteStoreConfigurationBuilder" passivation="false" fetch-state="false" purge="false" preload="false" shared="true">
<property name="remoteCacheName">loginFailures</property>
<property name="useConfigTemplateFromCache">work</property>
</store>
</distributed-cache>
```
3.7) The configuration of distributed cache `authenticationSessions` and other caches is left unchanged.
3.8) Optionally enable DEBUG logging under `logging` subsystem:
```xml
<logger category="org.keycloak.cluster.infinispan"> <logger category="org.keycloak.cluster.infinispan">
<level name="TRACE"/> <level name="DEBUG"/>
</logger>
<logger category="org.keycloak.connections.infinispan">
<level name="DEBUG"/>
</logger> </logger>
<logger category="org.keycloak.models.cache.infinispan"> <logger category="org.keycloak.models.cache.infinispan">
<level name="DEBUG"/> <level name="DEBUG"/>
</logger> </logger>
<logger category="org.keycloak.models.sessions.infinispan">
<level name="DEBUG"/>
</logger>
```
4) Copy the `NODE11` to 3 other directories referred later as `NODE12`, `NODE21` and `NODE22`.
5) Start `NODE11` :
```
cd NODE11/bin
./standalone.sh -c standalone-ha.xml -Djboss.node.name=node11 -Djboss.site.name=site1 \
-Djboss.default.multicast.address=234.56.78.100 -Dremote.cache.port=12232 -Djava.net.preferIPv4Stack=true \
-Djboss.socket.binding.port-offset=3000
```
6) Start `NODE12` :
````
cd NODE12/bin
./standalone.sh -c standalone-ha.xml -Djboss.node.name=node12 -Djboss.site.name=site1 \
-Djboss.default.multicast.address=234.56.78.100 -Dremote.cache.port=12232 -Djava.net.preferIPv4Stack=true \
-Djboss.socket.binding.port-offset=4000
```` ````
6) Setup Keycloak node2 . Just copy Keycloak to another location on your laptop and repeat steps 1-5 above for second server too. The cluster nodes should be connected. This should be in the log of both NODE11 and NODE12:
7) Run server 1 with parameters like (assuming you have virtual hosts "node1" and "node2" defined in your `/etc/hosts` ):
```
./standalone.sh -Djboss.node.name=node1 -b node1 -bmanagement node1
```
and server2 with:
```
./standalone.sh -Djboss.node.name=node2 -b node2 -bmanagement node2
```
8) Note something like this in both `KEYCLOAK_HOME/standalone/log/server.log` on both nodes. Note that cluster Startup Time will be same time on both nodes:
```
2016-11-16 22:12:52,080 DEBUG [org.keycloak.cluster.infinispan.InfinispanClusterProviderFactory] (ServerService Thread Pool -- 62) My address: node1-1953169551
2016-11-16 22:12:52,081 DEBUG [org.keycloak.cluster.infinispan.CrossDCAwareCacheFactory] (ServerService Thread Pool -- 62) RemoteStore is available. Cross-DC scenario will be used
2016-11-16 22:12:52,119 DEBUG [org.keycloak.cluster.infinispan.InfinispanClusterProviderFactory] (ServerService Thread Pool -- 62) Loaded cluster startup time: Wed Nov 16 22:09:48 CET 2016
2016-11-16 22:12:52,128 DEBUG [org.keycloak.cluster.infinispan.InfinispanNotificationsManager] (ServerService Thread Pool -- 62) Added listener for HotRod remoteStore cache: work
```
9) Login to node1. Then change any realm on node2. You will see in the node2 server.log that RealmUpdatedEvent was sent and on node1 that this event was received.
This is done even if node1 and node2 are NOT in cluster as it's the external JDG used for communication between 2 keycloak servers and sending/receiving cache invalidation events. But note that userSession
doesn't yet work (eg. if you login to node1, you won't see the userSession on node2).
WARNING: Previous steps works on Keycloak server overlay deployed on EAP 7.0 . With deploy on Wildfly 10.0.0.Final, you will see exception
at startup caused by the bug https://issues.jboss.org/browse/ISPN-6203 .
There is a workaround to add this line into KEYCLOAK_HOME/modules/system/layers/base/org/wildfly/clustering/service/main/module.xml :
``` ```
<module name="org.infinispan.client.hotrod"/> Received new cluster view for channel hibernate: [node11|1] (2) [node11, node12]
``` ```
7) Start `NODE21` :
```
cd NODE21/bin
./standalone.sh -c standalone-ha.xml -Djboss.node.name=node21 -Djboss.site.name=site2 \
-Djboss.default.multicast.address=234.56.78.101 -Dremote.cache.port=13232 -Djava.net.preferIPv4Stack=true \
-Djboss.socket.binding.port-offset=5000
```
It shouldn't be connected to the cluster with `NODE11` and `NODE12`, but to separate one:
```
Received new cluster view for channel hibernate: [node21|0] (1) [node21]
```
8) Start `NODE22` :
```
cd NODE22/bin
./standalone.sh -c standalone-ha.xml -Djboss.node.name=node22 -Djboss.site.name=site2 \
-Djboss.default.multicast.address=234.56.78.101 -Dremote.cache.port=13232 -Djava.net.preferIPv4Stack=true \
-Djboss.socket.binding.port-offset=6000
```
It should be in cluster with `NODE21` :
```
Received new cluster view for channel server: [node21|1] (2) [node21, node22]
```
9) Test:
9.1) Go to `http://localhost:11080/auth/` and create initial admin user
9.2) Go to `http://localhost:11080/auth/admin` and login as admin to admin console
9.3) Open 2nd browser and go to any of nodes `http://localhost:12080/auth/admin` or `http://localhost:13080/auth/admin` or `http://localhost:14080/auth/admin` . After login, you should be able to see
the same sessions in tab `Sessions` of particular user, client or realm on all 4 servers
9.4) After doing any change (eg. update some user), the update should be immediatelly visible on any of 4 nodes as caches should be properly invalidated everywhere.
9.5) Check server.logs if needed. After login or logout, the message like this should be on all the nodes `NODEXY/standalone/log/server.log` :
```
2017-08-25 17:35:17,737 DEBUG [org.keycloak.models.sessions.infinispan.remotestore.RemoteCacheSessionListener] (Client-Listener-sessions-30012a77422542f5) Received event from remote store.
Event 'CLIENT_CACHE_ENTRY_REMOVED', key '193489e7-e2bc-4069-afe8-f1dfa73084ea', skip 'false'
```
This is just a starting point and the instructions are subject to change. We plan performance improvements especially around performance. If you
have any feedback regarding cross-dc scenario, please let us know on keycloak-user mailing list referred from [Keycloak home page](http://www.keycloak.org/community.html).

2
testsuite/integration-arquillian/servers/wildfly-balancer/src/main/xslt/mod_cluster.xsl
View file

@ -40,7 +40,7 @@
</xsl:template> </xsl:template>
<!--add filter--> <!--add filter-->
<xsl:template match="//*[local-name()='filters']"> <xsl:template match="//*[local-name()='subsystem']/*[local-name()='filters']">
<xsl:copy> <xsl:copy>
<xsl:apply-templates select="node()|@*"/> <xsl:apply-templates select="node()|@*"/>
<mod-cluster <mod-cluster