HA mode with JDBC_PING shows warning in the logs after migration to 4.8.3 from 3.4.3
by abhishek raghav
Hi
After the migration of keycloak HA configurations from 3.4.3.Final to
4.8.3.Final, I am seeing some WARNINGS on one of the nodes of keycloak
immediately after the keycloak is started with 2 nodes. This occurs after
every time when the cluster is scaled up or whenever infinispan is trying
to update the cluster member list.
I am using JDBC_PING to achieve clustering in keycloak.
Below is the stacktrace -
2019-04-24 12:20:43,687 WARN
>> [org.infinispan.topology.ClusterTopologyManagerImpl]
>> (transport-thread--p18-t2) [dcidqdcosagent08] KEYCLOAK DEV 1.5.RC
>> ISPN000197: Error updating cluster member list:
>> org.infinispan.util.concurrent.TimeoutException: ISPN000476: Timed out
>> waiting for responses for request 1 from dcidqdcosagent02
>
> at
>> org.infinispan.remoting.transport.impl.MultiTargetRequest.onTimeout(MultiTargetRequest.java:167)
>
> at
>> org.infinispan.remoting.transport.AbstractRequest.call(AbstractRequest.java:87)
>
> at
>> org.infinispan.remoting.transport.AbstractRequest.call(AbstractRequest.java:22)
>
> at
>> java.util.concurrent.FutureTask.run(FutureTask.java:266)
>
> at
>> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180)
>
> at
>> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
>
> at
>> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
>
> at
>> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
>
> at java.lang.Thread.run(Thread.java:748)
>
> Suppressed: org.infinispan.util.logging.TraceException
>
> at
>> org.infinispan.remoting.transport.Transport.invokeRemotely(Transport.java:75)
>
> at
>> org.infinispan.topology.ClusterTopologyManagerImpl.confirmMembersAvailable(ClusterTopologyManagerImpl.java:525)
>
> at
>> org.infinispan.topology.ClusterTopologyManagerImpl.updateCacheMembers(ClusterTopologyManagerImpl.java:508)
>
>
Now after I searched, I really did not see anyone reported such error on
keycloak but there is similar bug reported in WILDLFY 14 and is categorized
as a blocker in WILDLFY 14.This bug is already fixed in WILDLFY 15.
https://issues.jboss.org/browse/WFLY-10736?attachmentViewMode=list
Now since keycloak 4.8 is also based on WILDLFY 14, these WARNINGS could be
because of this blocker in WILDFLY 14.
What should I do to get rid this error. Is this really a problem in
keycloak 4.8.3.Final. Did anyone notice any such issue while running
keycloak 4.8.3 in HA mode.
Is there a workaround to fix this.
One more thing we noticed is - It is regarding a property in JDBC_PING
protocol we are using in our 3.4.3 setup i.e. "clear_table_on_view_change"
but it is no more supported in 4.8 version. and thus the JGROUPSPING table
is filled up with lot of stale entries. Is there a workaround to clear the
table after view change in 4.8 also.
Thanks
Abhishek
5 years, 4 months
Further enhancements in token endpoint with SPI
by luke@code-house.org
Dear developers,
I follow mailing list and I been debugging some bits and pieces of Keycloak since several years. Lately I found a great contribution on device flow/auth which involves changes in Token endpoint.
As there are various grant types in OAuth 2 token endpoint is currently conjunction of several IF-ELSE statements. Would it perhaps make sense to separate actual handling of requests to smaller code units which will enable further plug-in ability in OAuth endpoints?
A name which reasonates for me now is TokenEndpointGrantHandlerSPI or similar.
Kind regards,
Łukasz
—
http://code-house.org
5 years, 4 months