-
Bug
-
Resolution: Not a Bug
-
Blocker
-
None
-
8.1.0.GA-CR1, 8.1.0.Beta, EAP-XP-6.0.0.CR1
-
False
-
-
False
-
Regression
-
-
-
-
-
-
-
-
Update - 20250508
As of today, after applying the suggested changes to the cache configuration, the issue described below (i.e. cache inconsistencies) does not happen anymore, but the test is failing quite regularly due to JBoss EAP pods being unable to connect to the RHDG service after scale up, as described in the latest comments.
The issue summary has been updated and the priority has been increased to Critical as a consequence. As a side note, it also affects JBoss EAP XP 6 test deliverables.
As a side note, the initial number of JBoss EAP replicas has been updated from 1 to 2, but this didn't change anything.
— Original description (OBSOLETE) —
An EAP 8.1.0 Beta + Red Hat Datagrid 8.5.3.GA interoperability test on OpenShift fails intermittently, supposedly due to cache inconsistencies:
org.assertj.core.error.AssertJMultipleFailuresError: Multiple Failures (2 failures) -- failure 1 -- [Get value from session on pod eap-1-bnktg] Expecting: "2" to match pattern: "21" at Eap8WebCacheOffloadedToOperatorRhdgTests.lambda$clusters$2(Eap8WebCacheOffloadedToOperatorRhdgTests.java:224) -- failure 2 -- [Get value from session on pod eap-1-pd4j8] Expecting: "2" to match pattern: "21"
The deployment is built via the EAP Maven plugin with the cloud-default-config layer, plus the web-clustering, ejb, and ejb-dist-cache, and excluding the ejb-local-cache layer.
The infinispan subsystem is configured to connect via HotRod:
/socket-binding-group=standard-sockets/remote-destination-outbound-socket-binding=rhdg:add(host=${env.JDG_HOST}, port=${env.JDG_PORT})
/subsystem=infinispan/remote-cache-container=rhdg-container:add(default-remote-cluster=data-grid-cluster)
/subsystem=infinispan/remote-cache-container=rhdg-container/remote-cluster=data-grid-cluster:add(socket-bindings=[rhdg])
/subsystem=infinispan/cache-container=web/invalidation-cache=rhdg-cache:add()
/subsystem=infinispan/cache-container=web/invalidation-cache=rhdg-cache/store=hotrod:add(remote-cache-container=rhdg-container,fetch-state=false,purge=false,passivation=false,shared=true)
/subsystem=infinispan/cache-container=web:write-attribute(name=default-cache,value=rhdg-cache)
/subsystem=infinispan/remote-cache-container=rhdg-container:write-attribute(name=properties, value={infinispan.client.hotrod.auth_realm=default,infinispan.client.hotrod.use_auth=true,infinispan.client.hotrod.auth_username=${env.CACHE_USERNAME},infinispan.client.hotrod.auth_password=${env.CACHE_PASSWORD},infinispan.client.hotrod.auth_server_name=rhdg-host,infinispan.client.hotrod.sasl_properties.javax.security.sasl.qop=auth,infinispan.client.hotrod.sasl_mechanism=SCRAM-SHA-512,infinispan.client.hotrod.sni_host_name=rhdg-host,infinispan.client.hotrod.ssl_hostname_validation=false,infinispan.client.hotrod.trust_store_path=/var/run/secrets/kubernetes.io/serviceaccount/service-ca.crt,})
The test logic is about creating clusters of both EAP and RHDG services, where an EAP web session cache is offloaded to the RHDG service, and checking the expected values are stored in the cache when accessing them from different pods.
The priority of this issue is set to Major (therefore not a blocker for 8.1.0 Beta) because:
1. the issue is intermittent
2. the overall configuration (layers + infinispan subsystem) could be maybe improved according to the latest versions used.
Feel free to reach out for any additional details.
- is cloned by
-
-
- Closed
-