[ https://issues.jboss.org/browse/ISPN-8411?page=com.atlassian.jira.plugin.... ] Emond Papegaaij edited comment on ISPN-8411 at 10/17/17 5:31 AM: ----------------------------------------------------------------- [~NadirX] Can you point me to the location of this cluster-wide clear command? The documentation of {{Cache.clear}} states it's local and should not be used: "_Note: This should never be invoked in production unless you can guarantee no other invocations are ran concurrently. If the cache is transactional, it will not interact with the transaction._" However, the implementation seems to execute a ClearCommand and participate in the transaction. Also note that Infinispan's own JCache (JSR 107) implementation uses the same iterate and remove technique: https://github.com/infinispan/infinispan/blob/9.1.1.Final/jcache/remote/s... [~rvansa] Yes, we are using transactional invalidation caches. We took this configuration from the Infinispan 8.2 documentation and the default configuration provided by WildFly (10.1 and 11.0). From those we concluded that transactional caches are advised. Is this wrong? Prior to using Infinispan, we used EH Cache 2, which was read-write. was (Author: papegaaij): [~NadirX] Can you point me to the location of this cluster-wide clear command? The documentation of {{Cache.clear}} states it's local and should not be used: "_Note: This should never be invoked in production unless you can guarantee no other invocations are ran concurrently. If the cache is transactional, it will not interact with the transaction._" However, the implementation seems to execute a ClearCommand and participate in the transaction. Also note that Infinispan's own JCache (JSR 107) implementation uses the same iterate and remove technique: https://github.com/infinispan/infinispan/blob/9.1.1.Final/jcache/remote/s... [~rvansa] Yes, we are using transaction invalidation caches. We took this configuration from the Infinispan 8.2 documentation and the default configuration provided by WildFly (10.1 and 11.0). From those we concluded that transactional caches are advised. Is this wrong? Prior to using Infinispan, we used EH Cache 2, which was read-write. -- This message was sent by Atlassian JIRA (v7.5.0#75005)