[ https://issues.jboss.org/browse/JGRP-2253?page=com.atlassian.jira.plugin.... ] Sibin Karnavar reopened JGRP-2253: ---------------------------------- I would like to reopen this case. In AWS environment sometimes the FD_SOCK is not working when we kill the process using kill-9 or while terminating the EC2 instance from the console. Most of the times, It detects the connection breakage immediately with in 3 seconds post crashing the node ( e.g kill -9 -> not gracefully shutting down a node from the cluster). Sometimes, it do not detect immediately. it takes up to 12 to 13 seconds. So I am assuming that FD configurations are helping to detect the node failure instead of FD SOCK in this case. There are no warnings related to FD SOCK in the log files. TCP: Configurations jgroup-ext-addr is the IP address of the each nodes. <TCP external_addr="${jgroup-ext-addr}" bind_addr="match-interface:eth0" bind_port="7803" port_range="0" diagnostics_port="7806" recv_buf_size="20M" send_buf_size="10M" max_bundle_size="64k" enable_diagnostics="true" thread_naming_pattern="cl" thread_pool.enabled="true" thread_pool.min_threads="2" thread_pool.max_threads="16" thread_pool.keep_alive_time="5000" /> FD protocols: <FD_SOCK start_port="7804" external_addr="${jgroup-ext-addr}" bind_addr="match-interface:eth0" client_bind_port="7805"/> <FD timeout="3000" max_tries="3" /> <VERIFY_SUSPECT timeout="3000" /> Thanks -- This message was sent by Atlassian JIRA (v7.5.0#75005)