[ https://issues.jboss.org/browse/JBIDE-21847?page=com.atlassian.jira.plugi... ] Viacheslav Kabanovich edited comment on JBIDE-21847 at 3/11/16 3:01 PM: ------------------------------------------------------------------------ [~fbricon], is it possible that when IPortForwardable.stop() calls destroy process and returns, local sockets still need some time to get free, and the check for !hasPortInUse() in updatePortForwardingAllowed happens too early? It would be definitely so if table would show 'in use', but it does not, which unfortunately proves nothing because update of cells takes place much later. Added: AbstractOpenShiftBinaryCapability.stop() calls process.destroyForcibly() and javadoc to that method says: {code} * <p>Note: The subprocess may not terminate immediately. * i.e. {@code isAlive()} may return true for a brief period * after {@code destroyForcibly()} is called. This method * may be chained to {@code waitFor()} if needed. {code} That supports my guess. was (Author: scabanovich): [~fbricon], is it possible that when IPortForwardable.stop() calls destroy process and returns, local sockets still need some time to get free, and the check for !hasPortInUse() in updatePortForwardingAllowed happens too early? It would be definitely so if table would show 'in use', but it does not, which unfortunately proves nothing because update of cells takes place much later. -- This message was sent by Atlassian JIRA (v6.4.11#64026)