[
https://issues.apache.org/jira/browse/ARTEMIS-4189?focusedWorklogId=848293&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-848293
]
ASF GitHub Bot logged work on ARTEMIS-4189:
-------------------------------------------
Author: ASF GitHub Bot
Created on: 01/Mar/23 11:29
Start Date: 01/Mar/23 11:29
Worklog Time Spent: 10m
Work Description: gaohoward commented on code in PR #4385:
URL: https://github.com/apache/activemq-artemis/pull/4385#discussion_r1121564380
##########
artemis-commons/src/main/java/org/apache/activemq/artemis/utils/NetworkUtil.java:
##########
@@ -0,0 +1,84 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ * <p>
+ * http://www.apache.org/licenses/LICENSE-2.0
+ * <p>
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.activemq.artemis.utils;
+
+import java.net.InetAddress;
+import java.net.UnknownHostException;
+
+/**
+ * This is a wrapper class of some InetAddress methods
+ * so we can mock them with static values in test environment
+ */
+public class NetworkUtil {
+
+ public static final InetAddressWrapper[] EMPTY_ADDRESSES = {};
+ public static TestMode testMode = null;
Review Comment:
> I havent looked at the code yet, but have a more general wonder about this
change.
>
> Not all that long ago we effectively made an opposite kind of change in
the failover bits of qpid-jms to resolve an issue where its use of resolved IPs
for hostnames actually broke things. We made it stop resolving failover hosts
to their IPs for comparison and just use the hostname string values as given,
because resolving them like it previously did meant it would often be de-duping
away entirely distinct hostname values that all just happened to resolve to the
same IP, which is a lot more common these days than when the original code was
written, due to use of TLS SNI in general and specifically in cloud
environments (e.g connections hitting some form of gateway that uses SNI to
determine which 'host' is actually being connected to).
It probably is not that complex for this use case. By the time this kind of
comparison is called the hostnames have been compared already. I'm not sure if
broker can be setup to distinguish 2 hostnames as different hosts even if they
are of same ip? I mean can 2 broker in a cluster binding to same ip and same
port?
Issue Time Tracking
-------------------
Worklog Id: (was: 848293)
Time Spent: 50m (was: 40m)
> NettyConnection.isSameTarget() should compare host names by IPs
> ---------------------------------------------------------------
>
> Key: ARTEMIS-4189
> URL: https://issues.apache.org/jira/browse/ARTEMIS-4189
> Project: ActiveMQ Artemis
> Issue Type: Bug
> Components: Broker
> Affects Versions: 2.28.0
> Reporter: Howard Gao
> Assignee: Howard Gao
> Priority: Major
> Time Spent: 50m
> Remaining Estimate: 0h
>
> The method just compares the host names using string equals. However when one
> host name is in IP form (like "10.7.2.2") it should try to resolve the
> addresses of the host name and make sure they are equal if one of the IP
> addresses matches. Otherwise it may return wrong comparison result.
> Also when comparing localhost it should take care of the case where the host
> is absent from the transport configuration.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
