You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@activemq.apache.org by "Justin Bertram (Jira)" <ji...@apache.org> on 2019/09/25 16:09:00 UTC
[jira] [Comment Edited] (ARTEMIS-2486) Multiple connector-ref in
broadcast-group is not working
[ https://issues.apache.org/jira/browse/ARTEMIS-2486?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16937880#comment-16937880 ]
Justin Bertram edited comment on ARTEMIS-2486 at 9/25/19 4:08 PM:
------------------------------------------------------------------
As you noted, there *is* an issue in the parsing logic where it won't read multiple {{connector-ref}} elements properly. However, that should be irrelevant at this point as the *need* to configure multiple {{connector-ref}} elements was eliminated years ago. Clients are updated automatically about a node's backup once they connect to a master with no need for static configuration. This bit of logic was leftover from that change and introduced this bug.
bq. I'm trying to pass multiple connectors so I can "publish" both the master and the slave to speed up the failover if the master is dying. (I'm not sure if that's the way to go, but OK)
I don't think this bug is related to the speed (or lack thereof) of failover as any client should already be informed of the backup soon after it connects. I recommend you follow-up on the [ActiveMQ User Mailing List|http://activemq.apache.org/contact/#mailing] or possibly [Stack Overflow|https://stackoverflow.com/questions/tagged/activemq-artemis] and describe your use case, the behavior you expect, the behavior you actually observe, and what you've tried so far to resolve it.
was (Author: jbertram):
As you noted, there *is* an issue in the parsing logic where it won't read multiple {{connector-ref}} elements properly. However, that should be irrelevant at this point as the *need* to configure multiple {{connector-ref}} elements was eliminated years ago. Clients are updated automatically about a node's backup once they connect to a master with no need for static configuration. This bit of logic was leftover from that change and introduced this bug.
bq. I'm trying to pass multiple connectors so I can "publish" both the master and the slave to speed up the failover if the master is dying. (I'm not sure if that's the way to go, but OK)
I don't think this bug is related to the speed (or lack thereof) of failover as any client should already be informed of the backup soon after it connects. I recommend you follow-up on the [ActiveMQ User Mailing List|http://activemq.apache.org/contact/#mailing] or possibly Stack Overflow and describe your use case, the behavior you expect, the behavior you actually observe, and what you've tried so far to resolve it.
> Multiple connector-ref in broadcast-group is not working
> --------------------------------------------------------
>
> Key: ARTEMIS-2486
> URL: https://issues.apache.org/jira/browse/ARTEMIS-2486
> Project: ActiveMQ Artemis
> Issue Type: Bug
> Reporter: Laurent Bigonville
> Priority: Major
>
> In my broadcast group, I'm trying to pass multiple connectors so I can "publish" both the master and the slave to speed up the failover if the master is dying. (I'm not sure if that's the way to go, but OK)
> My configurations looks like:
> {code:xml}
> <connectors>
> <connector name="DL02130V-61616">tcp://dl02130v.example.com:61616</connector>
> <connector name="DL02131V-61616">tcp://dl02131v.example.com:61616</connector>
> </connectors>
> <broadcast-groups>
> <broadcast-group name="bg-group-amq-mainqueues-rel1ldv">
> <local-bind-address>10.122.54.210</local-bind-address>
> <group-address>231.7.7.12</group-address>
> <group-port>9876</group-port>
> <broadcast-period>5000</broadcast-period>
> <connector-ref>DL02130V-61616</connector-ref>
> <connector-ref>DL02131V-61616</connector-ref>
> </broadcast-group>
> </broadcast-groups>
> {code}
> But if I'm inspecting the MBean or looking at the packet on the network, I see that the first connector is broadcasted twice.
> Looking at the code, I'm seeing the following:
> {code:java}
> private void parseBroadcastGroupConfiguration(final Element e, final Configuration mainConfig) {
> String name = e.getAttribute("name");
> List<String> connectorNames = new ArrayList<>();
> NodeList children = e.getChildNodes();
> for (int j = 0; j < children.getLength(); j++) {
> Node child = children.item(j);
> if (child.getNodeName().equals("connector-ref")) {
> String connectorName = getString(e, "connector-ref", null, Validators.NOT_NULL_OR_EMPTY);
> connectorNames.add(connectorName);
> }
> }
> }
> {code}
> Shouldn't
> {code:java}
> String connectorName = getString(e, "connector-ref", null, Validators.NOT_NULL_OR_EMPTY);{code}
> be
> {code:java}
> String connectorName = getString(child, "connector-ref", null, Validators.NOT_NULL_OR_EMPTY);{code}
> instead?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)