[jira] [Commented] (ARTEMIS-2867) NetworkHealthCheck should not cache IPs, and always use DNS

["ASF subversion and git services (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/ARTEMIS-2867?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17172747#comment-17172747 ] 

ASF subversion and git services commented on ARTEMIS-2867:
----------------------------------------------------------

Commit 9842f45a49afb81ff81d16fcb51bd0c0bb717b62 in activemq-artemis's branch refs/heads/master from Clebert Suconic
[ https://gitbox.apache.org/repos/asf?p=activemq-artemis.git;h=9842f45 ]

ARTEMIS-2867 Do not cache IPs on DNS Entries for NetworkHealthCheck

In case of a DNS outage, the pinger should still fail
If we cache the InetAddress this would not be possible


> NetworkHealthCheck should not cache IPs, and always use DNS
> -----------------------------------------------------------
>
>                 Key: ARTEMIS-2867
>                 URL: https://issues.apache.org/jira/browse/ARTEMIS-2867
>             Project: ActiveMQ Artemis
>          Issue Type: Bug
>          Components: Broker
>    Affects Versions: 2.14.0
>            Reporter: Clebert Suconic
>            Priority: Major
>             Fix For: 2.15.0
>
>
> The network health check is using DNS, however it resolves the address and only stores the InetAddress and always use the resolved IP.
>  
> We should instead always use the DNS server, so in case the DNS server is down, the server is down.
>  
> Some users reported split brain after the DNS server was unreachable and while the IPs were ok.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)