[
https://issues.apache.org/jira/browse/FLINK-2821?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15714835#comment-15714835
]
ASF GitHub Bot commented on FLINK-2821:
---------------------------------------
Github user mxm commented on the issue:
https://github.com/apache/flink/pull/2917
Thanks for checking out the code!
>If different parts of the code or the JDK do a subtle change of behavior
(i.e. resolve the InetAddress), then some nodes may have a hostname in the URL,
others an address.
Fair point. Let's remove InetAddress and use a String instead.
>If two machines have a slightly different network configuration
(especially concerning preferences to represent/encode IPv6 addresses) and the
users set the IP address as the JobManager host, then they might create
different Akka URLs and the machines cannot talk to each other again.
A problem of the old method was that hostnames might resolve differently
depending on the container context. So that method was not reliable either.
Using IP addresses will work with the new method. They just have to be
consistent on all node configurations. I don't think that is a problem because
a configuration is usually created once and then copied over to all nodes.
>I am a bit unsure how to proceed from here. Is there any way we can keep
using IP addresses in the Akka URLs? Or does that just inherently not make
sense with "dynamic hostnames" as they are used in container environments?
For this PR to address the core problems of FLINK-2821, we have to avoid
resolving the hostname because the IP address would represent the internal
container address which may be unreachable from the outside; even if it were
resolvable, Akka would drop the messages because of its exact URL match policy.
Using the hostname which now acts as a purely logical address ultimately is a
more reliable way across different network environments. We just have to make
sure we document the new requirement that the JobManager address
(JOB_MANAGER_IPC_ADDRESS) is consistent across all Flink cluster nodes.
> Change Akka configuration to allow accessing actors from different URLs
> -----------------------------------------------------------------------
>
> Key: FLINK-2821
> URL: https://issues.apache.org/jira/browse/FLINK-2821
> Project: Flink
> Issue Type: Bug
> Components: Distributed Coordination
> Reporter: Robert Metzger
> Assignee: Maximilian Michels
>
> Akka expects the actor's URL to be exactly matching.
> As pointed out here, cases where users were complaining about this:
> http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/Error-trying-to-access-JM-through-proxy-td3018.html
> - Proxy routing (as described here, send to the proxy URL, receiver
> recognizes only original URL)
> - Using hostname / IP interchangeably does not work (we solved this by
> always putting IP addresses into URLs, never hostnames)
> - Binding to multiple interfaces (any local 0.0.0.0) does not work. Still
> no solution to that (but seems not too much of a restriction)
> I am aware that this is not possible due to Akka, so it is actually not a
> Flink bug. But I think we should track the resolution of the issue here
> anyways because its affecting our user's satisfaction.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
