[
https://issues.apache.org/jira/browse/MESOS-6517?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15840124#comment-15840124
]
Avinash Sridharan commented on MESOS-6517:
------------------------------------------
[[email protected]] don't think asking the Framework/agent to set the IP
address/ hostname would make sense here, since, as [~jieyu] mentioned what do
you do if a container is joining multiple networks. I am fine with LOCALHOST
semantics since it is consistent, the only challenge being if the containers
bind to specific IP address in their network namespace. While a corner case,
this is of course a possibility. If we want to address this problem we should
be inferring all the IP addresses within the container's network namespace and
polling on each one of them and reporting health status on each of them. We
fail if we don't have reachability on any of the IP addresses.
> Health checking only on 127.0.0.1 is limiting.
> ----------------------------------------------
>
> Key: MESOS-6517
> URL: https://issues.apache.org/jira/browse/MESOS-6517
> Project: Mesos
> Issue Type: Improvement
> Reporter: Alexander Rukletsov
> Labels: health-check, mesosphere
>
> As of Mesos 1.1.0, HTTP and TCP health checks always use 127.0.0.1 as the
> target IP. This is not configurable. As a result, tasks should listen on all
> interfaces if they want to support HTTP and TCP health checks. However, there
> might be some cases where tasks or containers will end up binding to a
> specific IP address.
> To make health checking more robust we can:
> * look at all interfaces in a given network namespace and do health check on
> all the IP addresses;
> * allow users to specify the IP to health check;
> * deduce the target IP from task's discovery information.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
