[
https://issues.apache.org/jira/browse/AURORA-1785?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15617740#comment-15617740
]
Jing Chen commented on AURORA-1785:
-----------------------------------
We are not able to leverage {color:blue}HttpService{color} by injecting it to
{color:blue}CuratorServiceDiscoveryModule{color} @Provides methods , since the
state is not {color:blue}State.RUNNING{color} at that time of injection.
{color:blue}httpService.getAddress(){color} throws
{color:red}IllegalStateException{color} at line 356 in _JettyServerModule.java_
{quote}Preconditions.checkState(state() == State.RUNNING);{quote}which leads to
failure of election. We are thinking about moving latch creation out of the
constructor of {color:blue}CurratorSingletonService{color} or access
{color:blue} HttpService {color} in a *lazy* way.
> Populate curator latches with scheduler information
> ---------------------------------------------------
>
> Key: AURORA-1785
> URL: https://issues.apache.org/jira/browse/AURORA-1785
> Project: Aurora
> Issue Type: Task
> Reporter: Zameer Manji
> Assignee: Jing Chen
> Priority: Minor
> Labels: newbie
>
> If you look at the mesos ZK node for leader election you see something like
> this:
> {noformat}
> u'json.info_0000000104',
> u'json.info_0000000102',
> u'json.info_0000000101',
> u'json.info_0000000098',
> u'json.info_0000000097'
> {noformat}
> Each of these nodes contains data about the machine contending for
> leadership. It is a JSON serialized {{MasterInfo}} protobuf. This means an
> operator can inspect who is contending for leadership by checking the content
> of the nodes.
> When you check the aurora ZK node you see something like this:
> {noformat}
> u'_c_2884a0d3-b5b0-4445-b8d6-b271a6df6220-latch-0000000774',
> u'_c_86a21335-c5a2-4bcb-b471-4ce128b67616-latch-0000000776',
> u'_c_a4f8b0f7-d063-4df2-958b-7b3e6f666a95-latch-0000000775',
> u'_c_120cd9da-3bc1-495b-b02f-2142fb22c0a0-latch-0000000784',
> u'_c_46547c31-c5c2-4fb1-8a53-237e3cb0292f-latch-0000000780',
> u'member_0000000781'
> {noformat}
> Only the leader node contains information. The curator latches contain no
> information. It is not possible to figure out which machines are contending
> for leadership purely from ZK.
> I think we should attach data to the latches like mesos.
> Being able to do this is invaluable to debug issues if an extra master is
> added to the cluster.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
