aurora-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Bill Farner (JIRA)" <>
Subject [jira] [Commented] (AURORA-840) Add an FAQ for cluster operators
Date Thu, 16 Oct 2014 00:22:34 GMT


Bill Farner commented on AURORA-840:

[~dmontauk] the follow-up question from [~kevints] is the next step:

The scheduler uses that zookeeper path (or whatever url you pass to
-mesos_msater_address) to find the mesos master is should register with.

Is there a mesos master running with a command-line that looks something

exec /usr/local/sbin/mesos-master \
    --zk=zk://$ZK_HOST:2181/mesos/master \

A scheduler that is not registered indicates that it can't find a master, or the master is
not replying.  You'll want to confirm that a mesos master is running and announcing itself
at {{/mesos/master}}.  My guess is that is the case, since you don't have log lines like:
I1016 00:03:16.370030 28692 detector.cpp:138] Detected a new leader: (id='1436')
I1016 00:03:16.370208 28700 group.cpp:659] Trying to get '/home/mesos/prod/master/info_0000001436'
in ZooKeeper
I1016 00:03:16.371140 28690 detector.cpp:433] A new leading master (UPID=master@
is detected

> Add an FAQ for cluster operators
> --------------------------------
>                 Key: AURORA-840
>                 URL:
>             Project: Aurora
>          Issue Type: Story
>          Components: Documentation
>            Reporter: Bill Farner
> There are a number of common stumbling blocks to getting a new cluster running, we should
start documenting them to help folks self-serve.  Off the top of my head:
> - Replicated log not initialized
> - Scheduler not registered

This message was sent by Atlassian JIRA

View raw message