Turns out it was the nodename that was the culprit. The suffix part of the
nodename, @cluster.example.com, needs to be resolvable to that specific host
for ping to work. So had to switch to n...@node.cluster.example.com. Though it
looks strange, at least the service boots up properly.
Attempted
Just to add to this, seems like even after waiting for 15 mins, the start
script fails because it is unable to ping the node. Tracing through the
scripts, we arrive at the nodetool in /usr/lib/riak/erts-5.10.3/bin which is
calling {net_kernel:hidden_connect_node(TargetNode), net_adm:ping(Target
System: Ubuntu 14.04.2 LTS.
From a different vm (10.IP.IP.IP2) after exposing 8087 on 10.IP.IP.IP1:
1> {ok, Pid} = riakc_pb_socket:start(“10.IP.IP.IP1", 8087).
{ok,<0.35.0>}
2> riakc_pb_socket:is_connected(Pid).
true
3>
On 4/15/15, 5:53 PM, "Jon Meredith"
mailto:jmered...@basho.com>> wrote:
Wh
What distribution/ version of Linux are you running on the VM Shawn?
On Wed, Apr 15, 2015 at 5:11 PM, Shawn Debnath wrote:
> Hi there,
>
> Building out a new cluster (for the first time) and package cloud pushed
> down 2.1.0-1. I have gone through and installed all the necessary packages,
> c
To add, this is running in a VM in its entirety (i.e. not docker containers)
and I am using the Erlang distribution that comes packaged with riak. Java 7
SDK was the only thing that I installed separately.
On 4/15/15, 4:11 PM, "Shawn Debnath"
mailto:sh...@debnath.net>> wrote:
Hi there,
Buildi
