Re: cant create cluster on ec2

2014-02-03 Thread David Montgomery
Hi Jonathan Just be clear...you could not cluster? I wili try the lastest stable release cuase I am giving up on 1.0.0RC. Thanks On Sun, Feb 2, 2014 at 10:10 AM, Jonathan Goodwin jonathanwgood...@gmail.com wrote: Hello David and David, I took a look at the logs / configs posted and mine

Re: cant create cluster on ec2

2014-02-02 Thread David Montgomery
If this is what you mean my trace in the logs at org.elasticsearch.common.netty.channel.socket.nio.NioClientBoss.processConnectTimeout(NioClientBoss.java:137) at org.elasticsearch.common.netty.channel.socket.nio.NioClientBoss.process(NioClientBoss.java:83)

Re: cant create cluster on ec2

2014-02-01 Thread Jonathan Goodwin
Hello David and David, I took a look at the logs / configs posted and mine are mostly similar (though i'm not using any rivers) and I'm having the same issue. I've tested on 0.90.9 and 0.90.7, I hadn't tried the 1.0.0RC yet because the logstash website suggested I stick to 0.90.9 If you need

Re: cant create cluster on ec2

2014-01-31 Thread David Montgomery
Hi, In chef I used -N flag which changes the host name when using chef-client. I removed the flag and built again. No difference. I have two nodes running. Both have the name cluster name, ports are 100% wide open...and wow..ES cant talk despite both running in the same region. So...ES...any

Re: cant create cluster on ec2

2014-01-27 Thread David Pilato
Hi David, I think you should have nodes running under the same security group. In that case, you don't have to open 9300-9400 ports to public. If they don't belong to the same group or if you want to reach them from outside ec2 platform (public IP), then you need to set host_type to public_ip.

Re: cant create cluster on ec2

2014-01-27 Thread David Pilato
Could you change to TRACE level discovery package? -- David ;-) Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs Le 27 janv. 2014 à 18:52, David Montgomery davidmontgom...@gmail.com a écrit : Hi, I have the same sec group. Just staying that not having ports open is not an issue. ES is

Re: cant create cluster on ec2

2014-01-27 Thread David Montgomery
Hi, As an fyi..I can telnet into the other machine. Its only ES that does not work despite following the online docs. If ES aws auto discovery does not work l assuming assume that http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/modules-discovery-zen.html will work?

Re: cant create cluster on ec2

2014-01-27 Thread David Montgomery
I assume that means this in the logging.yml fike? # discovery discovery: TRACE I do have it on...what should I be looking for in the logs? On Tuesday, January 28, 2014 3:47:10 AM UTC+8, David Montgomery wrote: Hi, As an fyi..I can telnet into the other machine. Its only ES that does not

Re: cant create cluster on ec2

2014-01-27 Thread David Pilato
Yes. Could you gist it (and remove first keys / secrets of course)? -- David ;-) Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs Le 27 janv. 2014 à 20:53, David Montgomery davidmontgom...@gmail.com a écrit : I assume that means this in the logging.yml fike? # discovery discovery:

Re: cant create cluster on ec2

2014-01-27 Thread David Montgomery
I have to say..it this a know problem? I removed the keys. So.. should the servers still cluster? Wellits not.again given that the two machine can talk to each other via telent, in the same region, same sec group restarted services etcno go..no cluster Sohow do I

Re: cant create cluster on ec2

2014-01-27 Thread David Pilato
Is it a known problem? Not yet. That's the reason I'm asking for logs. I would like to understand what is happening here. So full logs on both nodes would help. The way you set unicast is fine to me. But please, could you share with me (even through Direct mail) your elasticsearch.yml file and

Re: cant create cluster on ec2

2014-01-27 Thread David Montgomery
sure..let me the logs together and will post two files ..one log file for each of the servers. Thanks On Tue, Jan 28, 2014 at 6:25 AM, David Pilato da...@pilato.fr wrote: Is it a known problem? Not yet. That's the reason I'm asking for logs. I would like to understand what is happening here.