1.3.4 startup hangs for 10min and fails

[ibussieres]

I am using elasticsearch 1.3.4 and cloud-aws 2.3.0, the startup of elastic search through the default recipe and it hangs, then fails after 10 minute wait. When I run sudo service elasticsearch start from the server, it works fine.

[2014-11-10 21:13:09,569][INFO ][node                     ] [search1.localdomain] version[1.3.4], pid[3811], build[a70f3cc/2014-09-30T09:07:17Z]
[2014-11-10 21:13:09,569][INFO ][node                     ] [search1.localdomain] initializing ...
[2014-11-10 21:13:09,589][INFO ][plugins                  ] [search1.localdomain] loaded [cloud-aws], sites [head]
[2014-11-10 21:13:13,473][INFO ][node                     ] [search1.localdomain] initialized
[2014-11-10 21:13:13,474][INFO ][node                     ] [search1.localdomain] starting ...
[2014-11-10 21:13:13,587][INFO ][transport                ] [search1.localdomain] bound_address {inet[/0.0.0.0:9300]}, publish_address {inet[/10.37.146.237:9300]}
[2014-11-10 21:13:13,605][INFO ][discovery                ] [search1.localdomain] opsworks-elasticsearch/WmE9X-6-RkuyrfO0BYZGCA
[2014-11-10 21:13:17,759][INFO ][cluster.service          ] [search1.localdomain] new_master     [search1.localdomain][WmE9X-6-RkuyrfO0BYZGCA][search1.localdomain][inet[/10.37.146.237:9300]]{aws_availability_zone=us-east-1b, max_local_storage_nodes=1}, reason: zen-disco-join (elected_as_master)
[2014-11-10 21:13:17,783][INFO ][http                     ] [search1.localdomain] bound_address {inet[/0.0.0.0:9200]}, publish_address {inet[/10.37.146.237:9200]}
[2014-11-10 21:13:17,783][INFO ][node                     ] [search1.localdomain] started
[2014-11-10 21:13:18,760][INFO ][gateway                  ] [search1.localdomain] recovered [1] indices into cluster_state
[2014-11-10 21:23:13,589][INFO ][node                     ] [search1.localdomain] stopping ...
[2014-11-10 21:23:13,648][INFO ][node                     ] [search1.localdomain] stopped
[2014-11-10 21:23:13,648][INFO ][node                     ] [search1.localdomain] closing ...
[2014-11-10 21:23:13,653][INFO ][node                     ] [search1.localdomain] closed

After looking around, I realized this is a regular Chef Command Timeout after 600s. Shouldn't the start and restart commands ignore any timeouts ? It should start elasticsearch in the background, am I wrong ?

[jasonwilson]

Having a similar issue with 1.5 and 2.5.0, were you able to resolve this?