Adding EC2 settings to elasticsearch.yml stops listening to port 9200

Elastic Stack Elasticsearch
1

I am running ES 0.90.4 on an Ubuntu 12.04 LTS EC2 instance.

Starting this instance, it starts listening to port 9200:
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$ sudo lsof -i TCP | grep 9200
java 1982 elasticsearch 102u IPv6 12069 0t0 TCP *:9200 (LISTEN)
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$

If I add the EC2 plugin settings to /etc/elasticsearch/elasticsearch.yml
and restart elasticsearch, it stops listening to port 9200:
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$ sudo lsof -i TCP | grep 9200
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$

If I yank the EC2 settings from /etc/elasticsearch/elasticsearch.yml and
restart elasticsearch, it starts listening to 9200 again.

Is this expected behaviour? Does it start listening to a different port
instead?

The settings that are added to /etc/elasticsearch/elasticsearch.yml are
the settings mentioned in http://www.elasticsearch.org/tutorials/elasticsearch-on-ec2/

The settings are:

cluster.name: myuniquename-cluster
cloud:
aws:
access_key:
secret_key:
discovery:
type: ec2
gateway:
type: s3
s3:
bucket: myuniquename

Thanks!

  • Steinar

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/upzchaa3mdxj.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.

2

How long did you wait after starting with AWS settings before checking the
lsof output? I've found that startup time on AWS can be quite slow, due to
the fact that ES is describing all your instances and looking for other ES
instances to cluster with. If you have a lot of instances, this could take
quite a while.

I'd suggest enabling the logging as described here:

Then watch the logs as the node comes up. There may be some clues there as
to what's going on.

Ross

On Friday, 20 December 2013 14:04:56 UTC-7, Steinar Bang wrote:

I am running ES 0.90.4 on an Ubuntu 12.04 LTS EC2 instance.

Starting this instance, it starts listening to port 9200:
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$ sudo lsof -i TCP | grep
9200
java 1982 elasticsearch 102u IPv6 12069 0t0 TCP *:9200
(LISTEN)
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$

If I add the EC2 plugin settings to /etc/elasticsearch/elasticsearch.yml
and restart elasticsearch, it stops listening to port 9200:
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$ sudo lsof -i TCP | grep
9200
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$

If I yank the EC2 settings from /etc/elasticsearch/elasticsearch.yml and
restart elasticsearch, it starts listening to 9200 again.

Is this expected behaviour? Does it start listening to a different port
instead?

The settings that are added to /etc/elasticsearch/elasticsearch.yml are
the settings mentioned in
Elasticsearch Platform — Find real-time answers at scale | Elastic

The settings are:

cluster.name: myuniquename-cluster
cloud:
aws:
access_key:
secret_key:
discovery:
type: ec2
gateway:
type: s3
s3:
bucket: myuniquename

Thanks!

  • Steinar

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/123a7260-c8cd-4203-bd95-f796a5b3ea69%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

3

The EC2 plugin doesn't mess with networking settings, so the behaviour
you're describing is weird.

After ES starts, can you connect to it via curl localhost:9200?

Also, don't use the now deprecated S3 gateway, apart maybe from strictly
testing purposes -- it's unsupported and unpredictable.

As suggested by Ross, maybe it takes time to ping all the instances in your
AWS account -- do you set the discovery.ec2.groups settings to filter the
list?

Karel

On Friday, December 20, 2013 10:04:56 PM UTC+1, Steinar Bang wrote:

I am running ES 0.90.4 on an Ubuntu 12.04 LTS EC2 instance.

Starting this instance, it starts listening to port 9200:
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$ sudo lsof -i TCP | grep
9200
java 1982 elasticsearch 102u IPv6 12069 0t0 TCP *:9200
(LISTEN)
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$

If I add the EC2 plugin settings to /etc/elasticsearch/elasticsearch.yml
and restart elasticsearch, it stops listening to port 9200:
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$ sudo lsof -i TCP | grep
9200
ubuntu@ip-172-31-31-127:/var/lib/elasticsearch$

If I yank the EC2 settings from /etc/elasticsearch/elasticsearch.yml and
restart elasticsearch, it starts listening to 9200 again.

Is this expected behaviour? Does it start listening to a different port
instead?

The settings that are added to /etc/elasticsearch/elasticsearch.yml are
the settings mentioned in
Elasticsearch Platform — Find real-time answers at scale | Elastic

The settings are:

cluster.name: myuniquename-cluster
cloud:
aws:
access_key:
secret_key:
discovery:
type: ec2
gateway:
type: s3
s3:
bucket: myuniquename

Thanks!

  • Steinar

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/f07a9191-8fea-4660-a38c-85825c5237db%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

4

Karel Minařík karel.minarik@gmail.com:

The EC2 plugin doesn't mess with networking settings, so the behaviour
you're describing is weird.

I got the same behaviour when /var/lib/elasticsearch/elasticsearch was a
dangling symlink to /mnt/var/lib/elasticsearch/elasticsearch (from an
experiment done in a previous run of the EC2 instance with indexing to
/dev/xvdb ), so I thought it might be storage related...?

After ES starts, can you connect to it via curl localhost:9200?

Well, I did
telnet localhost 9200
instead.

But no success there.

Also, don't use the now deprecated S3 gateway, apart maybe from strictly
testing purposes -- it's unsupported and unpredictable.

Ah, Ok. Is there something else I could use instead when I run out of
space on the instance itself? Because I fill up the disk on the
instance pretty quickly.

I did an experiment indexing to the non-persistent /dev/xvdb "disk", and
that worked fine wrt. to storage size and feeding rate. But this isn't
something that can be used with EC2 instances that are stopped and
started on demand.

As suggested by Ross, maybe it takes time to ping all the instances in your
AWS account -- do you set the discovery.ec2.groups settings to filter the
list?

Right! I will try giving it more time.

Thanks!

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/87iouimtt0.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.

5

Ross Simpson simpsora@gmail.com:

How long did you wait after starting with AWS settings before checking the
lsof output?

A minute or two or three, or maybe five. In any case much longer than
the normal ES startup time on the EC2 instance.

I've found that startup time on AWS can be quite slow, due to the fact
that ES is describing all your instances and looking for other ES
instances to cluster with. If you have a lot of instances, this could
take quite a while.

Ok, thanks! I will try giving it a bit longer time and see what happens.

I'd suggest enabling the logging as described here:
Elasticsearch Platform — Find real-time answers at scale | Elastic
Then watch the logs as the node comes up. There may be some clues there as
to what's going on.

Thanks for the heads up! Looking at the verbosity logging was my next
angle of approach, becuse what I found in the logs wasn't very
informative. It looked like a successful startup (except nothing was
listening on port 9200).

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/87eh56mtm7.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.

6

I got the same behaviour when /var/lib/elasticsearch/elasticsearch was a
dangling symlink (...) so I thought it might be storage related...?

Hmm, still, Elasticsearch should complain in the logs when the data dir is
not accessible. Might be related to S3 gateway...

Is there something else I could use instead when I run out of
space on the instance itself? Because I fill up the disk on the
instance pretty quickly.

Very good option on EC2 are EBS volumes, notably the IOPS (optimized or
high I/O) ones. You can easily create snapshots, unmount the volume and
mount it at different instance, etc.

(Some people also use ephemeral disks (which are huge for reasonably
large instances), and simply use lots of replicas to guard against data
loss...)

Karel

P. S.

The Elasticsearch Platform — Find real-time answers at scale | Elastic tutorial,
though quite old, is well worth thorough reading for understanding all the
mechanics on EC2. There's another
tutorial, Elasticsearch Platform — Find real-time answers at scale | Elastic,
which automates many of the steps with Chef.

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/c33370fe-7ec4-4a2a-941a-b77cd8a3203c%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

7

Steinar Bang sb@dod.no:
Ross Simpson simpsora@gmail.com:

Ok, thanks! I will try giving it a bit longer time and see what
happens.

I gave it 1h 40minutes last night, but all that came out in the log file
was:
[2013-12-21 19:39:42,205][INFO ][node ] [Stem Cell] version[0.90.7], pid[1317], build[36897d0/2013-11-13T12:06:54Z]
[2013-12-21 19:39:42,205][INFO ][node ] [Stem Cell] initializing ...
[2013-12-21 19:39:42,229][INFO ][plugins ] [Stem Cell] loaded [river-datagen, cloud-aws], sites [head]
[2013-12-21 19:39:44,591][DEBUG][discovery.zen.ping.multicast] [Stem Cell] using group [224.2.2.4], with port [54328], ttl [3], and address [null]
[2013-12-21 19:39:44,596][DEBUG][discovery.zen.ping.unicast] [Stem Cell] using initial hosts , with concurrent_connects [10]

The last two lines were new after I changed the logging config. All of
the lines were written immediately after startup, and nothing more was
written.

At the end of the 1h40 minutes, nothing was listening on port 9200.

I'd suggest enabling the logging as described here:
Elasticsearch Platform — Find real-time answers at scale | Elastic
Then watch the logs as the node comes up. There may be some clues
there as to what's going on.

Here are the logging changes I did:

*** logging.yml 2013/12/21 19:21:25 1.1
--- logging.yml 2013/12/22 16:50:32 1.2

*** 8,21 ****
com.amazonaws: WARN

# gateway

! #gateway: DEBUG
#index.gateway: DEBUG

# peer shard recovery
#indices.recovery: DEBUG

# discovery

! #discovery: TRACE

index.search.slowlog: TRACE, index_search_slow_log_file
index.indexing.slowlog: TRACE, index_indexing_slow_log_file

--- 8,22 ----
com.amazonaws: WARN

# gateway

! gateway: DEBUG
#index.gateway: DEBUG

  • org.apache: WARN

    peer shard recovery

    #indices.recovery: DEBUG

    discovery

! discovery: TRACE

index.search.slowlog: TRACE, index_search_slow_log_file
index.indexing.slowlog: TRACE, index_indexing_slow_log_file

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/upzc8uvcn7j8.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.

8

Steinar Bang sb@dod.no:

I gave it 1h 40minutes last night, but all that came out in the log file
was:
[2013-12-21 19:39:42,205][INFO ][node ] [Stem Cell] version[0.90.7], pid[1317], build[36897d0/2013-11-13T12:06:54Z]
[2013-12-21 19:39:42,205][INFO ][node ] [Stem Cell] initializing ...
[2013-12-21 19:39:42,229][INFO ][plugins ] [Stem Cell] loaded [river-datagen, cloud-aws], sites [head]
[2013-12-21 19:39:44,591][DEBUG][discovery.zen.ping.multicast] [Stem Cell] using group [224.2.2.4], with port [54328], ttl [3], and address [null]
[2013-12-21 19:39:44,596][DEBUG][discovery.zen.ping.unicast] [Stem Cell] using initial hosts , with concurrent_connects [10]

The last two lines were new after I changed the logging config. All of
the lines were written immediately after startup, and nothing more was
written.

I removed the S3 settings and after 1h30min nothing is still listening
on port 9200.

One more line was written to the log (all lines immediately after
startup):
[2013-12-22 17:07:43,419][INFO ][node ] [Destiny] version[0.90.7], pid[1255], build[36897d0/2013-11-13T12:06:54Z]
[2013-12-22 17:07:43,420][INFO ][node ] [Destiny] initializing ...
[2013-12-22 17:07:43,444][INFO ][plugins ] [Destiny] loaded [river-datagen, cloud-aws], sites [head]
[2013-12-22 17:07:45,813][DEBUG][discovery.zen.ping.multicast] [Destiny] using group [224.2.2.4], with port [54328], ttl [3], and address [null]
[2013-12-22 17:07:45,818][DEBUG][discovery.zen.ping.unicast] [Destiny] using initial hosts , with concurrent_connects [10]
[2013-12-22 17:07:46,971][DEBUG][gateway.local ] [Destiny] using initial_shards [quorum], list_timeout [30s]

I removed the EC2 discovery settings and the log filled up with lots of
stuff, including attempted start of my river (failing because nothing
was listening on the expected port). ES was listening on port 9200 when
I tested.

I guess I will have to read the EC2 document closely and try to get a
feeling for what it is supposed to work (amazon EC2 instances are a new
experience for me).

Thanks!

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/upzc4n60n2qo.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.

9

Karel Minařík karel.minarik@gmail.com:

Very good option on EC2 are EBS volumes, notably the IOPS (optimized
or high I/O) ones. You can easily create snapshots, unmount the volume
and mount it at different instance, etc.

Ok, I will study up on EBS volumes. Thanks!

(Some people also use ephemeral disks (which are huge for reasonably
large instances), and simply use lots of replicas to guard against
data loss...)

I only have access to a single EC2 instance for doing a small test, and
need to stop and start it on demand, so unfortunately this is not an
option.

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/upzczjnslnyt.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.

10

What is river-datagen?

Note that multicast should be disabled by cloud-aws plugin. So this line should not appear:

using group [224.2.2.4], with port [54328], ttl [3], and address [null]

Wondering if you have something wrong in elasticsearch.yml file.

--
David :wink:
Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs

Le 22 déc. 2013 à 19:45, Steinar Bang sb@dod.no a écrit :

Steinar Bang sb@dod.no:

I gave it 1h 40minutes last night, but all that came out in the log file
was:
[2013-12-21 19:39:42,205][INFO ][node ] [Stem Cell] version[0.90.7], pid[1317], build[36897d0/2013-11-13T12:06:54Z]
[2013-12-21 19:39:42,205][INFO ][node ] [Stem Cell] initializing ...
[2013-12-21 19:39:42,229][INFO ][plugins ] [Stem Cell] loaded [river-datagen, cloud-aws], sites [head]
[2013-12-21 19:39:44,591][DEBUG][discovery.zen.ping.multicast] [Stem Cell] using group [224.2.2.4], with port [54328], ttl [3], and address [null]
[2013-12-21 19:39:44,596][DEBUG][discovery.zen.ping.unicast] [Stem Cell] using initial hosts , with concurrent_connects [10]

The last two lines were new after I changed the logging config. All of
the lines were written immediately after startup, and nothing more was
written.

I removed the S3 settings and after 1h30min nothing is still listening
on port 9200.

One more line was written to the log (all lines immediately after
startup):
[2013-12-22 17:07:43,419][INFO ][node ] [Destiny] version[0.90.7], pid[1255], build[36897d0/2013-11-13T12:06:54Z]
[2013-12-22 17:07:43,420][INFO ][node ] [Destiny] initializing ...
[2013-12-22 17:07:43,444][INFO ][plugins ] [Destiny] loaded [river-datagen, cloud-aws], sites [head]
[2013-12-22 17:07:45,813][DEBUG][discovery.zen.ping.multicast] [Destiny] using group [224.2.2.4], with port [54328], ttl [3], and address [null]
[2013-12-22 17:07:45,818][DEBUG][discovery.zen.ping.unicast] [Destiny] using initial hosts , with concurrent_connects [10]
[2013-12-22 17:07:46,971][DEBUG][gateway.local ] [Destiny] using initial_shards [quorum], list_timeout [30s]

I removed the EC2 discovery settings and the log filled up with lots of
stuff, including attempted start of my river (failing because nothing
was listening on the expected port). ES was listening on port 9200 when
I tested.

I guess I will have to read the EC2 document closely and try to get a
feeling for what it is supposed to work (amazon EC2 instances are a new
experience for me).

Thanks!

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/upzc4n60n2qo.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/4A129CF1-6742-4884-BFB5-2DF29328755B%40pilato.fr.
For more options, visit https://groups.google.com/groups/opt_out.

11

David Pilato david@pilato.fr:

Wondering if you have something wrong in elasticsearch.yml file.

Except for adding the EC2 stuff from the EC2 page, it is the file of the
.deb package:
Elasticsearch Platform — Find real-time answers at scale | Elastic

Also, if I don't attempt to use either EC2 discovery or S3 gateway, ES
works fine.

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/87lhzclfyq.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.

12

So there is probably something wrong with your elasticsearch.yml file.
Could you gist it?

Preserve formating, just replace aws credentials with XXXX.

--
David :wink:
Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs

Le 22 déc. 2013 à 22:43, Steinar Bang sb@dod.no a écrit :

David Pilato david@pilato.fr:

Wondering if you have something wrong in elasticsearch.yml file.

Except for adding the EC2 stuff from the EC2 page, it is the file of the
.deb package:

Also, if I don't attempt to use either EC2 discovery or S3 gateway, ES
works fine.

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/87lhzclfyq.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/659CF695-EEC7-4B83-9AAA-BA0F7C507FD6%40pilato.fr.
For more options, visit https://groups.google.com/groups/opt_out.

13

David Pilato david@pilato.fr:

So there is probably something wrong with your elasticsearch.yml file.
Could you gist it?

Everything seems to be commented out, except for the stuff I've added in
from the EC2 document...?

Preserve formating, just replace aws credentials with XXXX.

Done. I also switched the name of the cluster and the S3 bucket.

Leaving in place either of the clustring settings or the S3 bucket
settings causes the startup problems.

Could the problem be a mismatch between the EC2 plugin and the current
ES...?

Oh, and I'm using OpenJDK instead of Oracle Java. Is that a potential
complication?

Thanks!

  • Steinar

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/87a9frlwej.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.

14

Could you try to add to elasticsearch.yml file:

cloud.enabled: true

and restart?

Also, your logs show that you are using elasticsearch 0.90.7, right?
Did you install was plugin 1.16.0?

bin/plugin -install elasticsearch/elasticsearch-cloud-aws/1.16.0

--
David Pilato | Technical Advocate | Elasticsearch.com
@dadoonet | @elasticsearchfr

Le 23 décembre 2013 at 11:00:35, Steinar Bang (sb@dod.no) a écrit:

David Pilato david@pilato.fr:

So there is probably something wrong with your elasticsearch.yml file.
Could you gist it?

Everything seems to be commented out, except for the stuff I've added in
from the EC2 document...?

Preserve formating, just replace aws credentials with XXXX.

Done. I also switched the name of the cluster and the S3 bucket.

Leaving in place either of the clustring settings or the S3 bucket
settings causes the startup problems.

Could the problem be a mismatch between the EC2 plugin and the current
ES...?

Oh, and I'm using OpenJDK instead of Oracle Java. Is that a potential
complication?

Thanks!

  • Steinar

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/87a9frlwej.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/etPan.52b8158d.5ff87e05.111%40MacBook-Air-de-David.local.
For more options, visit https://groups.google.com/groups/opt_out.

15

David Pilato david@pilato.fr:

Also, your logs show that you are using elasticsearch 0.90.7, right?

Correct.

Did you install was plugin 1.16.0?

No, I had aws plugin 1.4.0.

bin/plugin -install elasticsearch/elasticsearch-cloud-aws/1.16.0

I did that and now it runs and listens on port 9200 with the EC2
clustring and S3 gateway settings in place.

Thanks!

(actually I first uninstalled 1.4.0. I don't know if that's necessary
or not).

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/87eh52iqp4.fsf%40dod.no.
For more options, visit https://groups.google.com/groups/opt_out.