Oculus Elasticsearch RequestError IndexMissingException

[patrickshan]

Hi there,
when I tried to open the link behind the data points path in skyline which redirects me to oculus, I always get the IndexMissingException. Is that caused by the cron job which pulls data from skyline to oculus?
This is my configuration file config/config.yml:
root@ip-10-118-150-210:/opt/oculus/config# cat config.yml
results_explain: 0
elasticsearch:
servers:
- "http://127.0.0.1:9201"
- "http://127.0.0.1:9202"
index: "mini" #same as "MINI_NAMESPACE" in skyline settings.py ?
timeout: 30
phrase_slop: 20
scorers:
dtw:
radius: 5
scale_points: 25
euclidian:
scale_points: 25
skyline:
host: "ec2-5-42-4-244.compute-1.amazonaws.com"
port: 6379
listener_port: 1500 #skyline app http server listen port ?
metric_prefix: "metrics" # same as "FULL_NAMESPACE" in skyline settings.py ?
redis:
host: "127.0.0.1"
port: 6379

[jonlives]

@patrick try setting the "index" parameter to "metrics". This is the name of the actual index in ES to use and is unrelated to the skyline namespace.

Your metric_prefix is correct :)

[patrickshan]

hi jonlives, Thanks for your reply. It seems the search function is still broken even after I changed "index".

NoMethodError at /search
undefined method `fingerprint' for nil:NilClass
file: elasticsearch.rb location: get_fingerprint line: 311

I am wondering if the listener_port for skyline is right. Currently I am using the webapp port for skyline in oculus configuration. Is this all right ? Or it should be some other port such as the port for horizon.

And this is the logs for import cron job:
Active ES Server: http://ec2-23-2-1-156.compute-1.amazonaws.com:9201
Next ES Server: http://ec2-23-2-1-156.compute-1.amazonaws.com:9202
Recreating indexes
Creating redis jobs...
Getting unique metric names
Found 16262 metric names
33 workers working
0 process_redis_metrics jobs left to run
Setting active search server to http://ec2-23-2-1-156.compute-1.amazonaws.com:9202
Oculus import finished in 5.792149139 seconds

[jonlives]

@patrickshan apols for the late reply on this one - is it possible for you to check if you have any metrics in elasticsearch after the import has completed?

[banzayats]

I have the same problem. When I click "Collections" or perform any search I get the same error:
(404) {"error":"IndexMissingException[[collections] missing]","status":404}
I used plugin "head" to see if there any data in elasticsearch. Yes there is index "metrics" and it has some metrics.
Besides, I've checked log files:
import.log:

Active ES Server: http://logserver:9200                    
Next ES Server: http://testserver:9200                  
Recreating indexes                                                 
http://testserver:9200: metrics                         
Creating redis jobs...                                             
Getting unique metric names                                        
Found 413 metric names                                             
0 workers working                                                  
0 process_redis_metrics jobs left to run                           
Setting active search server to http://testserver:9200  
Oculus import finished in 5.373652176 seconds

resque_stdout:

I broke on mini.nginx_response_time 
...
I broke on mini.nginx_response_time 

Can someone explain why there are e lot of messages "I broke on mini.nginx_response_time" and why no workers (0) working?

[jonlives]

@banzayats Can you please confirm that you performed the last step in the "Web App Install" section of the README:

"The final step is to log onto the Oculus admin console (http://yourserver:3000/admin, username admin and password admin), then click "Reinitialize Collections." Click OK to confirm."

This step creates and initializes the collections index. If you've performed this step and are still seeing the above error, please let me know.

[banzayats]

Sorry, my fault again.
So I've initialized Collections and it worked!

[jonlives]

Excellent!

[jonlives]

Closing as appears to be solved