Bug 1340785

Description of problem:
I am using https://docs.openshift.com/enterprise/3.1/using_images/xpaas_images/jws.html and have recognized that the following access log is written 

$ ls
localhost_access_log.2016-05-26.txt
$ pwd
/opt/webserver/logs

I am wondering 
* will the access log be visible inside the EFK stack?
* what will happen when the amount/size of access log is growing overtime. How will docker deal with this?


Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:
I have installed jws in my lab an did some the check:
- access log file:
root@master1 # oc exec jws-app-1-79ven -- cat /opt/webserver/logs/localhost_access_log.2016-05-30.txt
::1 - MMSGkRf7 [30/May/2016:03:25:29 -0400] "GET /manager/jmxproxy/?get=Catalina%3Atype%3DServer&att=stateName HTTP/1.1" 200 64
::1 - MMSGkRf7 [30/May/2016:03:25:39 -0400] "GET /manager/jmxproxy/?get=Catalina%3Atype%3DServer&att=stateName HTTP/1.1" 200 64
::1 - MMSGkRf7 [30/May/2016:03:25:49 -0400] "GET /manager/jmxproxy/?get=Catalina%3Atype%3DServer&att=stateName HTTP/1.1" 200 64
::1 - MMSGkRf7 [30/May/2016:03:25:59 -0400] "GET /manager/jmxproxy/?get=Catalina%3Atype%3DServer&att=stateName HTTP/1.1" 200 64
::1 - MMSGkRf7 [30/May/2016:03:26:09 -0400] "GET /manager/jmxproxy/?get=Catalina%3Atype%3DServer&att=stateName HTTP/1.1" 200 64
....

- No access logs stored in logs on host machine. So, these logs wont be collected by fluentd:

root@worknode2 # grep -R -i jmxproxy ./*
root@worknode2 # pwd
/var/lib

Looks like log rotation for /opt/webserver/logs is not configured as well. System can run out of memory.

Expected results:


Additional info: