MySQL Bugs: #56064: Aggregator should send data to the service manager once every 60 sec

Bug #56064	Aggregator should send data to the service manager once every 60 sec
Submitted:	17 Aug 2010 19:40	Modified:	28 Jan 2011 22:49
Reporter:	Diego Medina	Email Updates:
Status:	Closed	Impact on me:	None
Category:	MySQL Enterprise Monitor: Aggregator	Severity:	S2 (Serious)
Version:	all	OS:	Any
Assigned to:	Mark Leith	CPU Architecture:	Any

Description:
Currently the aggregator "aggregates" data in 10 seconds chunks, this should be by default 60 seconds, as this is what the service manager expects

How to repeat:
1- Install and start the service manager, agent and aggregator
2- Enable the query analyzer on the UI
3- Enable the log-level on the aggregator to debug
4- restart the aggr. to create the proper log level
5- send the aggr some json data (see attached file)

$ while true ; do 
  curl -H"Content-Type: application/json" http://agent:mysql@127.0.0.1:14000/cta/aggregate -T data-2.3.txt  ; 
done ;

6- notice how the aggregator will send data every 10 seconds, and not every 60 seconds (looking at the log)

$ grep "46ad1a40-200c-43bc-8a63-35eec19e9aa2.sugarcrm.ba8e4ee06c56967fb06afe12"  mysql-monitor-agent.log  | awk '{print $2 }'

Mark Leith writes: 
Pushed:

 470 Mark Leith	2010-08-19
     Bug#56064 / EM-4733 - Aggregator should send data to the service manager once every 60 seconds
      - added --aggr-flush-interval so that this is configurable, defaults to 60 (seconds)

   modified:
     src/aggr.c
     src/aggr.h
     src/aggr_mainloop.c

Mark Leith writes: 
This should be in a build now, not sure why it was not marked as such.

Diego Medina writes: 
Verified fixed on 2.3.0.2020