graphite-dev team mailing list archive

[Tahir Raza <question249363@xxxxxxxxxxxxxxxxxxxxx>]

New question #249363 on Graphite:
https://answers.launchpad.net/graphite/+question/249363


We are using graphite, collectd, sensu to capture many metrics from various clients that push data into Whisper via RabbitMQ.

We are running the Graphite (0.9.10) in a VM (ESXi host), 8 G ram, 8 CPU, Ubuntu 12.04 

As we render graphs and more graphs, eventually the graphite server stops responding until I restart apache/carbon-cache.
We like to have at least 10-15 graphs that update at least once every 60 sec. These graphs do have complex computations. Just to give an overview, we are monitoring an internal cloud, over 20 nodes, with over 800 VMs and of course many metrics for each VM.
I have been trying to resolve this for over a month now, some help over skype or GoTo Meeting will be appreciated.

I notice from 'iostat -x -d' that disk seems very busy, if thats the case how can I make it better? Graphite page talked about memcache, how do I verify that its being used..

iostat 

Device:         rrqm/s   wrqm/s     r/s     w/s    rkB/s    wkB/s avgrq-sz avgqu-sz   await r_await w_await  svctm  %util
sdc               0.00     0.00    0.00  943.00     0.00  3772.00     8.00   143.48      159.10      0.00      159.10   1.06   100.00




-- 
You received this question notification because you are a member of
graphite-dev, which is an answer contact for Graphite.