graphite-dev team mailing list archive

Thread
Date

[Question #173457]: Intermittently broken aggregation

To: graphite-dev@xxxxxxxxxxxxxxxxxxx
From: Kevin Clark <question173457@xxxxxxxxxxxxxxxxxxxxx>
Date: Thu, 06 Oct 2011 14:55:42 -0000
Reply-to: question173457@xxxxxxxxxxxxxxxxxxxxx
Sender: bounces@xxxxxxxxxxxxx

New question #173457 on Graphite:
https://answers.launchpad.net/graphite/+question/173457

I've got two boxes with an aggregator and relay each. Services push to one of the two boxes, and then get forwarded to several storage boxes running carbon-cache. The other day I started to see graphs like this:

https://skitch.com/kevinclark/f62n7/overloaded-aggregation

And it turned out the aggregation machine was being overwhelmed. I took some load off and made it a larger instance, and things mostly resolved, but now one of three aggregation stats is still unhappy:

https://skitch.com/kevinclark/f6a9h/aggregation

It's on the same box, and there's no indication in the logs that it can't keep up. Any ideas?

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