graphite-dev team mailing list archive

Thread
Date

[Question #219163]: persistent hashing losing metrics?

To: graphite-dev@xxxxxxxxxxxxxxxxxxx
From: FRLinux <question219163@xxxxxxxxxxxxxxxxxxxxx>
Date: Mon, 14 Jan 2013 16:41:13 -0000
Reply-to: question219163@xxxxxxxxxxxxxxxxxxxxx
Sender: bounces@xxxxxxxxxxxxx

New question #219163 on Graphite:
https://answers.launchpad.net/graphite/+question/219163

I have a setup with 4 physical boxes all using carbon relay with persistent hashing enabled. My understanding is that when a carbon relay becomes unavailable and the carbon caches are also off, the metrics should be taken on by another carbon relay but this does not seem to be the case. Am I missing someting? 

This is my setup, I am curious if I missed anything in the configuration, i cannot find anything on the internet pointing to a potential mistake.

I have a physical load balancer up front which works accordingly when it sees that one relay is down. 

Here is a graph of the current situation: http://frlinux.net/weblog/graphite_hashing_metrics_lost.png

Notice the few weeks gap where two relays were down and only half of the metrics were sent to carbon caches.

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