graphite-dev team mailing list archive

Thread
Date

[Question #133636]: queue size - writes don't keep up with volume

To: graphite-dev@xxxxxxxxxxxxxxxxxxx
From: bhardy <question133636@xxxxxxxxxxxxxxxxxxxxx>
Date: Thu, 11 Nov 2010 17:31:37 -0000
Reply-to: question133636@xxxxxxxxxxxxxxxxxxxxx
Sender: bounces@xxxxxxxxxxxxx

New question #133636 on Graphite:
https://answers.launchpad.net/graphite/+question/133636

We have two linux servers running both carbon and carbon-relay for redundant storage of data and clustering.  About 200k metrics are received each minute and after about two hours the writes to disk slows considerably.  Its as if the writes cannot keep up with the volume we are receiving and thus the queue continues to increase.  We are using SAN mounts for this purpose.  Is there anything that would cause this issue?  We feel that our carbon.conf entries are pretty generous as to creates and updates.  Please advise.

Lines from the carbon.conf file:
MAX_QUEUE_SIZE = 500000
MAX_CACHE_SIZE = inf
MAX_UPDATES_PER_SECOND = 100000
MAX_CREATES_PER_MINUTE = inf

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