graphite-dev team mailing list archive

[chrismd <question173171@xxxxxxxxxxxxxxxxxxxxx>]

Question #173171 on Graphite changed:
https://answers.launchpad.net/graphite/+question/173171

chrismd posted a new comment:
Thanks for pointing out those issues, I've fixed the loadRelayRules call
and destinationName issue in trunk. Don't worry about that stacktrace
though, it only affects the reporting of carbon's internal metrics. I
would suggest trying to identify the resource bottleneck when you're at
400k metrics/minute because its quite possible your system can handle
more (ie. if you're cpu-bound on a daemon that could be sharded into
multiple instances to utilize more cores). If it cannot then clustering
is indeed the way to go, I would also suggest moving away from carbon-
relay.py in favor of having clients send data via carbon-client.py
(which is basically a client-side relay).

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