Graphite remote-storage in Prometheus

[zybjcdl]

Hi all:

Our current monitoring system is using collectd+graphite+seyren+pagerduty structure.

Now the service will be changed to delivered in kubernetes. So we want to use prometheus as the monitoring system.

But we still want to use graphite + seyren+ pagerduty system to keep the monitoring system consistent as before.

So I wondered instead of using prometheus + alertmanager

Could we get the data from prometheus ,and put it to graphite? That will be prometheus + graphite + seyren+ pagerduty.

I see from here
https://github.com/prometheus/prometheus/blob/b66ce9c5cd4a0cc624d9d29da9c6733d1236f382/CHANGELOG.md

There is one feature in 2016-01-18:

0.16.2 / 2016-01-18

• [FEATURE] Add Graphite remote-storage integration

Does that mean we could store prometheus data into graphite?

If so, how could we use this feature?

Please help! Thank you so much!

[dominikschulz]

Looking at the code I'd say it should be as easy as running Prometheus with these options:
-storage.local.engine=none -storage.remote.graphite-address=graphiteHost:graphitePort.

This will completely disable local prometheus storage. If you still want local storage for e.g. queries you can omit the storage.local.engine option.

[zybjcdl]

Thanks dominikschulz. it works.

[zybjcdl]

@dominikschulz One more question. Is there any limitation on this option?
I find when the graphite is in the same cluster as prometheus, this option works.
But if the graphite is in another cluster, this option doesn't work.

[brian-brazil]

There shouldn't be any such limit.

[infa-ddeore]

-storage.local.engine=none is removed in 2.x versions, seems there is no alternate option in 2.x

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.