Prometheus AlertManager vs Grafana v.4.0.2 Alerting

[Oleg Mayko]

Hello All, 

We are working on monitoring setup and check the eco-system for Prometheus. 

With the release of the new grafana alerting feature , does it make sense to setup Alertmanager ?

In other words, are there any benefits of usage of Alertmanager instead of Grafana alerts?

Thank you,

Regards,

Oleg 

[Brian Brazil]

Grafana's alerting is focused on the UI being simple first, and the features it offers are thus quite basic right now. It only allows for simple threshold alerts based on a single metric, whereas Prometheus offers more sophisticated possibilities via PromQL. The alertmanager supports alert aggregation into a notification, generic way to get alerts out to systems not directly supported, HA, silencing and so on.

If you've very simple alerting needs Grafana is an option, for anything more advanced than that Prometheus is a better choice.

--

Brian Brazil

www.robustperception.io

[frederic...@coreos.com]

- Alertmanager can deduplicate, group, silence, inhibit alerts

- Alertmanager can do real high availability

     - Alertmanager/Prometheus communicate alerts with a protocol specifically engineered for high availability

     - Gossip protocol for deduplication with multiple instances

- Separation of concerns, when your dashboard is down, your alerting goes down - not with Alertmanager

- Alertmanager has its configuration in a config file which is much easier to operate, version, apply than a local sqlite database that has to be kept in sync if you have multiple instances of Grafana

Note that some of these things you likely won't hit in an evaluation phase but are very important when you go into production.