What's New in Grafana v4.2 | Grafana documentation

Important: This documentation is about an older version. It's relevant only to the release noted, many of the features and functions have been updated or replaced. Please view the current version.

Documentation

Docs Home

Guides

What's New in Grafana v4.2

Enterprise Open source

Whats new in Grafana v4.2

Grafana v4.2 Beta is now available for download. Just like the last release this one contains lots bug fixes and minor improvements. We are very happy to say that 27 of 40 issues was closed by pull requests from the community. Big thumbs up!

Release Highlights

Hipchat: Adds support for sending alert notifications to hipchat #6451, thx @jregovic
Telegram: Added Telegram alert notifier #7098, thx @leonoff
LINE: Add LINE as alerting notification channel #7301, thx @huydx
Templating: Make $__interval and $__interval_ms global built in variables that can be used in by any datasource (in panel queries), closes #7190, closes #6582
Alerting: Adds deduping of alert notifications #7632
Alerting: Better information about why an alert triggered #7035
Orgs: Sharing dashboards using Grafana share feature will now redirect to correct org. #6948
Full changelog

New alert notification channels

This release adds five new alert notifications channels, all of them contributed by the community.

Hipchat
Telegram
LINE
Pushover
Threema

Templating

We added two new global built in variables in grafana. $__interval and $__interval_ms are now reserved template names in grafana and can be used by any datasource. We might add more global built in variables in the future and if we do we will prefix them with $__. So please avoid using that in your template variables.

Dedupe alert notifications when running multiple servers

In this release we will dedupe alert notificiations when you are running multiple servers. This makes it possible to run alerting on multiple servers and only get one notification.

We currently solve this with sql transactions which puts some limitations for how many servers you can use to execute the same rules. 3-5 servers should not be a problem but as always, it depends on how many alerts you have and how frequently they execute.

Next up for a better HA situation is to add support for workload balancing between Grafana servers.

Alerting more info

You can now see the reason why an alert triggered in the alert history. Its also easier to detect when an alert is set to alerting due to the no_data option.

Improved support for multi-org setup

When loading dashboards we now set an query parameter called orgId. So we can detect from which org an user shared a dashboard. This makes it possible for users to share dashboards between orgs without changing org first.

We aim to introduce dashboard groups sometime in the future which will introduce access control and user groups within one org. Making it possible to have users in multiple groups and have detailed access control.

Upgrade & Breaking changes

If your using https in grafana we now force you to use tls 1.2 and the most secure ciphers. We think its better to be secure by default rather then making it configurable. If you want to run https with lower versions of tls we suggest you put a reserve proxy in front of grafana.

If you have template variables name $__interval or $__interval_ms they will no longer work since these keywords are reserved as global built in variables. We might add more global built in variables in the future and if we do, we will prefix them with $__. So please avoid using that in your template variables.