Hi!
Im running Chronograf beta 7 at the moment.
In the Host List view, I get a lot of servers marked as down. Which is reasonable since those hosts do not exist anymore.
Will old servers stay in the Host List forever, or will they “expire” after a while?
We add and remove servers alot, and just considering if I need to make any changes to Influx, or if the view simply shows just the last 24 hours or similar.
@hasselrot There is a major feature for InfluxDB, tsi, which should help address this problem by not showing older hosts in the active index.
You could also take care of this using Retention Policies and downsampling in the database. Otherwise you are going to need to DROP SERIES FROM "telegraf"."autogen"./.*/ WHERE host = '<host>' to remove the dead hosts.
If you could open an issue for this over on chronograf I know they would appreciate it.
@jackzampolin Thanks alot!
I will read up on that and also create an issue for Chronograf.
Just for sharing, this is the issue I opened: Host lists and autoscaling · Issue #1177 · influxdata/chronograf · GitHub
Hi,
Is this possible also to have in Grafana if I’m not using TICK Stack with Kapacitor, Cronograf?
My configuration is based on: Telegraf + InfluxDB + Grafana
[Server side]
telegraf-1.5.0-1.x86_64
grafana-4.6.3-1.x86_64
influxdb-1.4.2-1.x86_64
I’m trying something like in Nagios | Centreon, to have on home page all Services that are in WARNING/ CRITICAL status, somehow a filtering in order to see only the problems.
It is feasible/ possible to have on the same Dashboard(with template or without template) where you have the services KPIs a top row where you can have some notifications panels (singlestat, etc) where you can see the services in WARNING/ CRITICAL status?
Or is needed to have a separate Dashboard for this?
Any recommendation or docs for this, really appreciated.
Kind Regards,
