chart scaling lower part

I’m running a “Debian GNU/Linux 11 (bullseye)”
installed via deb Index of /repos/edge/debian/ bullseye/
deb Index of /repos/repoconfig/debian/ bullseye/

Bare Metal Server with netdata Agent installed netdata v1.40.0-70-nightly

Vhost (web_log_apache_vhosts.requests_by_vhost
I’m sorry, but I have a usage problem or a comprehension problem.
I want to enlarge the lower area, I want to be able to see all domains at once, the scaling unfortunately only works for the upper chart.
I understand the displayed help so that it should work.

vhost-scaling-2

Both:
Firefox 102.12.0esr (64-Bit)
Chrome Version 114.0.5735.198 (Offizieller Build) (64-Bit)
same behavior

Just to be clear, it’s not a deal breaker, something can be worked out, it’s just a bit unattractive that it doesn’t seem to be possible to make the scaling work on the entire chart. Especially if you display a lot of domains it is hard to find the right ones.

Yeah the resizer handle changes the height of the graph essentially.
What I can propose is for you to use cloud, given you connect the agent you are using. The experience there is a bit different since you can hover on a point and check right away in a popover the closest point along with it’s value.

As you can see in the image, I’m hovering on the graph on workers dimension and the popover shows as many dimensions as it can, given my viewport, but tries to focus on my pointer. In this case there are 16 total dimensions, it’s showing 13 (around/closest to my selection) and has more above and below that I can reach/see if a hover higher or lower in the graph.

Thank you for your reply!
For me it is a bit complicated.
I use the agent and only the agent in the private environment for a root server.
Professionally I manage a netdata cloud. Yes, there is much more comfortable.
Unfortunately, it is so that when I click on cloud in my private agent I am immediately recognized that I use the company cloud. This is not so pleasant, so I mix private and professional, which is not very nice. Well I could of course get into the habit of strictly using 2 different browsers. ICh would find it easier if the scaling of the lower area could work. But that’s complaining on a high level, because let’s not kid ourselves, netdata is already really cool.

Ah got it, ok then what I would suggest is to be a bit patient, since we are about to release the cloud-ui on the agent dashboards. As a matter of fact you can play around a bit if you change the pathname of your local agent to /v2 for example instead of seeing localhost:19999 go to localhost:19999/v2.

This is a pre-alpha version and most things don’t work (for example alerts don’t work at all in this version), but if you just want to use the overview of your local agent with the cloud capabilities in place, you can give it a try.

Of course any input will be appreciated. I believe by mid-July it will go live and be the default UI for the local agent.

2 Likes

Looks great, thanx for the preview hint :star_struck:

Both, the pre alpha & the current Cloud are having a little issue.

You asked for comments about the dashboard.
To reproduce the small error, simply search for “cont” or even “containers”. You will only find the entry below Docker, but not Containers &VMs.
I find this strange. Is there a reason for this?

Hey, @Bernd. cgroups.plugin collects containers and VMs metrics (by reading/sys/fs/cgroup/*). So all the metrics names start with cgroup or k8s.cgroup (see cgroups.plugin metrics.csv). That is why nothing is found when using containers.

ok, i understand.
It’s not a big deal, it just caught my eye.
However, the view of Containers & vms is quite ingenious