Dashboard auto refresh memory usage

When I turn on auto-refresh on any of my dashboards the memory usage just keeps growing. This happens in IE and Chrome and eventually things go bad.

Is this a known problem, or is it something I am causing with the type of visualizations included in my dashboard??

I've just created some sample dashboards with different combinations of Pie Charts, Bar Charts, Line Charts and Searches.....

What version are you on? OS?

I downloaded the latest version about 2 weeks ago.

I have kibana 4.1.2 and elasticsearch 1.7.2 running on ubuntu 14.

I'm viewing the dashboard in chrome and ie on a windows 10 machine.

It might not have been obvious, so just to clarify - the memory issue is in my browser not on the server side.

I can also confirm the issue exists. My machine is windows 7 and I have used Chrome (45.0.2454.101 m), Firefox (40.0.2, 41.0.1) , Internet explorer (v11.0.22 (I can't even open one of my visualisations without gettings an Oops window and it keeps getting into a loop, showing the oops windows and refreshing the screen to show the oops window again)

The chart I'm having problems with is the vertical chart with two filters and showing 1 hour time intervals. The graph is showing a weeks worth of logstash data. If I use the auto refresh the memory usage keeps on increasing, both in chrome and firefox.However if I reload the whole page then the memory usage does not increase.

I can't go live with this bug considering kibana is supposed to be running on our wallboards and auto refreshing the graphs :frowning:

I have the same problem.

Ubuntu
14.04

Kibana
Version 4.1.1
Build 7489
Commit SHA bd4fcb6

It also happens in kibana 4.1.3 efe34565243462e5180fa199e864a3e4e72b5bdb

I followed the contributions instructions to get Kibana working from source

There seems to be an open issue for this, or at least a very similar issue, on GitHub. If you can add information about the versions you are seeing this issue in and what your configuration/dashboards look like it may help in resolving the issue.

Thanks Christian, I've done that.