"Your query is taking a while" notification prompted too early on dashboard load


Whenever my dashboard has to reload, a little pop-up notification appears after about 5 seconds in the lower-right of the screen that informs me, "Your query is taking a while," and asks if I want to "Run beyond timeout." Indeed I do, so I make that selection, but then sometime later (if the dashboard is still loading), the notification will appear again, asking me to click "Run beyond timeout" again. Here's what the pop-up looks like:

Screen Shot 2020-08-24 at 10.19.26 AM copy

Since it isn't ideal for this pop-up to appear to outside users viewing our dashboards (as they might think something is going wrong), I'm trying to extend the time duration for this notification to appear. To attempt doing this, I extended the elasticsearch.requestTimeout and elasticsearch.shardTimeout settings in the kibana.yml configuration file, and restarted Kibana. However, the problem still persists and doesn't seem to change the delay time for the notification pop up to appear.

Here are the settings I used in my Kibana.yml:

# Time in milliseconds to wait for responses from the back end or Elasticsearch. This value
# must be a positive integer.
elasticsearch.requestTimeout: 3000000

# Time in milliseconds for Elasticsearch to wait for responses from shards. Set to 0 to disable.
elasticsearch.shardTimeout: 0

Am I missing some other timeout setting that controls this? Is there another setting that controls the time at which the pop-up is triggered to appear?

[Version information]: Elastic stack 7.9.0, running on Ubuntu as systemctl service.

We are aware of this issue,: https://github.com/elastic/kibana/issues/75321

Thanks for pointing me to the relevant issue. It looks like this will be resolved in 7.10 or 7.11, so I will await their releases soon.

@wylie Do you know if there are any workarounds for this in the meantime? Does Kibana provide any control of these timeouts?

I don't think Kibana offers any customization of these, but as you can see that is the plan. The only workaround I can think of (which I haven't tested) is to use very short timeouts, so that Elasticsearch requests will time out in less than 30 seconds.

You mean reducing elasticsearch.requestTimeout to below 30s? This would only workaround cases where the dashboard can load under 30s too though, right?

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.