Add request_timeout to advanced settings

[markwalkom]

Would it be possible to move this into the web based, advanced settings under management?

I'm working with a few hundred PDFs in a containerised environment and changing this in the .yml is a bit of a hassle unless I modify the underlying container template. And I need to change it as using discover times out due to the size of the data.

So this all got me thinking that it would be useful to allow this to be dynamically changed, rather than restarting the service. I know this might be tricky given the timeout is used on service start, but thought I'd ask :)

[wylieconlon]

I believe you can work around this by following the instructions for customizing kibana.yml in our Docker guide: https://www.elastic.co/guide/en/kibana/current/docker.html

Specifically, you could set an environment variable like docker run -e "ELASTICSEARCH_REQUESTTIMEOUT=60000" kibana

[timroes]

We discussed this issue and would close it, since we're not having plans of moving the request timeout into advanced settings. We try to keep settings that are more of an "administrator" perspective for the cluster into the kibana.yml so they can only be changed by the person "in charge" of the setup.

Despite the fact that we're not planning to move that actual setting, we're planning to have better timeout handling in general in the future, and hopefully will be more flexible (and allow users to retry with longer timeouts). You can follow that meta issue for rewriting Courier: #34022. Nevertheless even in that scenario I would expect the kibana.yml to have a final say in limiting this in case the administrator wants to.