Hello it seems we keep running into a timeout after trying to look back at a days work of data. It works fine 12 hours or less. Once i try to query 24hrs or more , i get a time out in Kibana. I am using Kibana 6.1.1 . the data set we are looking at is roughly 100gb.
You could consider upping the default value of elasticsearch.requestTimeout from 30000 milliseconds in your kibana.yml. This might allow you use a larger window. Are you using scripted fields? This might be the cause of things taking so long.
is there a recommendation to change the value to, as I dont want to over do it since this then can not time out when there is an actual issue? I am not using scripted fields..
I would maybe try doubling it to 60000 and see if it helps. I would not recommend going above 2 minutes (120000).
ok will give that a go. Thanks!
What is the size of your cluster? What hardware are you using? How many visualisations do you have on the dashboard that is timing out? What type of visualisation?
we have 12 ES nodes 4 LS nodes and 1 kibana Node. ES has 40 cores, 250 gb ram
How many visualisations do you have on the dashboard that is timing out? What type of visualisation?
two visualizations, one is a Timelion one and the other is a table.
Sounds odd. What is the load on the cluster? What does CPU usage, disk I/O and iowait look like?
its all low, 2% CPU load is ~1.03
Anything in the Elasticsearch logs, e.g. long or frequent GC?
i havent touched GC that already comes with ES. So i would assume that is not an issue? Logs are clear of anything else.
Are there any log messages about GC in the logs?
There is nothing about GC
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.