-
Notifications
You must be signed in to change notification settings - Fork 174
Dashboard instant Gateway Timeout #1679
Comments
+1 |
Currently the only real workaround is to increase the default timeout in the web-interface.conf to something higher. |
I can't find a parameter called timeout.DEFAULT. In which .conf file should it be? I am on 1.2.2 |
It's not listed in the default config file. |
OK Thx, that solved the problem. i've set it to 50s |
I was running into a timeout issue with system/indecies requiring me to bump the timeout to 15s. I have roughly 400 32gb indexes ~ 2400 shards This has made large dashboards way less clunky in terms of random errors as well. Thanks @kroepke |
Hi,
Dashboards with many queries over a long time period will lead to an instant Gateway-Timeout until the queries are done.
only log information i get:
When the Dashboard gets opened the Server load is High but it is responding normaly.
Http timeouts are changed to 30s
I cant find a Gateway-Timeout to configure, it looks like there is no Timeout, bcs its popping up instantly.
Update: The Widgets only shows N/A and as error that the API-Call failed, after 2 Seconds it says That the Gateway timed out, and after a Page Refresh the Widgets are filled with Data.
Is there some Option to set the Widget timeout or Web-Interface to Server timeout?
Would it be helpful if you didn't trigger all Widgets at the same time? I have no Problem with waiting a few seconds with a rolling symbol than getting 20 Gateway Timeout messages or even get thrown out to the /disconnect page...
Update 2: If you ran into the same issues
lb_recognition_period_seconds = 0
does the job. The Web Interface then will wait for the Widget to get loaded.BUT the problem of getting thrown to /disconnect page stays.
Update 3: I ran into the same Problem every time, here are some new Stacktraces and my Server config:
Config:
The text was updated successfully, but these errors were encountered: