Project

General

Profile

Actions

Bug #29423

closed

Default task polling is too frequent at scale

Added by Adam Ruzicka over 5 years ago. Updated about 5 years ago.

Status:
Closed
Priority:
Normal
Assignee:
Category:
-
Target version:
-
Fixed in Releases:
Found in Releases:

Description

Cloned from https://bugzilla.redhat.com/show_bug.cgi?id=1817728

The default polling back-off algorithm can still cause an overload of polling during long running tasks:

def poll_intervals
[0.5, 1, 2, 4, 8, 16]
end

when there are large numbers of tasks, you can get 'storms' of status checks during long running tasks.

Updated and expanded array has shown improvement in working conditions when allowing longer tasks a greater time period of checking in:

[0.5, 1, 2, 4, 8, 16, 32, 64, 128, 256, 512, 1024]

We are going to update the default array to the above value and offer a configuration setting so it can be tuned to specific environments.

Actions

Also available in: Atom PDF