Agents not joining/ losing agents when executing scenarios with more number of guests
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
shaker |
New
|
Undecided
|
Unassigned |
Bug Description
When executing tests with a large number of master-slave pairs (>=10), agents fail to join/ are lost leading to error. Hardware performance was verified to make sure that the hardware wasn't a bottleneck. Agent join timeout and loss timeout were also increased but to no use.
The scenario was: (Image-CentOS7)
description:
This scenario launches several pairs of VMs on the same compute node.
VM are plugged into the same private network. Useful for testing performance
degradation when the number of VMs grows.
deployment:
template: l2.hot
accommodation: [pair, single_room, density: 20, compute_nodes: 2]
execution:
tests:
-
title: Iperf TCP
class: iperf_graph
time: 60
-
title: Iperf UDP
class: iperf
udp: 1
threads: 8
bandwidth: 1000M
The debug log shows that several agents were lost.
Reviewed: https:/ /review. openstack. org/199931 /git.openstack. org/cgit/ stackforge/ shaker/ commit/ ?id=760bcfc3a22 ee52aae02463d97 48d4591ff0e70a
Committed: https:/
Submitter: Jenkins
Branch: master
commit 760bcfc3a22ee52 aae02463d9748d4 591ff0e70a
Author: Ilya Shakhat <email address hidden>
Date: Thu Jul 9 10:41:05 2015 +0300
Split out UDP tests
UDP tests produce significant load on controllers, they should
be configured accordingly to scale of the cloud. This patch reduces
amount of UDP traffic in stock scenarios, tests with old values are
extracted to udp_* files.
Partial bug 1470892
Change-Id: I410bb3be0f7e26 7cd4aba4e95a375 9e9ad6bc5c0