Skip to content
This repository was archived by the owner on Feb 28, 2025. It is now read-only.
This repository was archived by the owner on Feb 28, 2025. It is now read-only.

benchmark teardowns threads for quite a long time #105

Open

Description

@timurb

I use a simple test with 3 consecutive gets and it runs fine when I run it as fl-run-test but it hangs when I try to run it as fl-run-bench:

$  fl-run-bench test_geogrep.py MyTest.test_creds                                                                    * master 139aeb0
No handlers could be found for logger "FunkLoad"
========================================================================
Benching MyTest.test_creds
========================================================================
No test description
------------------------------------------------------------------------

Configuration
=============

* Current time: 2013-11-09T14:09:59.040055
* Configuration file: /home/timur/funkload/MyTest.conf
* Log xml: /home/timur/git/funkload/simple-bench.xml
* Server: http://int.myserver.net
* Cycles: [50, 75, 100, 125]
* Cycle duration: 10s
* Sleeptime between request: from 0.0s to 0.5s
* Sleeptime between test case: 0.01s
* Startup delay between thread: 0.01s

Benching
========

* setUpBench hook: ... done.

Cycle #0 with 50 virtual users
------------------------------

* setUpCycle hook: ... done.
* Current time: 2013-11-09T14:10:04.650614
* Starting threads: .................................................. done.
* Logging for 10s (until 2013-11-09T14:10:15.297702): ................................................................................................................... done.
* Waiting end of threads: FFFFF............................................F...... done.
* Waiting cycle sleeptime 1s: ... done.
* tearDownCycle hook: ... done.
* End of cycle, 137.54s elapsed.
* Cycle result: **SUCCESSFUL**, 115 success, 0 failure, 0 errors.

Cycle #1 with 75 virtual users
------------------------------

* setUpCycle hook: ... done.
* Current time: 2013-11-09T14:12:22.190175
* Starting threads: ........................................................................... done.
* Logging for 10s (until 2013-11-09T14:12:33.168238): ......................................................................................................... done.
* Waiting end of threads: ......................

You see, the whole test have taken probably a dozen of seconds (I've checked that by tailing the XML report) while waiting for the end of threads have taken several minutes. At first I even thought it was hanging.

Is it ok that it takes so long? Can this be somehow improved?

I use funkload 1.16.1-4 from Ubuntu repository but can test from HEAD if needed.

Activity

  1. nomeata commented on Apr 7, 2014

    @nomeata

    Is there any chance that you have long-running request? I am under the impression (but purely from observing funkload here; I am not a developer) that every thread finishes its current request (or even test) and then checks if it should stop itself.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions