[Red5] Re d5 dropping requests when there are more than 10 requests

[Mondain]

The info you have is a good start, but providing a test to along with it and
creation of a ticket/issue in our Jira system should be your next step.
Paul

On Fri, Apr 10, 2009 at 12:53 AM, vinady <nav1982 at gmail.com> wrote:

>
> Hi anyone there ? Please reply to my message
>
> vinady wrote:
> >
> > I am running Red5 server version 0.8 in a DELL machine which
> configuration
> > is
> > OS : Ubuntu 8.04 LTS Desktop Edition
> >
> > Processor : Core 2 Duo  3.0 GHZ, 6MB L2 Cache 1333 MHZ FSB
> > RAM - 4 GB
> > Hard Drive - 260 GB
> >
> > And I am testing it with jmeter(Red5Bench) which is running in a Dell
> > Inspiron Laptop having configuration
> > The Jmeter is running in a DELL Inspiron Laptop having
> > OS : Windows Vista
> > Processor : Core 2 Duo 2.0 GHZ
> > RAM : 2GB
> > Hard Drive : 160 GB
> >
> > The test is carried out for 3 different sized videos 3.20 MB(1 min), 17.5
> > MB(3 min 51 seconds), 27.4 MB(5 min 56 seconds)  independently for 5, 10
> > ,20 and 30 users. Here is the result upto 20 users.
> >
> > For 5 users
> >
> > file size     No of Users Disconnected        Total data transferred
>  Time
> > Consumed to complete all requests  Avg Mbit/sec
> > 3.20 MB             0                                            16.20 MB
> > ~2 mins 30 seconds                                     0.746
> > 17.5 MB             0                                            92.96 MB
> > ~14 minutes                                                 0.873
> > 27.4 MB             0                                            145.13
> MB
> > ~17 minutes                                                 1.143
> >
> > For 10 users
> >
> > file size     No of Users Disconnected        Total data transferred
>  Time
> > Consumed to complete all requests  Avg Mbit/sec
> > 3.20 MB             0                                            33.84 MB
> > -                                                                 2.068
> > 17.5 MB             1                                           176.78 MB
> > ~ 22 minutes                                                1.093
> > 27.4 MB             0                                            288.0 MB
> > ~ 12 minutes                                                3.245
> >
> > For 20 users
> >
> > file size     No of Users Disconnected        Total data transferred
>  Time
> > Consumed to complete all requests  Avg Mbit/sec
> > 3.20 MB             10                                            34.05
> MB
> > ~9 minutes                                                 0.524
> > 17.5 MB             5                                              278.40
> > MB                 ~ 11 minutes
> > 3.428
> > 27.4 MB             5                                              288.0
> > MB                  ~ 28 minutes
> > 1.967
> >
> > From the data it shows that its taking a lot of time to complete the
> > requests(which is shown by the Time Consumed to complete all requests
> > field in the above tables) compared to the video duration.
> >
> > If there are 10 or more requests for red5 server at a time it gives
> > following errors and disconnects some of the threads
> >
> > [INFO] [pool-4-thread-16]
> > org.red5.server.adapter.MultiThreadedApplicationAdapter - W3C
> > x-category:session x-event:connect c-ip:192.168.1.3 c-client-id:4
> > [INFO] [pool-4-thread-16]
> > org.red5.server.adapter.MultiThreadedApplicationAdapter - W3C
> > x-category:session x-event:connect c-ip:192.168.1.3 c-client-id:5
> > [WARN] [DefaultQuartzScheduler_Worker-7]
> > org.red5.server.net.rtmp.RTMPConnection - Closing RTMPMinaConnection from
> > 192.168.1.3 : 52936 to null (in: 3073 out 3073 ), with id 29087107 due to
> > long handshake
> > [INFO] [pool-4-thread-16]
> > org.red5.server.adapter.MultiThreadedApplicationAdapter - W3C
> > x-category:session x-event:connect c-ip:192.168.1.3 c-client-id:6
> > [WARN] [DefaultQuartzScheduler_Worker-9]
> > org.red5.server.net.rtmp.RTMPConnection - Closing RTMPMinaConnection from
> > 192.168.1.3 : 52938 to null (in: 3073 out 3073 ), with id 6040101 due to
> > long handshake
> > [WARN] [DefaultQuartzScheduler_Worker-10]
> > org.red5.server.net.rtmp.RTMPConnection - Closing RTMPMinaConnection from
> > 192.168.1.3 : 52940 to null (in: 3073 out 3073 ), with id 9243407 due to
> > long handshake
> > [WARN] [DefaultQuartzScheduler_Worker-1]
> > org.red5.server.net.rtmp.RTMPConnection - Closing RTMPMinaConnection from
> > 192.168.1.3 : 52939 to null (in: 3073 out 3073 ), with id 5757973 due to
> > long handshake
> > [WARN] [DefaultQuartzScheduler_Worker-2]
> > org.red5.server.net.rtmp.RTMPConnection - Closing RTMPMinaConnection from
> > 192.168.1.3 : 52941 to null (in: 3073 out 3073 ), with id 21018877 due to
> > long handshake
> > [WARN] [DefaultQuartzScheduler_Worker-3]
> > org.red5.server.net.rtmp.RTMPConnection - Closing RTMPMinaConnection from
> > 192.168.1.3 : 52943 to null (in: 3073 out 3073 ), with id 16253147 due to
> > long handshake
> >
>
> --
> View this message in context:
> http://www.nabble.com/Red5-dropping-requests-when-there-are-more-than-10-requests-tp22906092p22984540.html
> Sent from the Red5 - English mailing list archive at Nabble.com.
>
>
> _______________________________________________
> Red5 mailing list
> Red5 at osflash.org
> http://osflash.org/mailman/listinfo/red5_osflash.org
>



-- 
http://gregoire.org/
http://osflash.org/red5
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://osflash.org/pipermail/red5_osflash.org/attachments/20090410/a101b88d/attachment.html>