opensand team mailing list archive

[Andrei POPOV <nyandoma@xxxxxxxxx>]

Hi Julien,

Thanks for your letter!

Here are my answers:

On 25/07/2013 17:08, Julien BERNARD wrote:
> Hi Andrei,
> > While  carrying  out a series of experiments I faced a problem when at
> > some point Satellite Terminal stops working.
> > The  testbed  topology  is  presented in the attached file. I generate
> > multiple  traffic  flows (1 to 9) from  the  Sender  to  the 
> > Receiver  in the
> > direction  from GW to ST using iperf with MSS = 1400 bytes. The buffer
> > of the GW (max_fifo) reduced to the value of 2500.
> > At  some  point  of the experiments (after 40-100 seconds form flow(s)
> > start)  the  ST  stops  processing  traffic (sometimes ST indicator in
> > OpenSand  Manager  turns red, sometimes it's still green but TCP flows
> > don't flow).
>
> I tried to reproduce your problem in your conditions but among many 
> tests I only encounter a crash a couple of times and this is not 
> enough to investigate.
>
> My tests:
>  - I configure GW fifo size to 2500 and  start OpenSAND
>  - I launch 9 or more iperf flows with the command "iperf -c 
> 192.168.19.1 -p 3004 -M 1400 -t 500 -m" in a workstation behind the GW
> Am I missing something ? Do you have a problem for each test or is it 
> random ?
Yes, that's right. I use the same configuration and send traffic from WS 
behind the GW to WS behind the ST. And I don't run opensnad-network and 
opensand-daemon services on both workstations. I found that WS-daemon 
adds route to the network behind ST or GW. I use static routing instead. 
Please let me know if it's essential to run WS-daemon services and if 
they have another functions besides routing.
As I said in previous letters, all machines are virtual machines hosted 
on one single machine. And may be I should add that I use 1.0 version of 
OpenSand. The problem itself is random but in my case it can be 
reproduced on the second-forth experimental try after starting OpenSand. 
In fact, this problem appeared only when I started testing GW-to-ST 
direction. In ST-to-GW direction TCP worked fine with BE buffers in the 
range from 3000 to 30 000. With buffer less than 3000 TCP performance 
decreases and with very low buffers (don't remember exactly - may be 
50-100) iperf refused to establish connection and send traffic.
> > In  the  core.log  I  see  the  following errors after which ST stops
> > working:
> >
> > Jul 24 15:28:40 router st: [encap] [Mpeg.cpp:deencapsulate():193] 
> > ERR: [Mpeg::Context::deencapsulate] cannot create a burst of packets, 
> > drop packet
> > Jul 24 15:28:42 router st: [encap] [Mpeg.cpp:deencapMpeg():515] ERR: 
> > [Mpeg::Context::deencapMpeg] MPEG frame(s) lost (MPEG CC = 12, CTXT 
> > CC = 8), reset context, sync on PUSI
> > Jul 24 15:28:42 router st: [encap] [Mpeg.cpp:deencapMpeg():515] ERR: 
> > [Mpeg::Context::deencapMpeg] MPEG frame(s) lost (MPEG CC = 15, CTXT 
> > CC = 2), reset context, sync on PUSI
> > Jul 24 15:28:43 router st: [dvb_rcs_tal] 
> > [bloc_dvb_rcs_tal.cpp:connectToQoSServer():1138] 
> > [Tal][BlocDVBRcsTal::connectToQoSServer] service on TCP/12000 is not 
> > available
> > Jul 24 15:28:44 router st: [encap] [Mpeg.cpp:deencapMpeg():536] ERR: 
> > [Mpeg::Context::deencapMpeg] PUSI not set in current MPEG frame, drop it
> > .............................
> > Jul 24 15:28:56 router st: [encap] 
> > [../../src/common/IpPacketHandler.h:build():89] ERR: 
> > [IpPacketHandler::build] cannot get IP version
> > Jul 24 15:28:56 router st: [encap] [Ule.cpp:deencapUle():397] ERR: 
> > [Ule::Context::deencapUle] cannot create a IP packet, drop the ULE 
> > packet
> > Jul 24 15:28:56 router st: [encap] [Ule.cpp:deencapsulate():191] ERR: 
> > [Ule::Context::deencapsulate] cannot create a burst of packets, drop 
> > packet
> > Jul 24 15:28:56 router st: [encap] 
> > [../../src/common/IpPacketHandler.h:build():89]
> The errors seems to be in the encapsulation plugins, could you try 
> with GSE on forward link instead of MPEG/ULE ?
> You may have to install libopensand-gse-encap-plugin on hosts and 
> libopensand-gse-encap-plugin-manager on manager.
Ok, thanks for suggestion! I'll try with GSE and tell you the result.

Best regards,
Andrei



> > Could you please tell me what can cause this problem?
>
> I think there is a bug in OpenSAND appearing at high throughput but I 
> need a way to reproduce it easily in order to fix it.
>
> Best regards,