ndt-users list created

[Richard Carlson <>]

Hi Chris;

I pulled the 3.5.6 tarball down and tested it on 2 systems.  It ran properly.

Looking again at the trace Christoph sent I see the following:

690876 kbps outbound
>>> send_msg: type=5, len=6
Signal USR1(10) sent to child [3744]
Signal 10 received by process 3744
Received SIGUSRx signal terminating data collection loop for pid=3744
Sending pkt-pair data back to parent on pipe 8, 9
fwd.saddr = f006c0a:3451, rev.saddr = dd70b0a:3002
01:02:46.369960   10.108.0.15:3451 --> 10.11.215.13:3002 Collected pkt-pair data max = 334203
01:02:46.369960   10.11.215.13:3002 --> 10.108.0.15:3451 Collected pkt-pair data max = 81670
128 bytes read '  1 0 0 0 19 12738 129915 1642 108553 334203 0 6705 219.82 0 0 0 1 0 7' from monitor pipe
128 bytes read '  0 0 1 0 3 31 49754 81670 19693 11227 1 0 581.71 21 1 162358 0 1 7' from monitor pipe
>>> send_msg: type=6, len=0
Signal 11 received by process 3744
Signal 17 received by process 3741
 <------------------------->
 <-- S2C throughput test -->

[ test results and debugging info followed by]

Signal USR2(12) sent to child [3746]
Signal 12 received by process 3746
Received SIGUSRx signal terminating data collection loop for pid=3746
Sending pkt-pair data back to parent on pipe 8, 9
fwd.saddr = dd70b0a:3003, rev.saddr = f006c0a:3461
01:02:56.724367   10.11.215.13:3003 --> 10.108.0.15:3461 Collected pkt-pair data max = 18667
01:02:56.724367   10.108.0.15:3461 --> 10.11.215.13:3003 Collected pkt-pair data max = 65475
Read '  1 0 0 0 4 661 18667 6971 5501 5721 0 5377 976.37 0 0 0 1 0 7' from monitor pipe
Read '  0 0 0 1 367 9334 40681 26321 35413 65475 39990 34285 663.83 39864 40036 171967 0 39990 7' from monitor pipe
550764 kbps inbound

[ followed by the test results being sent to the client and ]

Signal 11 received by process 3746
Signal 17 received by process 3741
Protocol error!
>>> send_msg: type=7, len=61
S2C throughput test FAILED!

When I run tests on my servers I don't get the "Signal 11 receive by process xxxx" message.  Instead I see the lines 

pcap_loop exited recvfrom: Bad file descriptor

Pkt-Pair data collection ended, waiting for signal to terminate process

Signal 11 is a SIGSEGV (invalid memory reference) and I did make changes to the pcap routines to automatically detect which interface the server is using.  

According to yum I have 0.9.8-3.fc10_x86_64 on my development system and 0.9.4 on my production server.    

Given the code runs on my servers, I need to ask what version of the pcap lib development package are you using?

Rich

On Sep 1, 2009, at 7:39 AM, Chris Welti wrote:

Hi Rich,

I've just upgraded one of our NDT servers from 3.4.4 to 3.5.6.
I'm observing the same behaviour as listed in this thread:

** Starting test 1 of 1 **
Connected to: lsmp2  --  Using IPv4 address
Checking for Middleboxes . . . . . . . . . . . . . . . . . .  Done
checking for firewalls . . . . . . . . . . . . . . . . . . .  Done
running 10s outbound test (client-to-server [C2S]) . . . . . 91.78Mb/s
running 10s inbound test (server-to-client [S2C]) . . . . . . 85.01Mb/s
S2C throughput test: Received wrong type of the message
ERROR MSG: Server (S2C throughput test): Invalid S2C throughput received
S2C throughput test FAILED!
Server unable to determine bottleneck link type.
Information: Other network traffic is congesting the link

There are test results for both C2S and S2C tests, but an error is reported anyway.
On the server there are also error messages (Protocol error! S2C throughput test FAILED!)

Anyway, I've tried to install all the 3.5. versions before and it seems to me that this bug was introduced with v3.5.3.
Using v3.5.2 or v3.5.1 it works as a charm on the same system (debian, lenny, 2.6.26-web100)
v3.5.3, v3.5.4, v3.5.5 and v3.5.6 all produce the same error output as above.

Maybe that helps... Regards,
Chris

Galuschka Christoph wrote:

Hello Richard,

I'm currently running the tests with 2.6.30.5 and IE6. JAVA - if

relevant - is 1.6.11.

The funny thing also is, I do get results in the webbrowser, the client

just doesn't finish correctly.

I will repair the WAIT-comment thing on monday (sorry, I'm not the best

in C)

Regards

Christoph

-----------------------------------------

Ing. Christoph Galuschka

TIWAG-Tiroler Wasserkraft AG

Bereich IT/Betrieb und Services

Eduard-Wallnöfer-Platz 2

6010 Innsbruck

T: +43 (0)50607 21832

F: +43 (0)50607 41832

www.tiroler-wasserkraft.at

-----------------------------------------

Firmenbuchgericht Innsbruck, FN 44133b

Sitz der Gesellschaft: Innsbruck

DVR: 0164089

Richard Carlson

<>

1000 Oakbrook Dr

Ann Arbor, MI  48104

P: 734-352-7043

C: 630-251-4572

Richard Carlson

1000 Oakbrook Dr

Ann Arbor, MI  48104

P: 734-352-7043

C: 630-251-4572