[netperf-talk] NetPerf fails in Windows 2008 Server
Vinod Pademar
vinod323 at yahoo.com
Wed Jul 27 06:24:37 PDT 2011
Hi Rick,
Thank you a bunch for all the support that you have been extending.
After obtaining the updated file src/nettest_bsd.c from the trunk, I have built a fresh copy of "netperf.exe" for Windows 7, Win2K3 and Win2K8 R2 (64-bit) platforms.
Also, I have compiled a fresh installation on my Linux server (10.219.192.20) with the new src/nettest_bsd.c file.
The result is not what I am expecting (recv_response: partial response received: 0 bytes).
Atleast happy that no errors this time.
The query result with debug enabled attached for your reference. (same in all windows platforms)
Please let me know, if I need to this differently to get the desired result. Thank you.
Regards
Vinod
C:\>netperf.exe -f m -i 9,3 -t TCP_STREAM -H 10.219.192.20 -l 20 -c -- -s 131072
-S 131072
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.219.192.20 (10.219.192.20) port 0 AF_INET : +/-2.500% @ 99% conf.
recv_response: partial response received: 0 bytes
C:\>netperf.exe -d -f m -i 9,3 -t TCP_STREAM -H 10.219.192.20 -l 20 -c -- -s 131072 -S 131072
resolve_host called with host '10.219.192.20' port '(null)' family AF_UNSPEC
getaddrinfo returned the following for host '10.219.192.20' port '(null)' family AF_UNSPEC
cannonical name: '10.219.192.20'
flags: 4 family: AF_INET: socktype: SOCK_STREAM protocol IPPROTO_TCP addrlen 16
sa_family: AF_INET sadata: 0 0 10 219 192 20 0 0 0 0 0 0 0 0 17 0
scan_sockets_args called with the following argument vector
netperf.exe -d -f m -i 9 -t TCP_STREAM -H 10.219.192.20 -l 20 -c -- -s 131072 -S 131072
Program name: netperf.exe
Local send alignment: 8
Local recv alignment: 8
Remote send alignment: 8
Remote recv alignment: 8
Report local CPU 1
Report remote CPU 0
Verbosity: 1
Debug: 1
Port: 12865
Test name: TCP_STREAM
Test bytes: 0 Test time: 20 Test trans: 0
Host name: 10.219.192.20
remotehost is 10.219.192.20 and port 12865
resolve_host called with host '10.219.192.20' port '12865' family AF_INET
getaddrinfo returned the following for host '10.219.192.20' port '12865' family AF_INET
cannonical name: '10.219.192.20'
flags: 4 family: AF_INET: socktype: SOCK_STREAM protocol IPPROTO_TCP addrlen 16
sa_family: AF_INET sadata: 50 65 10 219 192 20 0 0 0 0 0 0 0 0 3 0
resolve_host called with host '0.0.0.0' port '0' family AF_UNSPEC
getaddrinfo returned the following for host '0.0.0.0' port '0' family AF_UNSPEC
cannonical name: '0.0.0.0'
flags: 4 family: AF_INET: socktype: SOCK_STREAM protocol IPPROTO_TCP addrlen 16
sa_family: AF_INET sadata: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 17 0
establish_control called with host '10.219.192.20' port '12865' remfam AF_INET
local '0.0.0.0' port '0' locfam AF_UNSPEC
bound control socket to 0.0.0.0 and 0
successful connection to remote netserver at 10.219.192.20 and 12865
complete_addrinfo using hostname 10.219.192.20 port 0 family AF_INET type SOCK_STREAM prot IPPROTO_TCP flags 0x0
getaddrinfo returned the following for host '10.219.192.20' port '0' family AF_INET
cannonical name: '10.219.192.20'
flags: 4 family: AF_INET: socktype: SOCK_STREAM protocol IPPROTO_TCP addrlen 16
sa_family: AF_INET sadata: 0 0 10 219 192 20 0 0 0 0 0 0 0 0 4 0
local_data_address not set, using local_host_name of '0.0.0.0'
complete_addrinfo using hostname 0.0.0.0 port 0 family AF_UNSPEC type SOCK_STREAM prot IPPROTO_TCP flags 0x1
getaddrinfo returned the following for host '0.0.0.0' port '0' family AF_UNSPEC
cannonical name: '0.0.0.0'
flags: 4 family: AF_INET: socktype: SOCK_STREAM protocol IPPROTO_TCP addrlen 16
sa_family: AF_INET sadata: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0
TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 10.219.192.20 (10.219.192.20) port 0 AF_INET : +/-2.500% @ 99% conf.
create_data_socket: socket 1912 obtained...
netperf: get_sock_buffer: send socket size determined to be 131072
netperf: get_sock_buffer: receive socket size determined to be 131072
send_tcp_stream: send_socket obtained...
recv_response: received a 256 byte response
remote listen done.
recv_response: received a 256 byte response
remote reporting results for 20.28 seconds
correction factor: 1.000000
calc_service_demand called: units_sent = 45350912.000000
elapsed_time = 0.000000
cpu_util = 0.000000
num cpu = 0
calc_service_demand using: units_sent = 45350912.000000
elapsed_time = 20.281250
cpu_util = 0.462250
num cpu = 2
calc_service_demand got: thruput = 2183.691834
servdem = 4.233652
calculate_confidence: itr 1; time 20.281250; res 17.888804
lcpu 0.462250; rcpu -1.000000
lsdm 4.233653; rsdm -1.000000
create_data_socket: socket 1912 obtained...
netperf: get_sock_buffer: send socket size determined to be 131072
netperf: get_sock_buffer: receive socket size determined to be 131072
send_tcp_stream: send_socket obtained...
recv_response: received a 256 byte response
remote listen done.
recv_response: received a 0 byte response
recv_response: partial response received: 0 bytes
C:\>
--- On Wed, 7/27/11, Rick Jones <rick.jones2 at hp.com> wrote:
From: Rick Jones <rick.jones2 at hp.com>
Subject: Re: [netperf-talk] NetPerf fails in Windows 2008 Server
To: "Vinod Pademar" <vinod323 at yahoo.com>
Cc: "netperf-talk at netperf.org" <netperf-talk at netperf.org>
Date: Wednesday, July 27, 2011, 4:08 AM
So, I managed to find the diffs that someone sent back in March - they tweak some things in src/nettest_bsd.c to deal with race conditions between the test thread and the one emulating SIGALRM. I honestly don't know if they will address the problem you are encountering, but I've gone ahead and checked them into the top of trunk.
Meanwhile, I'm wondering if there might be a slightly cleaner way to deal with the netlib.c compilation problem - perhaps with casts alone and not an additional variable?
rick
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.netperf.org/pipermail/netperf-talk/attachments/20110727/465167ce/attachment.html>
More information about the netperf-talk
mailing list