473,397 Members | 2,068 Online
Bytes | Software Development & Data Engineering Community
Post Job

Home Posts Topics Members FAQ

Join Bytes to post your question to a community of 473,397 software developers and data experts.

xmlrpclib hangs execution

Perhaps I should be a bit more specific. When using this code to connect
to a remote XML-RPC server (C++, xmlrpc++0.7 library):

import xmlrpclib
server = xmlrpclib.Server("http://10.10.101.62:29500")
print server.Connection_Request("roberto")

the Python command line 'hangs' until I kill the server. Then, the
correct output is suddenly displayed:

{'state': 0, 'id_session': '2Z3EUSLJFA13', 'port': 29501,
'str': 'Connection accepted. Session attached.'}

Yet a slightly simpler call works flawlessly:

import xmlrpclib
server = xmlrpclib.Server("http://10.10.101.62:29500")
print server.Info_RT('Master')

{'state': 0, 'str': 'Info_RT'}

Looking at the TCP packages (using Wireshark), the first
Connection_Request flow as follows (c=python client,s= xml-rpcserver):

1 c>s SYN
2 s>c SYN,ACK
3 c>s ACK
4 c>s PSH,ACK (setting content-type: text/xml)
5 s>c ACK
6 c>s PSH,ACK (xml-rpc request)
7 s>c ACK
8 s>c PSH,ACK (xml-rpc results, correct)
9 c>s ACK

At this point, Python 'hangs' until I kill (-9) the server (28 seconds
later), and then these last packages are send:

10 s>c FIN,ACK
11 c>s FIN,ACk
12 s>c ACK

After which Python continues and prints the xml-rpc results on the
screen. Strangely, before Python hangs, it has already received the
package with the results, but doesn't do anything with it.

For comparison, this is the package flow for the Info_RT method, which
returns within a second with the correct result on the Python command line:

1 c>s SYN
2 s>c SYN,ACK
3 c>s ACK
4 c>s PSH,ACK (setting content-type: text/xml)
5 s>c ACK
6 c>s PSH,ACK (xml-rpc request)
7 s>c ACK
8 s>c PSH,ACK (xml-rpc results, correct)
9 s>c FIN,ACK
10 c>s ACK
11 c>s FIN,ACk
12 s>c ACK

Notice the reverse order of packages 9 and 10. Could this be it? Is
Python waiting until the server sends a FIN package? But if so, why?

Looking at the Python trace, this might be it. A diff between the trace
of the killed Connection_Request and the Info_RT resulted only in
expected differences (due to different length of parameters). But when
diffing the traces of the killed and non-killed Connection_Request,
something else is visible. The killed Connection_Request ends with:

[...]
string:318: return sep.join(words)
xmlrpclib:866: self._type = "params"
xmlrpclib:769: try:
xmlrpclib:770: f = self.dispatch[tag]
xmlrpclib:771: except KeyError:
xmlrpclib:772: pass # unknown tag ?
xmlrpclib:1273: if sock:
xmlrpclib:1274: response = sock.recv(1024)
xmlrpclib:1277: if not response:
xmlrpclib:1278: break
xmlrpclib:1283: file.close()
socket:218: try:
socket:219: if self._sock:
socket:220: self.flush()
socket:232: if self._wbuf:
socket:222: self._sock = None
xmlrpclib:1284: p.close()
xmlrpclib:530: self._parser.Parse("", 1) # end of data
xmlrpclib:531: del self._target, self._parser # get rid of
circular references
xmlrpclib:1286: return u.close()
xmlrpclib:741: if self._type is None or self._marks:
xmlrpclib:743: if self._type == "fault":
xmlrpclib:745: return tuple(self._stack)
socket:225: try:
socket:226: self.close()
socket:218: try:
socket:219: if self._sock:
socket:222: self._sock = None
xmlrpclib:1386: if len(response) == 1:
xmlrpclib:1387: response = response[0]
xmlrpclib:1389: return response
__main__:26: print result
{'state': 0, 'id_session': 'QDLY3GP9FBM5', 'port': 29501, 'str':
'Connection accepted. Session attached.'}

But the non-killed Connection_Request looks like:

[...]
return sep.join(words)
xmlrpclib:866: self._type = "params"
xmlrpclib:769: try:
xmlrpclib:770: f = self.dispatch[tag]
xmlrpclib:771: except KeyError:
xmlrpclib:772: pass # unknown tag ?
xmlrpclib:1273: if sock:
xmlrpclib:1274: response = sock.recv(1024)
socket:225: try:
socket:226: self.close()
socket:218: try:
socket:219: if self._sock:
socket:220: self.flush()
socket:232: if self._wbuf:
socket:222: self._sock = None

(In the diff there is a gap between the xmlrpclib:1274 and socket:225
lines, as compared to the killed Connection_Request; perhaps Ctrl+C
still executes the "socket try close" lines?)

If you want to do your own 'diff' on the Python traces, or look at the
TCP packages in more details with Wireshark, here are the files for the
correctly working Info_RT and faulty Connection_Request methods:

http://ctw-bw008.ctw.utwente.nl/~arn...fort-python.ws
http://ctw-bw008.ctw.utwente.nl/~arn...-python-ws.txt
http://ctw-bw008.ctw.utwente.nl/~arn...fort-trace.txt

http://ctw-bw008.ctw.utwente.nl/~arn...nreq-python.ws
http://ctw-bw008.ctw.utwente.nl/~arn...-python-ws.txt
http://ctw-bw008.ctw.utwente.nl/~arn...nreq-trace.txt
http://ctw-bw008.ctw.utwente.nl/~arn...-hardbreak.txt

(The first three ConReq files all include killing the server after ~28
seconds, thus resulting in the correct output. In the fourth files,
tracing is stopped before the server is killed. It is therefor the trace
up to the 'hanging point'. Note that the first two Wireshark packages
are not TCP, but ARP, thus the file has a total of 14 packages, not 12
as above.)

Really, any help here is appreciated!

Kind regards,
Arno.

Arno Stienen wrote:
Hello all,

It has been a while, but I finally found the time to further investigate
my problems with connecting Python to a RTAI-XML server. As I cannot
tell if the problems I'm facing are caused by RTAI-XML (or more
precisely, the xmlrpc++0.7 library) or Python (xmlrpclib), I'm posting
this message on both the RTAI as the Python mailing list.

Just to recap, here are the symptoms I described earlier.

I am using two computers. One is running the RTAI-patched real-time
linux kernel, and this computer controls my robotic hardware. On this
RTAI computer I installed the RTAI-XML server, which uses incoming
XML-RPC calls to control real-time processes on the server. RTAI-XML is
programmed in C++, using the xmlrpc++0.7 library.

The second is a desktop machine (Windows XP or Ubuntu), from which a
XML-RPC connection can be made to the RTAI computer. Although in theory
it should make no difference if I use C, C++, Java or Python to make the
connection, in practise I can get the Java connection running, but have
problems using Python. And I really like to use Python to quickly create
interfaces!

When connecting with Python on the desktop computer to the RTAI-XML
server running on my RTAI machine, I used the following code.

import xmlrpclib
server = xmlrpclib.Server("http://10.10.101.62:29500")
print server.Info_RT('Master')

This results in the correct output:

{'state': 0, 'str': 'Info_RT'}

Yet, this is only an information call. The really interesting part
should comes when I ask the server to open a connection:

import xmlrpclib
server = xmlrpclib.Server("http://10.10.101.62:29500")
print server.Connection_Request("roberto")

But then nothing happens. Python just hangs and is unresponsive. Until,
I kill the RTAI-XML server on the RTAI computer! And then suddenly the
correct output appears on the desktop command line:

{'state': 0, 'id_session': '2Z3EUSLJFA13', 'port': 29501,
'str': 'Connection accepted. Session attached.'}

(Using an alternative Python XML-RPC library '#import pyxmlrpclib as
xmlrpclib' makes no difference to the result.)

So what happens? I did some experiments. I did some package sniffing
with Wireshark and I did Python tracing. I stored the sniffing results
in both the Wireshark as plain text format. The files for the first
Info_RT examples are here:

http://ctw-bw008.ctw.utwente.nl/~arn...fort-python.ws
http://ctw-bw008.ctw.utwente.nl/~arn...-python-ws.txt
http://ctw-bw008.ctw.utwente.nl/~arn...fort-trace.txt

The files for the second, not correctly working example are here:

http://ctw-bw008.ctw.utwente.nl/~arn...nreq-python.ws
http://ctw-bw008.ctw.utwente.nl/~arn...-python-ws.txt
http://ctw-bw008.ctw.utwente.nl/~arn...nreq-trace.txt
http://ctw-bw008.ctw.utwente.nl/~arn...-hardbreak.txt

The first three files all include killing the server after ~30 seconds,
thus resulting in the correct output. In the fourth files, tracing is
stopped before the server is killed. It is therefor the trace up to the
'hanging point'.

Any, really any, suggestions are appreciated!

Kind regards,
Arno Stienen.
May 28 '07 #1
0 1618

This thread has been closed and replies have been disabled. Please start a new discussion.

Similar topics

0
by: Larry | last post by:
I've had a production system running for a long time that uses xmlrpclib with timeoutsocket, and with my recent upgrade to 2.3 it's no longer able to use xmlrpclib with the xmlrpc servers I...
0
by: sashan | last post by:
Hi I'm having trouble using xmlrpclib. I register a function (or class) with the SimpleXMLRPCServer and initiate the server. I then create a ServerProxy object and connect to the xml-rpc...
2
by: p2esp | last post by:
Hello, I'm using the xmlrpclib module to contact an XMLRPC server that takes a long time to send results back. My client timeouts. The question is whether there is a way to have an xmlrpclib...
1
by: Gabriel PASTOR | last post by:
I'm trying to send object using xmlrpclib, but it seems that classes inheriting from object cannot be marshalled. Here is an example: -------- server.py -------- import xmlrpclib,...
3
by: Rune Froysa | last post by:
Trying something like:: import xmlrpclib svr = xmlrpclib.Server("http://127.0.0.1:8000") svr.test("\x1btest") Failes on the server with:: xml.parsers.expat.ExpatError: not well-formed (invalid...
0
by: Willi Langenberger | last post by:
Hi! We have an application server (Zope) and make heavy use of xml-rpc. One problem arised, when we tried to return a zope.DateTime instance. xmlrpclib (naturally) knows nothing about...
5
by: Jonathan Ballet | last post by:
Hello, I have developped a XMLRPC server, which runs under Gnu/Linux with python2.3. This server receives method calls from Windows client. The server got some parameters which are string,...
2
by: squid | last post by:
First off, I'm a python neophyte, but I'm fairly experienced with Java, C and PHP. I've been trying to use the xmlrpclib to perform remote calls against a service, and it works nicely. However,...
2
by: Pierre Rouleau | last post by:
Hi all, I have a consistent test case where os.popen3() hangs in Windows. The system hangs when retrieving the lines from the child process stdout. I know there were several reports related to...
0
BarryA
by: BarryA | last post by:
What are the essential steps and strategies outlined in the Data Structures and Algorithms (DSA) roadmap for aspiring data scientists? How can individuals effectively utilize this roadmap to progress...
1
by: nemocccc | last post by:
hello, everyone, I want to develop a software for my android phone for daily needs, any suggestions?
1
by: Sonnysonu | last post by:
This is the data of csv file 1 2 3 1 2 3 1 2 3 1 2 3 2 3 2 3 3 the lengths should be different i have to store the data by column-wise with in the specific length. suppose the i have to...
0
by: Hystou | last post by:
There are some requirements for setting up RAID: 1. The motherboard and BIOS support RAID configuration. 2. The motherboard has 2 or more available SATA protocol SSD/HDD slots (including MSATA, M.2...
0
marktang
by: marktang | last post by:
ONU (Optical Network Unit) is one of the key components for providing high-speed Internet services. Its primary function is to act as an endpoint device located at the user's premises. However,...
0
by: Hystou | last post by:
Most computers default to English, but sometimes we require a different language, especially when relocating. Forgot to request a specific language before your computer shipped? No problem! You can...
0
Oralloy
by: Oralloy | last post by:
Hello folks, I am unable to find appropriate documentation on the type promotion of bit-fields when using the generalised comparison operator "<=>". The problem is that using the GNU compilers,...
0
agi2029
by: agi2029 | last post by:
Let's talk about the concept of autonomous AI software engineers and no-code agents. These AIs are designed to manage the entire lifecycle of a software development project—planning, coding, testing,...
0
isladogs
by: isladogs | last post by:
The next Access Europe User Group meeting will be on Wednesday 1 May 2024 starting at 18:00 UK time (6PM UTC+1) and finishing by 19:30 (7.30PM). In this session, we are pleased to welcome a new...

By using Bytes.com and it's services, you agree to our Privacy Policy and Terms of Use.

To disable or enable advertisements and analytics tracking please visit the manage ads & tracking page.