Problem with httplib.HTTPConnection

[Chiara81000]

Hi everyone,
I hope someone can help me! I'm working on some N95 running Python for S60 1.4.1. Untill now the same code has worked. The accused code is the following:

import httplib, urllib
import socket

apid = socket.select_access_point()
apo = socket.access_point(apid)
socket.set_default_access_point(apo)

diz = {'par1' : 1000, 'par2':2000, 'par3':3000}
params = urllib.urlencode(diz)
headers = {"Content-type": "application/x-www-form-urlencoded", "Accept": "text/plain"}
conn = httplib.HTTPConnection("xxx.xxx.xxx.xxx")
conn.request("POST", "/script.php", params, headers) 
response = conn.getresponse()
var = response.read()
conn.close()

The problem is quite strange! Untill yesterday this code doesn't have given me any problem. From yesterday, it sometimes works, the other times I got (at the bold line) one of the following problems:


File "c:\resource\httplib.py", line 627, in connect
return SSLFile(self._shared, self._ssl, bufsize)
error: (56, 'Connection timed out')

File "c:\resource\socket.py", line 58, in socket
return _socketobject(_realsocketcall(family, type, proto, apo), family)
RuntimeError: Couldn't start RConnection

File "c:\resource\httplib.py", line 627, in connect
return SSLFile(self._shared, self._ssl, bufsize)
error: (0, 'Error')


I thought "maybe it was me, Maybe I have changed something without wanting in the code" BUT I tried also with some backup version of the code (that I know for sure that was working) and it's the same!!!
I tried with both WLAN and Packet data connection but nothing!

Thanks a lot
Chiara

[JOM]

it sometimes works, the other times ... problems

Reboot your phone, start python and run your script. Does it work?

If it does, script should be ok. The problem would be caused by the environment: what other apps are running at the same time, are they using network connection, how good network connection you have (signal strenght) etc.

Solution possibilities... Well, go back as far as you can until it works ALL THE TIME. Have you added something which effects timing, like second thread, timers, active objects etc.

Just generic network connection debugging, hope you'll find out a solution.

Cheers,

--jouni

[Chiara81000]

HI JOM!!! Thanks a lot for the fast and complete answer!

Reboot your phone, start python and run your script. Does it work?

Yes, it does! It has given me the answer immediately and I couldn't believe it because I had already tried to reboot the phone and to change Sim.

If it does, script should be ok.

I'm happy of this!

The problem would be caused by the environment: what other apps are running at the same time, are they using network connection, how good network connection you have (signal strenght) etc.

I was thinking the same...but at least the WLAN connection should have been good! I have a bad news... it started to give exactly the same errors.

Do you think that there could be any other problem? Can I ask you to check this post? What do you think about it?
He got my same error "RConnection" ... I'd want to be sure that is only fault of the environment!

http://discussion.forum.nokia.com/fo...ht=RConnection

Thanks a lot again
Chiara

[JOM]

the WLAN connection should have been good! I have a bad news... it started to give exactly the same errors.

First it works ok and then after a while it starts failing == something is getting slowly corrupted. Are you reserving more and more resources, never releasing them? Do you just create new connections and never close them? Python does have automatic memory & resource management, but sometimes it needs a little bit of help.

Can I ask you to check this post? What do you think about it? He got my same error "RConnection" ... I'd want to be sure that is only fault of the environment!

http://discussion.forum.nokia.com/fo...ht=RConnection

Heh, easy to check: miohtama is a guru, listen very carefully whatever he says and follow the instructions to the point

Maybe you could run your code through pylint just in case? I don't think it would find memory / resource leaks, but who knows. I've coded several years in several languages in several operating systems and 99% problem cases have been in my own code, or could be fixed in my code, instead of being "system defects". I've seen some, even hardware defects, but don't really believe in them.

Last night I spend over two hours debugging mysterious crash, which most likely was caused by reading array out-of-bounds. I added an extra buffer at the end and now everything works. The original code was supposed to be ok. I verified indexing a hundred times and then again. No idea what happened, never managed to get a usable traceback..

Very tired,

--jouni