ftplib strange behaviour

[siggy2]

Hi,
I'm using
Python 2.3.4 (#53, May 25 2004, 21:17:02) [MSC v.1200 32 bit (Intel)]
on win32
I've noticed a strange (= not deterministic) behaviour of ftplib.py:
sometimes (not always) it fails (after a variable number of minutes
from 15 to 130) downloading a 150 MB BINARY file
(a big gzipped ascii file) with the traceback reported
below.
IMVHO this is not a timeout error because my script import
timeoutsocket.py (__version__ = "$Revision: 1.1.22.1 $"
__author__ = "Timothy O'Malley <[email protected]>")
to explicitly trap that...
and whenever a timeout occurs it is correctly reported and handled
(timeout set to 240 seconds)


Traceback (most recent call last):
File "C:\mydir\myscript.py", line 77, in downloadFile
result = ftpObject.retrbinary('RETR '+name, f.write)
File "C:\Python23\lib\ftplib.py", line 386, in retrbinary
return self.voidresp()
File "C:\Python23\lib\ftplib.py", line 221, in voidresp
resp = self.getresp()
File "C:\Python23\lib\ftplib.py", line 207, in getresp
resp = self.getmultiline()
File "C:\Python23\lib\ftplib.py", line 193, in getmultiline
line = self.getline()
File "C:\Python23\lib\ftplib.py", line 183, in getline
if not line: raise EOFError


In ftplib.py I read this comment:
[CUT]
# Internal: return one line from the server, stripping CRLF.
# Raise EOFError if the connection is closed
def getline(self):
[CUT]

May anyone explain me what is this?
Could this error be explained only as a ftp server problem?
TIA!
bye,
PiErre

 

[Kartic]

PiErre,

I have used ftplib but never for such a huge file (assuming your
problem is related to the size of the file).

Have you tried downloading the file using another ftp client? Does that
download succeed? The reason I ask is because I have attempted
downloads from servers that terminate the connection after a certain
connection time, in the middle of a download!

That is the best I can help you out with the information.
Have a happy New Year.

Thanks,
--Kartic

 

[siggy2]

I have used ftplib but never for such a huge file (assuming your
problem is related to the size of the file).

we're doing that since python 2.2 from different site to different
hosts where our script was installed...
needless to say all the (few) problems we had came from the servers...

Have you tried downloading the file using another ftp client? Does that
download succeed? The reason I ask is because I have attempted
downloads from servers that terminate the connection after a certain
connection time, in the middle of a download!

After two weeks the sysadms of the ftp server
discovered a misconficuration in some router... Anyway
we had already change the script: we managed to trap any exception and

we added a "retry" loop.

That is the best I can help you out with the information.

Thank you for your help anyway.

bye,
PiErre