Path: csiph.com!fu-berlin.de!uni-berlin.de!not-for-mail
From: Antoon Pardon <antoon.pardon@rece.vub.ac.be>
Newsgroups: comp.lang.python
Subject: Re: ftplib throws: IndexError: tuple index out of range
Date: Sun, 14 Feb 2016 19:10:16 +0100
Lines: 50
Message-ID: <mailman.114.1455473488.22075.python-list@python.org>
References: <56C05CEF.5080304@rece.vub.ac.be> <n9q04l$ese$1@ger.gmane.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 7bit
X-Trace: news.uni-berlin.de A08WUYTmqlEGDCnr2Daj2gWw4vKzPv8j8Gi7DCE0RNGA==
Return-Path: <antoon.pardon@rece.vub.ac.be>
X-Original-To: python-list@python.org
Delivered-To: python-list@mail.python.org
X-Spam-Status: OK 0.003
X-Spam-Evidence: '*H*': 0.99; '*S*': 0.00; 'args.': 0.09; 'e.args[0]': 0.09; 'ftplib': 0.09; 'timeout': 0.09; 'tuple': 0.09; 'exception': 0.13; 'index': 0.13; '478': 0.16; 'eintr:': 0.16; 'fires': 0.16; 'frame,': 0.16; 'indexerror:': 0.16; 'innermost': 0.16; 'received :adsl-dyn.isp.belgacom.be': 0.16; 'received:io': 0.16; 'received:psf.io': 0.16; 'socket.py': 0.16; 'wrote:': 0.16; 'mechanism': 0.18; 'try:': 0.18; 'skip:" 30': 0.20; 'defined': 0.23; 'seems': 0.23; 'somewhere': 0.24; 'written': 0.24; 'header :In-Reply-To:1': 0.24; 'header:User-Agent:1': 0.26; 'error': 0.27; 'actual': 0.28; 'looks': 0.29; 'arguments,': 0.29; 'readline': 0.29; 'received:192.168.1.3': 0.29; 'raise': 0.29; 'starts': 0.29; 'program,': 0.29; 'received:be': 0.30; 'probably': 0.31; 'skip:s 30': 0.31; 'anyone': 0.32; 'class': 0.33; 'problem': 0.33; 'raised': 0.33; 'skip:/ 20': 0.33; 'file': 0.34; 'except': 0.34; 'problem.': 0.35; 'remote': 0.35; 'to:addr:python-list': 0.36; 'subject:: ': 0.37; 'thanks': 0.37; 'self': 0.38; 'data': 0.39; 'received:192': 0.39; 'to:addr:python.org': 0.40; 'some': 0.40; 'charset:windows-1252': 0.62; 'backup': 0.66; 'locals': 0.84; 'pardon': 0.84; 'received:195.238': 0.84; 'schreef': 0.84
X-Belgacom-Dynamic: yes
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: A2CWAQDUwcBW/1KZhG0NUYQMbbokgWcZhXQCgVcSAQEBAQEBAYVMAQEEeBELGAkWDwkDAgECAUUTCAK0cop6hAoBAQgCAR2GEYQ1g3mBBYNuBZZ5gTuEFIgGgiaMTo4+JgGEH2mIVgEBAQ
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Icedove/38.5.0
In-Reply-To: <n9q04l$ese$1@ger.gmane.org>
X-BeenThere: python-list@python.org
X-Mailman-Version: 2.1.21rc2
Precedence: list
List-Id: General discussion list for the Python programming language <python-list.python.org>
List-Unsubscribe: <https://mail.python.org/mailman/options/python-list>, <mailto:python-list-request@python.org?subject=unsubscribe>
List-Archive: <http://mail.python.org/pipermail/python-list/>
List-Post: <mailto:python-list@python.org>
List-Help: <mailto:python-list-request@python.org?subject=help>
List-Subscribe: <https://mail.python.org/mailman/listinfo/python-list>, <mailto:python-list-request@python.org?subject=subscribe>
Xref: csiph.com comp.lang.python:102919

Op 14-02-16 om 14:40 schreef Peter Otten:
> Antoon Pardon wrote:
> 
>> I have written a small backup program, that uses ftplib to make
>> remote backups. However recentely the program starts to regularly
>> raise IndexErrors, as far as I can see the problem is in socket.py
>> Can anyone shed some light?
>>
>> This is the traceback:
> 
> [...]
> 
>>   File "/usr/lib/python2.7/socket.py", line 478, in readline
>>     if e.args[0] == EINTR:
>> IndexError: tuple index out of range
> 
> The offending line seems to be part of
> 
>                 try:
>                     data = self._sock.recv(self._rbufsize)
>                 except error, e:
>                     if e.args[0] == EINTR:
>                         continue
>                     raise
> 
> 
>> Locals by frame, innermost last
> 
> [...]
> 
>> Frame readline in /usr/lib/python2.7/socket.py at line 478
>>                  buf = <cStringIO.StringO object at 0x7ff8c5d40298>
>>              buf_len = 0
>>                    e = timeout()
>>                 self = <socket._fileobject object at 0x7ff8c7b75b50>
>>                 size = 8193
> 
> It looks like the actual error is socket.timeout which is probably raised 
> from somewhere inside the stdlib without args.

I think I know what is going on. I have my own timeout mechanism at work
here, that works with signals and alarm. When the SIGALRM fires I just
raise the socket.timeout without arguments, which then causes the problem.

I defined my own timeout exception class and now raise that and the problem
disappeared. Thanks for the extra pair of eyes.

-- 
Antoon Pardon.