Too many read 0, while fetching msg '214'

[silasdb]

Hi there! In another list.remote(x) bug report (#198), you told me that the problem I reported was due to poor error handling and wasn't related to the other problem handled there. You then asked me to open a new issue. Here we go. Debug output is attached. Thanks!

debug.txt

[nicolas33]

Which version?

[silasdb]

6.5.6, under NetBSD 7.0.

[nicolas33]

Please, try latest stable.

[silasdb]

In latest stable (6.6.1) list.remove(x): x not in list disappears but Too many read 0 still shows up. (debug2.txt attached).

I had a filesystem crash and if I delete the entire folder and download everything again it works, but I believe OfflineIMAP was supposed to handle this or provide more useful error messages?

Thanks!

debug2.txt

[nicolas33]

ERROR: command: UID => socket error: <type 'exceptions.IOError'> - Too many read 0, while fetching msg '214' in folder 'INBOX.datacenter'. Max retry reached (2)

This Max retry reached (2) explains why it failed. This error should not be reproducible unless there is a lot of changes in this folder.

[silasdb]

Hi again! Thanks! I found and deleted problematic messages using the webmail. Isn't strange, though, that they "appeared" there? Is there any chance that offlineimap have pushed zero size or corrupted messages? Thanks!

[nicolas33]

Yes, that's possible. We can't tell without samples to retry.

[dolohow]

Exactly the same same error happens on some mails when using Python 3, yet offlineimap under Python 2 works like a charm.

ERROR: command: UID => socket error: <class 'OSError'> - Too many read 0, while fetching msg '4105' in folder 'INBOX' . Max retry reached (2)

Looks like a socket cannot be polled in bundled_imaplib2.py:1842 and it retries infinitely, but what causes that I have no idea.

while not (terminate or self.Terminate):                                                                              
    if self.state == LOGOUT:                                                                                          
        timeout = 10                                                                                                  
    else:                                                                                                             
        timeout = read_poll_timeout                                                                                   
    try:                                                                                                              
        r = poll.poll(timeout)                                                                                        
        if __debug__: self._log(5, 'poll => %s' % repr(r))                                                            
        if not r:                                                                                                     
            continue                                    # Timeout

EDIT: Could be some encoding issue?

[nicolas33]

I wonder that with Py3 the reader makes excessive reads while the full email got already downloaded if the recorded size of what was downloaded is wrongly too low. IOW, the email is already fully downloaded but the client didn't count the bytes correctly so it tries to read more data.

This is just an idea that could really be wrong.

[nicolas33]

Additional note: this could happen because str type was changed in Py3. I expect the correct type must be bytes.

[purpleidea]

I hit this issue on a F28 machine. It ran for a while, and then all of a sudden it now dies after a few minutes.

offlineimap --version
7.1.5

Thread 'Copy message from XXX@gmail.com-Remote:[Gmail]/All Mail' terminated with exception:
Traceback (most recent call last):
  File "/usr/lib/python2.7/site-packages/offlineimap/threadutil.py", line 160, in run
    Thread.run(self)
  File "/usr/lib64/python2.7/threading.py", line 757, in run
    self.__target(*self.__args, **self.__kwargs)
  File "/usr/lib/python2.7/site-packages/offlineimap/folder/Gmail.py", line 287, in copymessageto
    super(GmailFolder, self).copymessageto(uid, dstfolder, statusfolder, register)
  File "/usr/lib/python2.7/site-packages/offlineimap/folder/Base.py", line 839, in copymessageto
    message = self.getmessage(uid)
  File "/usr/lib/python2.7/site-packages/offlineimap/folder/Gmail.py", line 72, in getmessage
    data = self._fetch_from_imap(str(uid), self.retrycount)
  File "/usr/lib/python2.7/site-packages/offlineimap/folder/IMAP.py", line 784, in _fetch_from_imap
    imapobj.select(self.getfullIMAPname(), readonly=True)
  File "/usr/lib/python2.7/site-packages/offlineimap/imaplibutil.py", line 64, in select
    exc_info()[2])
  File "/usr/lib/python2.7/site-packages/offlineimap/imaplibutil.py", line 53, in select
    result = super(UsefulIMAPMixIn, self).select(mailbox, readonly)
  File "/usr/lib/python2.7/site-packages/offlineimap/bundled_imaplib2.py", line 1098, in select
    typ, dat = rqb.get_response('command: %s => %%s' % rqb.name)
  File "/usr/lib/python2.7/site-packages/offlineimap/bundled_imaplib2.py", line 201, in get_response
    raise typ(exc_fmt % str(val))
OfflineImapError: Server 'imap.gmail.com' closed connection, error on SELECT '[Gmail]/All Mail'. Server said: command: EXAMINE => socket error: <type 'exceptions.IOError'> - Too many read 0

[nicolas33]

@purcell open a new issue, please. You're running on python2.

[purcell]

@nicolas33 I think you meant to tag @purpleidea, not me. :-)

[thekix]

Hello,

I think this bug could be closed. If you run offlineimap with Python2 this error should not happen. If you run Python3, please, use offlineimap3.

Regards,
kix