nnimap

nnimap

[George McNinch]

This probably belongs in the thread "More IMAP testing, please"
but I'm not sure quite where, so...

Anyhow, I'm somewhat *erratically* having trouble with one of my imap
servers and the new nnimap.

From a "clean" start -- restart emacs, M-x gnus --  *usually* gnus hangs with 

        "Opening TLS connection to imap.tufts.edu...Done" 

in the minibuffer.  After killing it with C-g, I find two

*nnimap imap.tufts.edu nil  *nntpd**<n>

buffers.

Buffer *nnimap imap.tufts.edu nil  *nntpd** contains lots of FETCH's

Here is a <snipped> version:

       * FLAGS (\Answered \Flagged \Draft \Deleted \Seen gnus-forward gnus-save Junk $Label1 $Label2 $Label3 $La\
       bel4 $Label5 gnus-expire NonJunk $Forwarded $MDNSent gnus-dormant)^M
       * OK [PERMANENTFLAGS ()]  ^M
       * 17 EXISTS^M
       * 0 RECENT^M
       * OK [UIDVALIDITY 1088476165]  ^M
       * OK [UIDNEXT 22707]  ^M
       122 OK [READ-ONLY] Completed^M
       * 5 FETCH (FLAGS (\Flagged \Seen) UID 22671)^M
       * 6 FETCH (FLAGS (\Seen) UID 22676)^M
       * 7 FETCH (FLAGS (\Flagged \Seen) UID 22678)^M
       * 8 FETCH (FLAGS (\Answered \Seen) UID 22679)^M

       <snip>

       * 57 FETCH (FLAGS (\Seen) UID 62)^M
       * 58 FETCH (FLAGS (\Seen NonJunk) UID 63)^M
       * 59 FETCH (FLAGS (\Answered \Seen) UID 64)^M
       145 OK Completed^M
       * FLAGS (\Answered \Flagged \Draft \Deleted \Seen NonJunk $Label1)^M
       * OK [PERMANENTFLAGS ()]  ^M
       * 93 EXISTS^M
       * 0 RECENT^M
       * OK [UIDVALIDITY 1257445250]  ^M
       * OK [UIDNEXT 94]  ^M
       146 OK [READ-ONLY] Completed^M
       - Peer has closed the GNUTLS connection

I guess the "Peer has closed..." is my C-g.

Buffer *nnimap imap.tufts.edu nil  *nntpd**<2> contains

   * CAPABILITY IMAP4 IMAP4rev1 ACL QUOTA LITERAL+ NAMESPACE UIDPLUS CHILDREN BINARY UNSELECT LANGUAGE STARTTLS XSENDER X-NETSCAPE XSERVERINFO
   449 OK Completed


The last entries in *imap-log* (after
hanging up) are

        ...
        19:09:41 446 EXAMINE "Dept"
        19:09:41 447 UID FETCH 1:* FLAGS
        19:09:44 448 LOGIN "myusername" "mypasswd"
        19:09:44 449 CAPABILITY

And the last entries in *Messages* are

        Opening TLS connection to `imap.tufts.edu'...
        Opening TLS connection with `gnutls-cli -p imaps imap.tufts.edu'...done
        Opening TLS connection to `imap.tufts.edu'...done

If I comment out the imap.tufts.edu server from
"gnus-secondary-select-methods" in gnus.el and restart, I have no
trouble with my other (gmail) imap.


Is this the "race condition" that was mentioned in a different thread?
I did "git pull" about 10 minutes ago, and this problem persists for me.

A couple of times I *have* gotten a connection to imap.tufts.edu. I don't
seem to be able to predict when it will work and when it won't.

Unfortunately I don't know much about (and forget how to query in a
useful way...) the server imap.tufts.edu

Best,
-george

-- 
George McNinch <gmcninch@gmail.com>

Re: nnimap

[Sebastian Krause]

George McNinch <gmcninch@gmail.com> wrote:
> This probably belongs in the thread "More IMAP testing, please"
> but I'm not sure quite where, so...
>
> Anyhow, I'm somewhat *erratically* having trouble with one of my imap
> servers and the new nnimap.
>
> From a "clean" start -- restart emacs, M-x gnus --  *usually* gnus hangs with 
>
>         "Opening TLS connection to imap.tufts.edu...Done" 
>
> in the minibuffer.  After killing it with C-g, I find two
>
> *nnimap imap.tufts.edu nil  *nntpd**<n>
>
> buffers.

Sounds similar to the problem I've reported in
<http://thread.gmane.org/gmane.emacs.gnus.general/70892/focus=71031>. However,
I only get this problem after leaving Gnus idle for a while. I've
mailed Lars a backtrace yesterday, hope it helps.

Re: nnimap

[Lars Magne Ingebrigtsen]

George McNinch <gmcninch@gmail.com> writes:

> From a "clean" start -- restart emacs, M-x gnus --  *usually* gnus hangs with 
>
>         "Opening TLS connection to imap.tufts.edu...Done" 
>
> in the minibuffer.  After killing it with C-g, I find two

Can you (setq debug-on-error t) and `C-g' when it hangs, and then mail
me the backtrace and the contents of the *nnimap ... tufts* buffer?  And
the imap log buffer.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: nnimap

[Sebastian Krause]

Lars Magne Ingebrigtsen <larsi@gnus.org> wrote:
> George McNinch <gmcninch@gmail.com> writes:
>> From a "clean" start -- restart emacs, M-x gnus --  *usually* gnus hangs with 
>>
>>         "Opening TLS connection to imap.tufts.edu...Done" 
>>
>> in the minibuffer.  After killing it with C-g, I find two
>
> Can you (setq debug-on-error t) and `C-g' when it hangs, and then mail
> me the backtrace and the contents of the *nnimap ... tufts* buffer?  And
> the imap log buffer.

Did you get my two mails with the backtraces? The second one is
probably more useful.

Re: nnimap

[George McNinch]

>>>>> "Lars" == Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

    Lars> Can you (setq debug-on-error t) and `C-g' when it hangs, and
    Lars> then mail me the backtrace and the contents of the *nnimap
    Lars> ... tufts* buffer?  And the imap log buffer.

Huh. It "hangs" but I don't seem to get a backtrace upon `C-g'ing.

-g

-- 
George McNinch <gmcninch@gmail.com>
http://www.tufts.edu/~gmcnin01

Re: nnimap

[George McNinch]

>>>>> "George" == George McNinch <gmcninch@gmail.com> writes:

>>>>> "Lars" == Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
    Lars> Can you (setq debug-on-error t) and `C-g' when it hangs, and
    Lars> then mail me the backtrace and the contents of the *nnimap
    Lars> ... tufts* buffer?  And the imap log buffer.

    George> Huh. It "hangs" but I don't seem to get a backtrace upon
    George> `C-g'ing.

Maybe (setq debug-on-quit t) is the desired thing? That gives a
backtrace...

-- 
George McNinch <gmcninch@gmail.com>
http://www.tufts.edu/~gmcnin01

Re: nnimap

[Lars Magne Ingebrigtsen]

George McNinch <gmcninch@gmail.com> writes:

>     Lars> Can you (setq debug-on-error t) and `C-g' when it hangs, and
>     Lars> then mail me the backtrace and the contents of the *nnimap
>     Lars> ... tufts* buffer?  And the imap log buffer.
>
>     George> Huh. It "hangs" but I don't seem to get a backtrace upon
>     George> `C-g'ing.
>
> Maybe (setq debug-on-quit t) is the desired thing? That gives a
> backtrace...

Yes, sorry.  debug-on-quit.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen