Michael Albinus | 7 Aug 2005 18:26
Picon
Picon
Gravatar

Tramp 2.0.50

Hello,

Stable Tramp 2.0.50 has been released. It is mainly a bug fix release.

For production systems, it is recommended to upgrade to this
version. Recent Emacs 22.0.50 and XEmacs 21.5.b21 will be upgraded.

The download files are located at ftp://ftp.gnu.org/gnu/tramp/

Best regards, Michael.
Kai Großjohann | 8 Aug 2005 12:55
Picon

Re: Tramp on Windows XP

Please try PuTTY (or the plink tool that comes with it).  There is a
Tramp method "plink", so you only need to add the PuTTY directory to
%PATH% and then C-x C-f /plink:user <at> host:/some/file RET.

I hope.

The problem is that Windows does not have pseudo ttys.  So Emacs does
not use them for subprocesses.  And when ssh sees that a pseudo tty is
not used, then it won't allocate one on the remote end.  And this
causes the remote end to think it is talking to a program, and to omit
printing the shell prompt.  But the shell prompt is very important for
Tramp's operation.

Doing "ssh -t -t" (yes, twice), it is sometimes possible to force ssh
to allocate a pseudo tty on the remote end, but apparently this does
not work reliably on Windows.

Kai

Joel Gwynn <joelman <at> gmail.com> writes:

> Hmm.  eshell certainly does hang.  All I get is this message:
> Pseudo-terminal will not be allocated because stdin is not a terminal.
>
> What's that all about?
>
> On 6/11/05, John L. Utz III <john <at> utzweb.net> wrote:
>> I cant answer why emacs doesnt do the right thing in this case, and i am
>> surprised that you didnt get the imfamous 'terminal is not a TTY' ( google for
>> it :-( )
(Continue reading)

Nelson E Ingersoll | 11 Aug 2005 22:03

Hanging at mime-encode on Win2K -to- Tru64


   I have used ange-ftp for a long time.  I've decided to give TRAMP a
shot.

   I cannot get TRAMP to complete the login sequence onto my Tru64
system.  It does complete when I log onto a SunOS system.

   When logging onto my Tru64 system TRAMP hangs at the "... Perl
`mime-encode' implementations." line until I type a Ctrl-G.  No errors
are displayed.  Came to the mailing list and poked around looking for
help.  I found information about setting the 'chunksize' up.  So I did
it.   Alas, I am still unable to get past the "Perl 'mime-encode'
implementations'  I've set chucksize as low as 250 and as high as 100000.

   On the Tru64 system I normally log into and use the 'tcsh' shell.
However, I do the same thing on a SunOS system and I can use TRAMP with
the SunOS login.  Any thoughts?  I am a TRAMP neophyte.

- Nelson ...

PS - 'plink IS my default tramp-method.
     (setq tramp-default-method "plink")

     tramp-version's value is "2.0.49
     emacs-version's value is "22.0.50.1"
     Client OS:   Windows 2000
     Client SSH:  plink: Release 0.58

     Remote OS: Compaq Tru64 4.0d
     Remote SSH: openssh37
(Continue reading)

Michael Albinus | 12 Aug 2005 09:45
Picon
Picon
Gravatar

Re: Hanging at mime-encode on Win2K -to- Tru64

Nelson E Ingersoll <ningersoll <at> cso.atmel.com> writes:

>    I cannot get TRAMP to complete the login sequence onto my Tru64
> system.  It does complete when I log onto a SunOS system.
>
>    When logging onto my Tru64 system TRAMP hangs at the "... Perl
> `mime-encode' implementations." line until I type a Ctrl-G.  No errors
> are displayed.  Came to the mailing list and poked around looking for
> help.  I found information about setting the 'chunksize' up.  So I did
> it.   Alas, I am still unable to get past the "Perl 'mime-encode'
> implementations'  I've set chucksize as low as 250 and as high as 100000.

But it sounds like a chunksize related problem. AFAIR, True64 target
systems have been reported to suffer from this problem, and sending
the mimencode Perl function is the first bigger string sent to the
target system; therefore the problem happens first here.

Best would be if you set

     (setq tramp-debug-buffer t
           tramp-verbose 10)

Then rerun your test, and submit a bug report via "M-x tramp-submit-bug".
Both buffers  *tramp/foo* and *debug tramp/foo* would be helpful for
analyzing; please provide them too.

Best regards, Michael.
Nelson E Ingersoll | 12 Aug 2005 21:59

tramp (2.0.49); Failure to finish connection to Tru64 v4.0d system


Executed these commands manually:

    (setq tramp-default-method "plink")
    (setq tramp-debug-buffer t tramp-verbose 10)

then performed a simple

    C-x C-f /ingersol <at> ash:work/RET

to get a directory listing from which to start working.  Lots of
activity and then failure following the "Sending the Perl
`mime-encode' implementations."  sequence.

   This happens every time I attempt to use plink (v0.58) on my Win2K
system.  I can connect to a SunOS system without trouble.

--bug report follows this line--

Files attached:

tramp-submit-bug.txt
tramp_plink-ingersol <at> ash
debug_tramp_plink_ingersol <at> ash
Attachment (debug_tramp_plink_ingersol@ash): application/octet-stream, 8 KiB
tramp_executable /bin/perl
(Continue reading)

Kai Großjohann | 12 Aug 2005 22:21
Picon

Re: tramp (2.0.49); Failure to finish connection to Tru64 v4.0d system

Nelson E Ingersoll <ningersoll <at> cso.atmel.com> writes:

> to get a directory listing from which to start working.  Lots of
> activity and then failure following the "Sending the Perl
> `mime-encode' implementations."  sequence.

Could you set tramp-chunksize to a smaller value?  Report your
findings.

Kai
Michael Albinus | 15 Aug 2005 11:01
Picon
Picon
Gravatar

Re: tramp (2.0.49); Failure to finish connection to Tru64 v4.0d system

Nelson E Ingersoll <ningersoll <at> cso.atmel.com> writes:

> Executed these commands manually:
>
>     (setq tramp-default-method "plink")
>     (setq tramp-debug-buffer t tramp-verbose 10)
>
> then performed a simple
>
>     C-x C-f /ingersol <at> ash:work/RET
>
> to get a directory listing from which to start working.  Lots of
> activity and then failure following the "Sending the Perl
> `mime-encode' implementations."  sequence.
>
>    This happens every time I attempt to use plink (v0.58) on my Win2K
> system.  I can connect to a SunOS system without trouble.

As Kai already wrote, try a smaller tramp-chunksize. I've scanned
shortly the mail archive; Tru64 has been reported to be bitchy with
longer string chunks. Maybe we should give it a corresponding default
value.

JFTR: both tramp and tramp debug buffers have not been appended to
your mail. You did copy just the 
Attachment: application/octet-stream, 108 bytes
_______________________________________________
Tramp-devel mailing list
Tramp-devel <at> gnu.org
(Continue reading)

Kai Großjohann | 15 Aug 2005 11:40
Picon

Re: Hanging at mime-encode on Win2K -to- Tru64

Nelson E Ingersoll <ningersoll <at> cso.atmel.com> writes:

>    When logging onto my Tru64 system TRAMP hangs at the "... Perl
> `mime-encode' implementations." line until I type a Ctrl-G.  No errors
> are displayed.  Came to the mailing list and poked around looking for
> help.  I found information about setting the 'chunksize' up.  So I did
> it.   Alas, I am still unable to get past the "Perl 'mime-encode'
> implementations'  I've set chucksize as low as 250 and as high as 100000.

I think I told you to *decrease* the value, not *increase* it.

Maybe 250 wasn't low enough.  Please try 1 as the value, too.  Tramp
will be very slow.  But it will tell us whether this is indeed the
problem.

Thanks,
Kai
Nelson E Ingersoll | 16 Aug 2005 17:32

Re: Hanging at mime-encode on Win2K -to- Tru64

Good morning Kai and Michael,

   I have a hopefully helpful report to make.  I apologize for taking
several days to respond.  Lot's of work and not enough hands.  One has
to leave time for gaming and sleep!  9-)

   Following Kai's instructions I reduced the tramp-chunksize value to
1.  Good call.   It did take FOREVER to complete; however, complete it
did.  Following that success I started increasing tramp-chunksize 2X
iteratively until failure.  I restarted Emacs cleanly every test thereby
forcing the entire loading sequence to be performed.

   When I set tramp-chunksize to 128 there was failure.  Knowing full
well that 127 is really the boundary value, I dropped tramp-chunksize to
127 and re-tested.  Connection and subsequent communication worked with
tramp-chunksize set to 127.  Obviously something is limited to 127 bytes
under Tru64.  Presumably the next question is what is limited?

   I can report that, as far as Tru64 v4.0d is concerned, setting
tramp-chunksize to 127 works for me.  If there are other tests, or if
you wish me to provide one or more debugging logs please let me know.  I
am happy, no delighted, to help the community.

- Nelson Ingersoll

Kai Großjohann wrote:
> Nelson E Ingersoll <ningersoll <at> cso.atmel.com> writes:
> 
> 
>>   When logging onto my Tru64 system TRAMP hangs at the "... Perl
(Continue reading)

Kai Großjohann | 17 Aug 2005 11:10
Picon

Re: Hanging at mime-encode on Win2K -to- Tru64

Nelson E Ingersoll <ningersoll <at> cso.atmel.com> writes:

>    I can report that, as far as Tru64 v4.0d is concerned, setting
> tramp-chunksize to 127 works for me.  If there are other tests, or if
> you wish me to provide one or more debugging logs please let me know.  I
> am happy, no delighted, to help the community.

It is good that you have now found a value that works.  You seem to
think that there is a size limit of 127 somewhere.  Perhaps that is
not the case, and it is just a matter of time.  If you would like to
investigate further, you could try to tweak the time it waits after
each chunk.  With a larger wait time between chunks, it should be
possible to increase the chunksize somewhat.

Currently, it sleeps for 0.1 seconds between chunks.  You can find the
sleep-for call in tramp.el, in the function tramp-send-string.  You
can just go to that file, edit the 0.1 into something else (0.2,
perhaps, or 0.5 or 1.0) and then hit C-M-x with point still inside the
defun.  Then you can try again if a larger value of tramp-chunksize
works.

You should kill the *tramp/foo* and the *debug tramp/foo* buffers
between attempts, but other than that no restart of any kind should be
required -- just tweak the defun, tweak the tramp-chunksize value,
kill the buffers, and try again.  That is the beauty of an interactive
environment like Lisp -- very rapid tweak and try cycles :-)

You can use M-: (setq tramp-chunksize 42) RET to change the value of
tramp-chunksize.  Note that I really mean M-: and not M-x.

(Continue reading)


Gmane