home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Columbia Kermit
/
kermit.zip
/
newsgroups
/
misc.20041116-20060924
/
000170_fdc@columbia.edu_Mon Oct 17 14:36:58 2005.msg
< prev
next >
Wrap
Internet Message Format
|
2020-01-01
|
4KB
Path: newsmaster.cc.columbia.edu!not-for-mail
From: Frank da Cruz <fdc@columbia.edu>
Newsgroups: comp.protocols.kermit.misc
Subject: Re: ftp mput /recursive problem....
Date: 17 Oct 2005 18:36:42 GMT
Organization: Columbia University
Lines: 66
Message-ID: <slrndl7rpq.d73.fdc@sesame.cc.columbia.edu>
References: <dip2gm$ci$1$8300dec7@news.demon.co.uk> <slrndl29i0.ldi.fdc@sesame.cc.columbia.edu> <dirqev$7j$1$830fa7a5@news.demon.co.uk> <slrndl4vo5.mfv.fdc@sesame.cc.columbia.edu> <diuls4$fj3$1$830fa795@news.demon.co.uk> <3ZL4f.2292$h25.38@news-wrt-01.rdc-nyc.rr.com> <dj07ch$i4c$1$8302bc10@news.demon.co.uk>
Reply-To: fdc@columbia.edu
NNTP-Posting-Host: sesame.cc.columbia.edu
X-Trace: newsmaster.cc.columbia.edu 1129574202 9762 128.59.59.56 (17 Oct 2005 18:36:42 GMT)
X-Complaints-To: postmaster@columbia.edu
NNTP-Posting-Date: 17 Oct 2005 18:36:42 GMT
User-Agent: slrn/0.9.8.0 (SunOS)
Xref: newsmaster.cc.columbia.edu comp.protocols.kermit.misc:15414
On 2005-10-17, John Hackett <john.hackett@icon-is.co.uk> wrote:
: That was my first thought when I began this and had various
: conversations with Technical Support at LaCie: but then I ran the
: complete backup with gFTP against the LaCie and it worked fine.
:
Perhaps gFTP is issuing a different reportoire of commands to the server
and doesn't tickle the server bug, assuming there is one.
: Also, I
: have run this Kermit script against both the LaCie and another Fedora3
: server and both failed in the same way at the same point - which seems
: to suggest some problem with Kermit rather than the server.
:
Is it the same FTP server in both cases?
: As this was really a second line of backup I think I may give up on this
: and try a different approach - As I already have an cron'd rsync running
: to give server -> backup I think I'll just mount the LaCie on the backup
: server and use rsync to it instead. A bit less flexible for what I
: originally bought the Lacie for but unless anyone has any other bright
: ideas.....
:
Let's look at this transcript again:
:>>SAVED IN: /mirror/sarah/profile/Application Data/Skype/sarah_hac/
:>>---> CDUP
:>>250 CDUP command successful
:>>
:>>NOW: /mirror/sarah/profile/Application Data/Skype/
:>>---> CDUP
:>>250 CDUP command successful
:>>
:>>NOW: /mirror/sarah/profile/Application Data/
:>>---> CWD Real
:>>550 Real: No such file or directory
:>>---> MKD Real
:
Kermit has CDUP'd to the server's directory:
/mirror/sarah/profile/Application Data/
and now asks to CD to the "Real" subdirectory of that directory. The server
says there isn't one. But there is, right? So either one or more of the
prior CDUP commands didn't actually work (even thought the server said they
did, as Jeff said) or the server botched the CWD command, reporting the
target directory did not exist when it did. The next bit shows that the
former theory is the right one:
:>>257 "/mirror/sarah/profile/Application Data/Skype/Real" - Directory
:>>successfully created
:>>CREATED DIRECTORY Real
:>>---> CWD Real
:>>250 CWD command successful
:
The CDUP out of the Skype subdirectory was not successful, despite the
server's status report, exactly what Jeff said, which is why the CWD command
did not find the Real subdirectory and therefore Kermit had to ask the
server to create one.
If you can't change or fix the server, then I agree you'll have to give up
on Kermit, but I don't think it's Kermit's fault. Conceivably I could have
Kermit do a PWD after each CDUP to check that the server is not lying when
it says it worked, but that would slow things down a lot on certain kinds
of transfers.
- Frank