Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 18 Jun 2013 14:20:02 GMT
From:      John Hein <jhein@symmetricom.com>
To:        freebsd-ports-bugs@FreeBSD.org
Subject:   Re: ports/179676: deskutils/org-mode.el: checksum mismatch
Message-ID:  <201306181420.r5IEK2Hb082243@freefall.freebsd.org>

next in thread | raw e-mail | index | archive | help
The following reply was made to PR ports/179676; it has been noted by GNATS.

From: John Hein <jhein@symmetricom.com>
To: John Marino <draco@marino.st>
Cc: <bug-followup@FreeBSD.org>, <dryice@freebsd.org>
Subject: Re: ports/179676: deskutils/org-mode.el: checksum mismatch
Date: Tue, 18 Jun 2013 08:09:56 -0600

 John Marino wrote at 15:38 +0200 on Jun 18, 2013:
  > On 6/18/2013 15:27, John Hein wrote:
  > > Well, dryice@ updated it in ports/168944 (I never got any notice about that
  > > from him or from gnats - I'm not sure how that came to pass).
  > >
  > > I never saw or approved 168944.  The 168944 changes seems mostly fine,
  > > but I never got a 7.8.11 distfile, so I have no original to check
  > > against.
  > >
  > > That said, however, when I do a fresh 'make fetch' now, the tarball
  > > looks fine to me...
  > >
  > > John Marino, maybe you have a partial tarball (from an interrupted
  > > fetch, perhaps)?  Try moving it away and running 'make fetch' again.
  >
  > No, the dryice commit was in July-2012.  org-mode.el fetched and built 
  > fine as of two months ago.
 
 'No' to what?  You don't have a partial tarball?  Can you provide
 ls(1) or stat(1) output?  Manual sha256(1)?
 
 I know when it was committed.  I do understand you think an upstream
 re-roll may have occurred.  I am trying to check that for you and
 have documented what I've found so far.
 
 
  > 'make fetch' is meaningless.  You can fetch with a bad checksum.  It's
  > 'make checksum' that has to pass.
  >
  > The fetch was not interrupted.  There's no partial file.
  > "make checksum" continues to fail.
  >
  > John, please retry but with "make checksum" instead of "make fetch".
 
 I didn't say 'make fetch' ran the checksum.  I said the tarball looked
 fine to me and included the manual sha256(1) and ls(1) output to prove
 it.  But for your edification...
 
 % make checksum
 ===> Fetching all distfiles required by org-mode.el-emacs24-7.8.11_3 for building
 => SHA256 Checksum OK for org-7.8.11.tar.gz.
 
 Did you try to do what I asked - move away your tarball and re-run
 'make fetch' (or 'make checksum' if you wish)?  If so, is the
 tarball still failing the checksum?  Can you include your output
 (which will show the fetch URL).
 
 
 Do you have a record of your original fetch (i.e., the output of the
 fetch showing success and where it came from)?
 
 When I fetched, it came from http://orgmode.org/org-7.8.11.tar.gz
 
 % curl http://orgmode.org/org-7.8.11.tar.gz | sha256
   % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                  Dload  Upload   Total   Spent    Left  Speed
 100 3226k  100 3226k    0     0   126k      0  0:00:25  0:00:25 --:--:--  142k
 2f6a69aaff5cd66162a9a1a6df900c6050434bced1bbc7eea715b5579a0487c1
 
 That matches the sha256 in distinfo.
 
 If your tarball's size matches, but the sha256 doesn't, we can find out
 where with a hex dump of the file.  Post output of 'hd org-7.8.11.tar.gz'
 somewhere or send in private mail, and I will check against mine.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201306181420.r5IEK2Hb082243>