SCIENTIFIC-LINUX-DEVEL Archives

September 2007

SCIENTIFIC-LINUX-DEVEL@LISTSERV.FNAL.GOV

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jan Iven <[log in to unmask]>
Reply To:
Date:
Wed, 19 Sep 2007 10:02:05 +0200
Content-Type:
text/plain
Parts/Attachments:
text/plain (43 lines)
On 18/09/07 20:54, Troy Dawson wrote:
> Hi Jan,
> Just because we didn't reply to this, didn't mean we didn't read it and 
> are looking at it.
> We've also found that the pam packages are this way too.  We're in the 
> process of rebuilding them.
> The two questions are,
> Where do we put them when we've recompiled them?  In the errata?  They 
> will definatly make it into SL 5.1.

(or SL4.6, I guess/hope :-)

I don't think they will get applied if they are in the errata (unless 
you change the release number, of course), since the "base" repository 
will usually come first. Which means we have the optiosn to
- replace in base (nasty, CD/DVD is different, but helps people who'd 
like to newly install these RPMS, and nobody else would notice
- bump release and put into errata
- defer to next release, keep in errata until then and tell people to 
manually overwrite in their repo if they see trouble.

> and Is there a nice way to check all these packages?  I'd hate to find 
> them all one by one as we get errors.

Hm. Quick+dirty, looking at man pages only (not sure which other files 
get silently "ignored" on a multilib install)

cd ..../en/os/x86_64/SL/RPMS;
for i in *.i?86.rpm; do
  j=${i%%.i?86.rpm}.x86_64.rpm;
  if [ ! -e $j ]; then
     next;
  fi;
  echo $i - $j;
  rpm -qp --qf '[%{FILENAMES} %{FILEMD5S}\n]' $i | grep '/man/'|sort > 
${TMP:-/tmp}/$i.files;
  rpm -qp --qf '[%{FILENAMES} %{FILEMD5S}\n]' $j | grep '/man/' | sort > 
${TMP:-/tmp}/$j.files;

  diff -u ${TMP:-/tmp}/$i.files  ${TMP:-/tmp}/$j.files || echo "Warning: 
$i/$j different";
done

ATOM RSS1 RSS2