On Thu, 2006-08-31 at 16:43 +0100, Will McDonald wrote:
On 31/08/06, Bowie Bailey Bowie_Bailey@buc.com wrote:
Will McDonald wrote:
From your second link:
Yes, I did get lots of duplicate packages but I've already fixed that. Basically, what I did was to run Adam Stoke's script to remove duplcate packages. It can be located at http://astokes.org/?q=node/49
That website is no longer working. Is there still a copy of this script floating around?
Not that I could find unfortunaltey. I've just done...
[root@willspc ~]# for package in `rpm -qa --qf "%{NAME}.%{arch}\n" | sort | uniq -d | grep -v kernel`; do rpm -q $package; done
Which lists all base duplicated packages (though at least kernel and gpg-pubkey are this way by design I believe). Then manually removed the older packages. You could THORETICALLY do...
# for duppackage in `rpm -qa --qf "%{NAME}.%{arch}\n" | sort | uniq -d | grep -Ev '(kernel)|(gpg-pubkey)'`; do for olderpackage in `rpm -q $duppackage | head -1` do rpm -e $olderpackage; done done
WARNING: ^^^^ That's entirely untested and reliant on the ordering from rpm output, which looks consistent but I couldn't *swear* to it. So be careful! :)
Will. _______________________________________________
personally, I would remove the new package and re-update in case something in cleanup is required for proper operations.
what does everyone think.
Not trying to jinx myself, but I have never had this problem using centos on hundreds of servers.
Thanks, Johnny Hughes