On Tue, 26 Aug 2008, Konstantin Olchanski wrote:

> On Tue, Aug 26, 2008 at 08:58:40AM -0700, Keith Lofstrom wrote:
>> http://blog.vipul.net/2008/08/24/redhat-perl-what-a-tragedy/
>
> Please read the explanation written by the person who fixed
> the relevant perl performance bugs:
>
> http://use.perl.org/~nicholas/journal/37274
>
> From reading his cvs commit comments, it looks like the issue is
> performance bugs fixed back in October 2007, but never propagated
> to the version packaged by Red Hat.
>
> I am surprised at the ranting & raving, as this situation has happened
> before and we know exactly how to deal with it:
>
> 1) 1st tier distributor (Red Hat) pushes a bum package (today it's perl,
> yesterday it was bum kernels)
>
> 2) 2nd tier distributor (SL or CENTOS) fixes it (bum kernels) or does
> not fix it due to manpower and experitise limitations (bum perl).

We did not fix anything because we did not know it was broken.

>
> 3) 3rd tier distributor (TRIUMF computing group) fixes it for
> the local site (or does not fix it)

Next time please tell us about this and we will put the new packages on 
the ftp site so that others can use it.

>
> 4) end user fixes it locally on his machine (build from source,
> installs fixed RH rpms linked from the bugzilla bug entry,
> or installs Fedora rpms, whatever).
>
>

-Connie Sieh