From: Allen Kistler on
Robert Heller wrote:
> gollum.deepsoft.com% sudo yum check-update
> Password:
> Setting up repositories
> Reading repository metadata in from local files
> primary.xml.gz 100% |=========================| 419 kB 01:28
> http://www.gtlib.gatech.edu/pub/fedora-epel/4/i386/repodata/primary.xml.gz: [Err
> no -1] Metadata file does not match checksum
> Trying other mirror.
> primary.xml.gz 100% |=========================| 419 kB 01:25
> http://mirror.hiwaay.net/pub/epel/4/i386/repodata/primary.xml.gz: [Errno -1] Met
> adata file does not match checksum
> Trying other mirror.
> primary.xml.gz 100% |=========================| 419 kB 01:26
> http://mirrors.usu.edu/epel/4/i386/repodata/primary.xml.gz: [Errno -1] Metadata
> file does not match checksum
> Trying other mirror.
> primary.xml.gz 100% |=========================| 419 kB 01:26
> http://mirror.unl.edu/fedora/epel/4/i386/repodata/primary.xml.gz: [Errno -1] Met
> adata file does not match checksum
> Trying other mirror.
> primary.xml.gz 100% |=========================| 419 kB 01:26
> http://mirrors.rit.edu/epel/4/i386/repodata/primary.xml.gz: [Errno -1] Metadata
> file does not match checksum
> Trying other mirror.
> primary.xml.gz 100% |=========================| 419 kB 01:25
> http://mirrors.xmission.com/fedora/epel/4/i386/repodata/primary.xml.gz: [Errno -
> 1] Metadata file does not match checksum
> Trying other mirror.
> primary.xml.gz 100% |=========================| 88 kB 01:12
> http://mirrors.tummy.com/pub/fedora.redhat.com/epel/4/i386/repodata/primary.xml.
> gz: [Errno 4] Socket Error: timed out
> Trying other mirror.
> primary.xml.gz 100% |=========================| 419 kB 01:26
> http://linux.nssl.noaa.gov/epel/4/i386/repodata/primary.xml.gz: [Errno -1] Metad
> ata file does not match checksum
> Trying other mirror.
> Error: failure: repodata/primary.xml.gz from epel: [Errno 256] No more mirrors t
> o try.
> heller has logged on pts/3 from local at 01-Jun-2008 7:46.
> gollum.deepsoft.com%
>

I got the same thing Friday on the RHN repositories. I waited. I tried
again. It was okay. Apparently Red Hat's createrepo runs were bombing
part way through.

I've seen it happen on my own repos when there's a corrupted package
that createrepo can't handle. You have to hunt down the one bad package
and replace it with a good copy. Of course, I don't *know* that that's
what's happening here, so YMMV.