Software Management call for RFEs
jzeleny at redhat.com
Mon May 27 08:24:33 UTC 2013
On 27. 5. 2013 at 09:43:15, Florian Weimer wrote:
> On 05/27/2013 09:32 AM, Jan Zelený wrote:
> > On 25. 5. 2013 at 09:34:32, Nico Kadel-Garcia wrote:
> >> On Wed, May 22, 2013 at 11:55 AM, Michael Ekstrand <michael at elehack.net>
> > wrote:
> >>> Performance improvement: improve scaling to 5K+ installed packages.
> >> * Amen. This is particularly compounded by poor caching default
> >> behavior, so that a few yum commands in a row each wind up reaching
> >> out to downloading metadata again, and again, and again.
> >> I think this can be addressed by moving the metadata updates to a
> >> different function, and calling it *separately* only as needed. The
> >> Debian "apt" tool does this quite effectively.
> > Unfortunately there is not much we can do about this. Debian has
> > completely
> > different repository policy - they keep all versions of packages in the
> > repo so there is no need to update metadata on client machines every
> > time.
> I don't quite understand this comment.
> Debian repository policy varies quite a bit. Some repositories keep old
> versions, some don't. Mostly the latter, actually, because not all
> repository managers (there a couple of implementations) can deal with
> multiple versions for a single package/architecture combination.
I'm sorry but the Debian repositories say otherwise, see Iceweasel for
Multiple old versions are kept in there. That's why they don't need to update
their metadata every single time - the old ones are simply still valid.
> As far as I can tell, the main difference is that apt-get and apt-cache
> read very few, relatively large files at the beginning, so they don't
> block on disk reads early.
> dpkg, on the other hand, uses a database scatter across many small files
> on disk, so you get the delay only when you actually install or remove
> any packages. At the beginning, this is quite fast, but eventually, the
> files will be scattered quite badly, and there is a considerable delay
> at this step.
This part is about disk read-write but that was not what I was writing about.
>From my experience users mostly complain about the metadata download which is
More information about the devel