Re: What is broken with software update?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 16 September 2015 at 18:46, Gordon Messmer <gordon.messmer@xxxxxxxxx> wrote:
> Never was sure why; I thought that's what packagekitd was *for*.

PackageKit needs to offer clients results with guaranteed latencies;
it's no good if the command-not-found handler takes 700ms to return
results, or gnome-software takes 15 seconds to show the first overview
screen. This is why some of the fancier stuff was turned off with the
yum backend. The problem with sharing a cache is that PackageKit
downloads an entire set of repository files at the same time, all
coherent to each other. It then atomically swaps the link in the cache
so that at any time you can query PackageKit for results and get
something sane in a small amount of time.

What dnf does, and yum did, is download only the files that it needs.
This is great for command line use where you might just want to check
for updates, but means that we have metadata from different
repositories with different ages, and most importantly different
metadata files in the same repo with different ages (repo files
reference each other internally). This means using dnf that you can do
something like a search with no additional downloads, but when you
come to install something it could be that half way through the
depsolve it works out that the filelists isn't new enough and starts
downloading a new metadata file, so it can re-depsolve and continue on
to install.

PackageKit can't do this, as we only promise network access during
certain method calls, and we have strict latency requirements for some
requests, where downloading at random points would shatter both
promises.

If PK and DNF could agree to download all metadata at a set time for
all repositories and do an atomic switch that would be great, and we
could certainly share a cache. Until then, it makes sense to have
separate locations.

What we can fix is downloading the same file twice in some cases, and
in that case either squid helps, or CAShe can fix that in the longer
term.

Richard.
-- 
users mailing list
users@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe or change subscription options:
https://admin.fedoraproject.org/mailman/listinfo/users
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines
Have a question? Ask away: http://ask.fedoraproject.org



[Index of Archives]     [Older Fedora Users]     [Fedora Announce]     [Fedora Package Announce]     [EPEL Announce]     [EPEL Devel]     [Fedora Magazine]     [Fedora Summer Coding]     [Fedora Laptop]     [Fedora Cloud]     [Fedora Advisory Board]     [Fedora Education]     [Fedora Security]     [Fedora Scitech]     [Fedora Robotics]     [Fedora Infrastructure]     [Fedora Websites]     [Anaconda Devel]     [Fedora Devel Java]     [Fedora Desktop]     [Fedora Fonts]     [Fedora Marketing]     [Fedora Management Tools]     [Fedora Mentors]     [Fedora Package Review]     [Fedora R Devel]     [Fedora PHP Devel]     [Kickstart]     [Fedora Music]     [Fedora Packaging]     [Fedora SELinux]     [Fedora Legal]     [Fedora Kernel]     [Fedora OCaml]     [Coolkey]     [Virtualization Tools]     [ET Management Tools]     [Yum Users]     [Yosemite News]     [Gnome Users]     [KDE Users]     [Fedora Art]     [Fedora Docs]     [Fedora Sparc]     [Libvirt Users]     [Fedora ARM]

  Powered by Linux