Owen Taylor wrote:
*however* - updates are more of a challenge, because we're still
changing things rapidly and the client software is tightly coupled to
the server software; when we add new features, there is often both a
client part and a server part.
[...]
Do people have other ideas about how it could work
Welcome to the world of distributed computing in which software updates
can not be done atomically. Even certain large telcos have had
problems with this.
How about running multiple servers, one for each supported but
incompatible version of the client software? You could use a common
primary server/URL with redirects to the version-specific server.
John
--
fedora-extras-list mailing list
fedora-extras-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/fedora-extras-list
