On 12/08/07, Leslie Satenstein <lsatenstein@xxxxxxxxx> wrote: > I guess then I should report then that pup is very unsociable and at times > hogs the network, blocking other applications from sharing eth0. Please keep discussions on-list. Applications do not get blocked from sharing eth0 or any other network interface, at least not without malice. Applications will saturate a connection if they so require - fire up a bittorrent share of f8t1, then try and browse the internet and you'll see what I mean. > Yes, I used the test environment to copy fc8 kernel to fc7. > I was noting what I was experiencing and that either pup was not put at the > back of the dispatch queue , eh? > and so, neither Firefox or evolution could > startup properly, and the latter twp timed out. > > My experience is other kernels, which supported multiplel round robin > queues, (Interactive, batch and I/O queue), and not with linux). > > Why then should pup (from my perspective) hang on to eth0 for so long. I > thought that tcp/ip could multi-thread. You're talking nonsense. Applications are multi-threaded, protocols know nothing of threads and do not implement them. Please also don't use words like backport unless you understand their meaning. > And yes, I wrote and I admitted that it was qualitative. I was hoping for > some feedback which would be informative. If I didn't think something was > amis, I would not have written. I guess it is best to wait for November; > why waste my time, and yours. Quite. You'll have people chasing rainbows with posts like this. Please also do not top-post. Chris -- http://www.chruz.com -- fedora-test-list mailing list fedora-test-list@xxxxxxxxxx To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-test-list
