[CentOS] A minor beef
crichey at gmail.com
Fri Nov 25 22:42:50 UTC 2005
On 11/25/05, Bryan J. Smith <thebs413 at earthlink.net> wrote:
[ very long but not relevant rant about legal/not-legal software snipped ]
> In what regard? Other than the recent 4.2 snafus, what retrieve hell
> have you experienced with _just_ CentOS (_no_ DAG)? I haven't
> experienced too many performance issues with CentOS either.
Read back through the posts again. I very clearly stated that I was
having trouble retrieving CentOS, not Dag packages.
[ more scarcely relevant stuff about repository hell snipped ]
I repeat - no repository hell, none, zip, nada.
I am not experiencing problems with repository interrelationships just
pure falure at certain (not easily predictable or repeatable) points
retrieving certain packages (one was a man update) from CentOS
resulting in abort of 'yum update' and/or total hang of yum (cancel,
clear lock and retry). The total process required 6 restarts over a
period of several hours, and finally the updates completed. I can't
repeat this process because the system in question is now up to date.
During this time I received no indication of inability to communicate
with other external sites on my other systems that are on the same
> So why post this if you don't know how to help?
Knowledge that a problem exists precedes any attempts to fix it.
> > I'm quite surprised to encounter
> > this type of delay during supposedly "normal" times.
> I don't know what to tell you, I haven't.
> The reality is that CentOS is pretty popular, and may not have the same
> support structure of some other projects.
That's about the most relevant part of this entire diatribe. You
haven't experienced the problem, but I did. If CentOS does not have
the same structure as some other projects, then what can we do to help
improve the structure? If we can't report a problem without a lengthy
lecture on the theory of package selection by several different
distros which has absolutely no bearing on the ability to retrieve
CentOS packages, I suppose we should just ignore the problem.
What I experienced might well be a one-off scenario, but I can't know
that without reporting the problem and inquiring if others are
experiencing similar problems. I did get a few reports to indicate
that no one else appears to be experiencing this problem, and I am
thankful for the concrete suggestions for circumventing the problem.
Debugging is twice as hard as writing the code ... If you write
the code as cleverly as possible, you are, by definition, not
smart enough to debug it.
More information about the CentOS