Yes, it turned out to be some malfunctioning switch hardware had
messed up the load balancing. Hopefully now we'll see both the new 
bandwidth, and reliable access, together.

-- Don

duncan.coutts:
> Just to let you and others know, the problem with the network has now
> been identified and, we think, fixed.
> 
> On Wed, 2008-09-03 at 23:28 +0200, Matthias Kilian wrote:
> > On Wed, Aug 27, 2008 at 11:22:33PM -0700, Don Stewart wrote:
> > > > - darcs.haskell.org is very slow, and sometimes not reachable at all, at
> > > >   least whenever I tried to darcs get, darcs pull or just wget
> > > >   something from it ;-)
> > > >   I don't know wether it has bad connectivity at all, or wether
> > > >   some bandwith-limiting is in action.
> > > 
> > > We just upgraded the bandwidth,
> > 
> > Yeah, to 1 *packet* per hour.
> > 
> > > but it still keeps getting saturated
> > > with hackage downloads. I'll raise the issue with the administrators.
> > 
> > Move this shit to another server, ASAP. Damit. I wasted ONE COMPLETE
> > WEEK (well, only my spare time after my day job) trying to get
> > current sources on my development machine and trying to write some
> > mirroring script (which doesn't work either, since darcs.haskell.org
> > just sucks if it comes to connectivity). I only wanted to build a
> > working ghc-HEAD to get this .cabal -> makefile snippet thing done,
> > but how could I do without a working (and current) ghc?
> > 
> > Sorry, but if the Haskell community backs on a main darcs repository
> > that JUST DOES NOT WORK, and if the most important compiler DOES
> > NOT CARE ABOUT PORTABILITY (I mean the HC bootstrapping, which has
> > been pushed from milestone to milestone), then I'm not any longer
> > interested at all.
> > 
> >             -- Kili
> > 
> 

_______________________________________________
Cvs-ghc mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-ghc

Reply via email to