On Sun, 08 Feb 2009 10:11:17 +1000 "James C. McPherson" <James.McPherson at Sun.COM> wrote:
> On Fri, 06 Feb 2009 15:45:08 -0800 > Mike Kupfer <Mike.Kupfer at Sun.COM> wrote: > > > Oops, forgot to respond to something... > > > > >>>>> "meem" == Peter Memishian <Peter.Memishian at sun.com> writes: > > > > meem> (I encourage those of you in MPK to try working with a remote > > meem> ON gate via Mercurial -- it is excruciatingly slow.) > > > > Are you using NFS or ssh? Yes, using Mercurial and NFS for ON is > > prohibitively slow. With ssh, my experience (with a repo in > > Broomfield) is that it's sluggish but faster than NFS would be. > > > My experience working with ON is that I can pull the entirety > of usr and usr/closed directly (SFBay -> Australia) in well > under an hour. Any time I've needed to do the recommit dance > across the Pacific it has taken quite an acceptable length of > time. Then again, I've always been using Mercurial with ssh, > since NFS across the Pacific has always been dog slow. Something > to do with the speed of the carrier pigeon protocol, iirc. I must also point out, however, that I do believe you have come across a problem with how we're doing things. Unfortunately I do not have any suggestion to solve it which has not already been talked about by others. (JBeck, I'm looking at you). James -- Senior Kernel Software Engineer, Solaris Sun Microsystems http://blogs.sun.com/jmcp http://www.jmcp.homeunix.com/blog
