[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Savannah-hackers] Re: savannah.gnu.org + subversions.gnu.org migration
Joel N. Weber II
[Savannah-hackers] Re: savannah.gnu.org + subversions.gnu.org migration complete
Mon, 28 May 2001 02:18:23 -0400
I'm a bit disappointed by the performance I get; if I try to check
a change to sysadmin.texi, I get about a four second delay from
Is it worse now than it used to be with the other machine?
I don't think so. I know that when we moved sysadmin.texi from being
in RCS to being in CVS I noticed a substantially increased delay, and
I don't think it's all that different now that subversions has moved.
There is some packet loss (5% to 10% maybe?) between me and
subversions, that I sent mail to Barton about a couple hours ago.
I'll probably call him at some point, because that's a much more
reliable way to get the message across than email. I don't think it's
unreasonable to expect that we should be able to get 0% packet loss,
but it does require us to get Global NAPS to understand why their
network is a bit saturated and fix it.
Other than that, it's hard to imagine that performance to subversions
could be worse than it was. traceroute tells me subversions is four
hops away. and the new machine is faster.
I think the real point is that remote cvs introduces a somewhat
nontrivial delay, that I find moderately annoying at times, especially
if it causes the emacs that I'm also using for zephyr and mail to hang
while I wait. And at the moment, I don't really understand whether
it's because the computers aren't fast enough, or the net connection
isn't fast enough, or what. (I suspect if we just wait five years and
don't change the software, the problem will go away with hardware and