[Planetlab-users] TCP port 4662
christin at SIMS.Berkeley.EDU
Mon Jul 11 12:37:25 EDT 2005
Thanks Neil, this seems to be working.
On Mon Jul 11, 2005, Neil Spring <nspring at cs.umd.edu> wrote:
> To find the "culprit",
> ssh <host you're having trouble with>
> grep 4662 /proc/scout/ports/summary
> curl -s http://localhost:3120 | grep <context id that has bound the
> mail <slicename>@slices.planet-lab.org
> at least, I think this is right.
> I don't know how to go from context id to name without using cotop...
> hints welcome.
> Note: an entry in the wiki is as binding as calling "shotgun" before
> entering a car.
> On Jul 10, 2005, at 7:43 PM, Nicolas Christin wrote:
> >I had reserved TCP port 4662 for some experiments with my
> >slice (ucb_p2pecon). It appears that, on some nodes at least,
> >(e.g., planetlab01.ethz.ch, planet-lab-1.csse.monash.edu.au,
> >planetlab1.millennium.berkeley.edu), 4662 is already bound (and not by
> >me). This seems to be a fairly new development as I ran some
> >without any problem last week.
> >I am not sure there is a way for me of seeing which process is using
> >4662, so if you're reading this, and are using 4662, please do
> >check the
> >wiki before binding...
> >For information, 4662 is the edonkey port -- and I am indeed using it
> >to crawl the ed2k network, as a follow up to the P2P
> >measurement project I have been involved in.
> >(See: http://p2pecon.berkeley.edu/pub/CWC-EC05.pdf)
> >I am not sure what the effects of switching to a different
> >"non-standard" port will be on my measurements. I'll use 14462
> >temporarily, and will accordingly reserve the port on the wiki, but I
> >*really* would like to see 4662 available if at all possible.
> >Users mailing list: Users at lists.planet-lab.org
> Users mailing list: Users at lists.planet-lab.org
More information about the Users