On 6/6/07, Michel <[EMAIL PROTECTED]> wrote: > > Ok I see. I wasn't aware that one sage process > (your "server process") could serve multiple users. > I will wait to see how things develop. > =============================== > In the setup I was planning it is definitely not true that any user > can kill > the server process. I wanted to have a bunch of users, say > > sageuser1----sageuser10 > > each with their own working directory > and running a copy of sage. So there would be (at most) 10 sage > processes. > > One controlling process (running as root) would monitor these > sage processes and restart (and kill) them as required. > > So sageuser5 would be able to kill sageprocess5 but nothing > else. Moreover sageprocess5 would be restarted immediately. > Everything is fully isolated. No security risk whatsoever. > > One would need a system with a lot of memory of course.
That's an interesting alternative model that I've thought some about. Instead of improving the server we just spawn a bunch of separate servers on separate ports. This is precisely what I did one year ago after I wrote the first version of the notebook and had 30 high school kids try *one* notebook server all at once -- crash. So I made one sage notebook server for each student and this worked fine. I'll have the next group of 30 high school students June 26-July 7, and this year I hope to serve them all with one server. William --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/ -~----------~----~----~----~------~----~------~--~---
