Hi, The better solution -- in the long run -- is that each SAGE worksheet process starts as a different more limited user. Unfortunately, this will require significant work to implement -- we'll likely do it at SD4 next week.
William On 6/6/07, Michel <[EMAIL PROTECTED]> wrote: > > Yep this solution seems to work quite well. My son remarked > that when restarting sage it is necessary to also kill all processes > run by sageuser. Otherwise sageuser could start a process which > would be on the lookout for new instances of sage and kill > these also! > > Michel > > > On Jun 6, 6:40 pm, Michel <[EMAIL PROTECTED]> wrote: > > On Jun 6, 2:04 pm, Michel <[EMAIL PROTECTED]> wrote: > > > > > Input from my son who is fascinated by security. > > > > > On my setup at least the notebook user can > > > kill the sage binary, needing manual intervention > > > to start it again. > > > > > How to guard against that? > > > > > Michel > > > > Well instead of starting > > > > su - sageuser sage -notebook > > > > inside the jail it seems one can run a little script (as root) > > > > while true > > do > > su - sageuser sage -notebook > > sleep 2 > > done > > > > The idea is that from the notebook one can kill the sageuser > > process but not the parent process (which is running as root). > > So the sageuser process is restarted. > > > > Perhaps sage should do this by itself. I.e. sage should have a daemon > > mode in which it drops priviledges. > > > > Michel > > > > > -- William Stein Associate Professor of Mathematics University of Washington http://www.williamstein.org --~--~---------~--~----~------------~-------~--~----~ 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/ -~----------~----~----~----~------~----~------~--~---
