aiju found a bug in port/sysproc.c sysrfork(). in the RFPROC|RFNOMNT case, it sets up->pgrp->noattach instead of p->pgrp->noattach setting the *parents* noattach flag instead of the childs pgrp.
http://code.google.com/p/plan9front/source/detail?r=7bbd45940626e92d4caf11620423a96a3fdc58ad -- cinap
