Date: Wed, 15 Nov 1995 05:22:26 -0800 From: David Greenman <davidg@Root.COM> To: jan@todonix.ping.de (Jan Wedekind) Cc: freebsd-bugs@freebsd.org Subject: Re: panic: nfsreq nogrps Message-ID: <199511151322.FAA00623@corbin.Root.COM> In-Reply-To: Your message of "Wed, 15 Nov 95 14:14:59 %2B0100." <m0tFhg3-0006zZC@todonix.ping.de>
next in thread | previous in thread | raw e-mail | index | archive | help
>> It's usually caused by Smail. Smail does a setgroups(0,foo) which isn't >> valid (you have to have at least one group in the list - the effective gid). >> This was fixed recently in the version of Smail in the ports tree. The bug in >> the kernel was fixed in -current just recently. The kernel fix is to disallow >> ngroups=0. >> >Thanks for the hint, but on that machine (with the panic) only >sendmail is running. > >On the other machine (mounted by nfs) smail is running, but >no directory-tree concerning mail configuring or delivery >is mounted by the other on. Then something else must be doing the setgroups() call. The kernel fix is attached. -DG Index: kern_prot.c =================================================================== RCS file: /home/ncvs/src/sys/kern/kern_prot.c,v retrieving revision 1.13 retrieving revision 1.14 diff -c -r1.13 -r1.14 *** 1.13 1995/10/08 00:06:07 --- 1.14 1995/11/04 10:50:55 *************** *** 407,413 **** if ((error = suser(pc->pc_ucred, &p->p_acflag))) return (error); ! if ((ngrp = uap->gidsetsize) > NGROUPS) return (EINVAL); pc->pc_ucred = crcopy(pc->pc_ucred); if ((error = copyin((caddr_t)uap->gidset, --- 407,414 ---- if ((error = suser(pc->pc_ucred, &p->p_acflag))) return (error); ! ngrp = uap->gidsetsize; ! if (ngrp < 1 || ngrp > NGROUPS) return (EINVAL); pc->pc_ucred = crcopy(pc->pc_ucred); if ((error = copyin((caddr_t)uap->gidset,
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199511151322.FAA00623>