Date: Fri, 13 Jan 2006 02:19:32 -0500 From: Kris Kennaway <kris@obsecurity.org> To: Robert Watson <rwatson@FreeBSD.org> Cc: freebsd-stable@freebsd.org Subject: Re: RELENG_6 devfs problem Message-ID: <20060113071932.GA62526@xor.obsecurity.org> In-Reply-To: <20060111210323.G28748@fledge.watson.org> References: <200601051154.32092.doconnor@gsoft.com.au> <20060111210323.G28748@fledge.watson.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--Nq2Wo0NMKNjxTN9z Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Jan 11, 2006 at 09:06:32PM +0000, Robert Watson wrote: >=20 > On Thu, 5 Jan 2006, Daniel O'Connor wrote: >=20 > >I just installed the latest mgetty and ran it, then found it wasn't=20 > >talking to my modem so I disabled it in /etc/ttys and killed the process. > > > >I ran fstat /dev/cuad0 to check it was dead and fstat got stuck in the= =20 > >devfs state. It seems that everything else was stuck too and now the=20 > >machine is alive but not very functional. > > > >What can I do to try and debug it? >=20 > Do you use devfs rule sets? I fixed a bug a week or so ago in devfs_rule= .c=20 > in HEAD, and MFC'd the fix today. The symptoms are much what you describ= e,=20 > and occur when an invalid rule is proposed -- the right error is returned= ,=20 > but a lock is not released resulting in it being unavailable when process= es=20 > try to access device nodes, resulting in eventual deadlock. It's=20 > devfs_rule.c:1.21 in HEAD, and devfs_rule.c:1.14.2.3 in RELENG_6. I beli= ve=20 > the bug does not exist in RELENG_5, as there were significant changes in= =20 > devfs locking for 6.0, and those haven't been merged to RELENG_5. >=20 > If this doesn't fix it, the normal debugging steps apply -- compile in DD= B,=20 > BREAK_TO_DEBUGGER, and WITNESS. When the wedge occurs, dump the lock sta= te=20 > (show alllocks), and we'll see if we can track down the problem. I don't use rulesets, and I encountered the same problem when using minicom to talk to serial ports. I didn't yet have time to debug it, but I can provide instructions for repeating it. Kris --Nq2Wo0NMKNjxTN9z Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (FreeBSD) iD8DBQFDx1SDWry0BWjoQKURAvR2AJ4h/Lhhhpda6WrVXuSof3wHTZXzGACgry3U V0qtOsHC0J/xQ4L5Q4YFAzY= =ml5h -----END PGP SIGNATURE----- --Nq2Wo0NMKNjxTN9z--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060113071932.GA62526>