freebsd-stable Digest, Vol 72, Issue 1
freebsd-sta...@freebsd.org
freebsd...@freebsd.org
To subscribe or unsubscribe via the World Wide Web, visit
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
or, via email, send a message with subject or body 'help' to
freebsd-sta...@freebsd.org
You can reach the person managing the list at
freebsd-st...@freebsd.org
When replying, please edit your Subject line so it is more specific
than "Re: Contents of freebsd-stable digest..."
Today's Topics:
1. two addresses and natd (Dmitry Tolstov)
2. Re: Unable to use a PS/2 keyboard after a boot without it.
(Jesper Wallin)
3. Re: How does CAM determine the order to number drives ?
(Pete French)
4. Re: How does CAM determine the order to number drives ?
(Scott Long)
5. Re: How does CAM determine the order to number drives ?
(Pete French)
6. Re: How does CAM determine the order to number drives ?
(Michael Grant)
7. nfs_lookup doesn't set PDIRUNLOCK? (Brian Buchanan)
8. Re: How does CAM determine the order to number drives ?
(Holger Kipp)
9. Re: Looking for ntp/PPS setup guide (Eric Brown)
10. Re: How does CAM determine the order to number drives ?
(Mike Bristow)
11. Re: Looking for ntp/PPS setup guide (Gregory Bond)
12. Re: Looking for ntp/PPS setup guide (Richard Perini)
13. Re: Unable to use a PS/2 keyboard after a boot without it.
(Marc Santhoff)
14. Re: Looking for ntp/PPS setup guide (Kevin Oberman)
15. Re: XFree86 i830M unresolved symbols (Zoran Kolic)
16. How to successfully mount /root if not in fstab? (Joel Hatton)
----------------------------------------------------------------------
Message: 1
Date: Sun, 1 Aug 2004 17:31:41 +0400
From: Dmitry Tolstov <di...@ibch.ru>
Subject: two addresses and natd
To: freebsd...@freebsd.org
Message-ID: <1851539458.2...@ibch.ru>
Content-Type: text/plain; charset=us-ascii
Hello everyone!
I am running 4.9-RELEASE-p11 and natd. I am planning to get one more
IP and I want it to be redirected to another box in the local network.
Would it possible to run natd as somthing like this:
natd -a <IP1> -p 8668 -redirect_address <localIP> <IP2>;
or run two natd
natd -a <IP1> -p 8668
natd -a <IP2> -p 8778 -redirect_address <localIP> <IP2>
with appropriate ipfw rules;
or it is better to find a switch and up one more interface?
Thank you.
-Dmitry.
------------------------------
Message: 2
Date: Sun, 1 Aug 2004 15:53:01 +0200 (CEST)
From: "Jesper Wallin" <jes...@hackunite.net>
Subject: Re: Unable to use a PS/2 keyboard after a boot without it.
To: "Chris Whitehouse" <ch...@childeric.freeserve.co.uk>
Cc: freebsd...@freebsd.org
Message-ID:
<1169.213.112.193.78...@mail.hackunite.net>
Content-Type: text/plain;charset=iso-8859-1
Heya again..
Once again, thanks for all the replies.. Sure, you can "hotplug" the keyboards as long
as you booted the machine with a keyboard connected.. my point is, if I need to reboot
the server (and no, they're not in range of eachother), I need to get a keyboard,
connect it, ssh back in, reboot it.. then unplug the keyboard when the boot is
complete.. which is kind of annoying and shouldn't be needed imo..
I've removed the "0x1" flags in /boot/device.hints but I havn't had time to restart the
system and find out if it works.. but the last thing which leave me is the hope.. :-P
Regards,
Jesper Wallin
> 01/08/04 00:51:38, "Jesper Wallin" <jes...@hackunite.net> wrote:
>
>>Hello..
>>
>>I run a few FreeBSD machines at home (both 4.10 and 5.2.1) and if I boot them up
>>*without* having the PS/2 keyboard connected, I can't connect it later on.. My servers
>>usually runs without screen and keyboard, power and network cable is the only thing
>>needed. :) But when I need to change something (like, take them down to
>>single-user-mode), then I need to reconnect the keyboard, reboot the machine and THEN I
>>can use the keyboard..
>
> If the machines are within reach of each other a kvm (keyboard video mouse) switch works
> well. You can get a passive switch or a fancy electronic one. The electronic
> one mimics the devices plugged in and allows booting a machine which is not selected.
> With the passive one you must select the machine you want to boot. I have a
> passive one and switch between various operating systems and between AT and PS2 type
> motherboards.
>
> At one place I used to work we booted machines with a keyboard plugged in but otherwise
> moved the few keyboards around between running machines according to
> need.
>
> Chris
>
------------------------------
Message: 3
Date: Sun, 01 Aug 2004 15:09:16 +0100
From: Pete French <petef...@ticketswitch.com>
Subject: Re: How does CAM determine the order to number drives ?
To: m.se...@infracaninophile.co.uk, mg-f...@grant.org
Cc: freebsd...@freebsd.org
Message-ID: <E1BrH1Y-...@dilbert.firstcallgroup.co.uk>
> You can wire down particular devices to a given SCSI bus, target and
> LUN. Thus in your (4.x) kernel config you can say:
>
> device scbus0 at ahc0
> device da0 at scbus0 target 0 unit 0
Thanks for this - it looked as though it was just what I needed. I built
a kernel with scbus0 placed at ahc? (ahc0 would not compile - but I only
have one Adaptec controller in the machine). I would have thought this
was sufficient as it would then scan scbus0 first and thus find the
Addapted conntected drive first. But it did not appear to do this - I
still had the ciss drives appearing first.
I assume for this that the second line is also necessary (which
surprises me), but at this point I had the lid of the machine
off so I simply moved cards around until the Adaptec was scanned
first by the kernel on startup. Which seemed simpler than putting
all the old controllers back and compiling yet another kernel :-)
But thanks for all the advice, (even if I didnt follow it to the letter
in the end). Very useful to know.
-pcf.
PS: For anyone interested in making a similar upgrade, the performance
increase of the 5300 over the 4200 is very impressive. Around 65%
on read performances from RAID-0+1 and 50% from RAID-5. Whether thats
due to the change of BSD driver as well as the change of controller I
don't know, but I am very happy with it!
------------------------------
Message: 4
Date: Sun, 1 Aug 2004 08:49:58 -0600 (MDT)
From: Scott Long <sco...@freebsd.org>
Subject: Re: How does CAM determine the order to number drives ?
To: Pete French <petef...@ticketswitch.com>
Cc: freebsd...@freebsd.org
Message-ID: <2004080108...@pooker.samsco.org>
Content-Type: TEXT/PLAIN; charset=US-ASCII
On Sun, 1 Aug 2004, Pete French wrote:
> > You can wire down particular devices to a given SCSI bus, target and
> > LUN. Thus in your (4.x) kernel config you can say:
> >
> > device scbus0 at ahc0
> > device da0 at scbus0 target 0 unit 0
>
> Thanks for this - it looked as though it was just what I needed. I built
> a kernel with scbus0 placed at ahc? (ahc0 would not compile - but I only
> have one Adaptec controller in the machine). I would have thought this
> was sufficient as it would then scan scbus0 first and thus find the
> Addapted conntected drive first. But it did not appear to do this - I
> still had the ciss drives appearing first.
You need to explicitely define 'ahc0' as so:
device ahc
device ahc0 /* declared for wiring */
device scbus
device scbus0 at ahc0
Wiring down the controller number doesn't usually affect drive ordering.
All drives on all controllers are scanned at once in parallel, and the
first to respond on any controller, regardless of the controller number,
get to be da0.
Scott
------------------------------
Message: 5
Date: Sun, 01 Aug 2004 16:08:04 +0100
From: Pete French <petef...@ticketswitch.com>
Subject: Re: How does CAM determine the order to number drives ?
To: petef...@ticketswitch.com, sco...@freebsd.org
Cc: freebsd...@freebsd.org
Message-ID: <E1BrHwS-...@dilbert.firstcallgroup.co.uk>
> Wiring down the controller number doesn't usually affect drive ordering.
> All drives on all controllers are scanned at once in parallel, and the
> first to respond on any controller, regardless of the controller number,
> get to be da0.
Ah, thats interesting. It explains why my kernel config didnt work as I
had only wired down scbus0 and not the actual drive. Thanks.
*thinks*
But it does beg the question as to why my problem has gone away by moving
the slot the Adaotec controller is in. If all controllers are scanned in
parallel then the pysical positioning in the PCI slots should not make
any difference surely ? Yet it now seems to be booting reliably every time
with the ahc0 drive being found first.
-pcf. [even more puzzled!]
------------------------------
Message: 6
Date: Sun, 1 Aug 2004 11:36:10 -0400
From: Michael Grant <mg-f...@grant.org>
Subject: Re: How does CAM determine the order to number drives ?
To: Scott Long <sco...@freebsd.org>
Cc: freebsd...@freebsd.org
Message-ID: <20040801153...@grant.org>
Content-Type: text/plain; charset=us-ascii
Now I'm a little confused. If you have 2 of the same scsi controller
(say 2 adaptec 2940s), how can you possibly distinguish them apart?
One problem I had once was one of the 2940s died and my second 2940
became my first.
I don't really see how one can fix this sort of thing in the kernel
unless you can specifically identify which controller is which.
I just want to toss this out as an idea: Would it be possible to put
some identifier somewhere on the drive itself when you partition the
drive such that the drive itself could always come up in the same
place regardless even if it moved to a different controller?
Michael Grant
On Sun, Aug 01, 2004 at 08:49:58AM -0600, Scott Long wrote:
> On Sun, 1 Aug 2004, Pete French wrote:
> > > You can wire down particular devices to a given SCSI bus, target and
> > > LUN. Thus in your (4.x) kernel config you can say:
> > >
> > > device scbus0 at ahc0
> > > device da0 at scbus0 target 0 unit 0
> >
> > Thanks for this - it looked as though it was just what I needed. I built
> > a kernel with scbus0 placed at ahc? (ahc0 would not compile - but I only
> > have one Adaptec controller in the machine). I would have thought this
> > was sufficient as it would then scan scbus0 first and thus find the
> > Addapted conntected drive first. But it did not appear to do this - I
> > still had the ciss drives appearing first.
>
> You need to explicitely define 'ahc0' as so:
>
> device ahc
> device ahc0 /* declared for wiring */
> device scbus
> device scbus0 at ahc0
>
> Wiring down the controller number doesn't usually affect drive ordering.
> All drives on all controllers are scanned at once in parallel, and the
> first to respond on any controller, regardless of the controller number,
> get to be da0.
>
> Scott
------------------------------
Message: 7
Date: Sun, 1 Aug 2004 11:29:22 -0700 (PDT)
From: Brian Buchanan <b...@holo.org>
Subject: nfs_lookup doesn't set PDIRUNLOCK?
To: sta...@freebsd.org
Message-ID: <2004080109595...@thought.holo.org>
Content-Type: TEXT/PLAIN; charset=US-ASCII
I think I've tracked down my problem with nullfs layered over NFS causing
deadlocks on 4.10. The nullfs code blindly trusts the lower layer to
correctly set the PDIRUNLOCK flag to know whether it has released the lock
on the parent directory, and NFS doesn't seem to have any code that
manipulates this flag. The result is that nullfs's VOP_LOOKUP doesn't
release the lock on the upper layer.
Everything else that I've looked at (FFS, SMBFS, MSDOSFS, etc.) seems to
play nice with PDIRUNLOCK, though I haven't actually tested with anything
other than FFS and NFS. NFS in -CURRENT also appears to use PDIRUNLOCK.
I thought I was missing something here, wondering "how did NFS ever
work?!", until I discovered that NFS in -STABLE just isn't locked. Oh.
But why do we have PDIRUNLOCK, anyway? Shouldn't everything be following
the locking protocol described in VOP_LOOKUP(9)?
Also, there appears to be a separate logic bug in the -CURRENT nullfs code
(and tjr's nullfs patch) that results in the same deadlock if the lower
layer did not export a lock.
Here's a patch against -STABLE that should make things better. This
is obviously not the right solution, but it is easier than hacking
on NFS.
nullfs in -CURRENT could use the "... dvp->v_vnlock == NULL || ..." part
of the fix, too, to fix the no-exported-lock bug. Of course, everything
in current might be exporting a lock by now.
Index: null_vnops.c
===================================================================
RCS file: /home/ncvs/src/sys/miscfs/nullfs/Attic/null_vnops.c,v
retrieving revision 1.38.2.6
diff -u -r1.38.2.6 null_vnops.c
--- null_vnops.c 31 Jul 2002 00:32:28 -0000 1.38.2.6
+++ null_vnops.c 1 Aug 2004 18:08:49 -0000
@@ -376,17 +376,26 @@
ldvp = NULLVPTOLOWERVP(dvp);
vp = lvp = NULL;
error = VOP_LOOKUP(ldvp, &lvp, cnp);
+
+ /*
+ * We should be relying on PDIRUNLOCK here, but we can't
+ * currently trust the underlying filesystem to set it.
+ * Instead, we'll determine whether the lower vnode should
+ * have been unlocked per VOP_LOOKUP(9).
+ *
+ * Also, only release our lock if it is not shared with
+ * the lower vnode.
+ */
+ if ((error == 0 || error == EJUSTRETURN) &&
+ !((flags & ISLASTCN) && (flags & LOCKPARENT)) &&
+ (dvp->v_vnlock == NULL || dvp->v_vnlock != ldvp->v_vnlock))
+ VOP_UNLOCK(dvp, LK_THISLAYER, p);
+
if (error == EJUSTRETURN && (flags & ISLASTCN) &&
(dvp->v_mount->mnt_flag & MNT_RDONLY) &&
(cnp->cn_nameiop == CREATE || cnp->cn_nameiop == RENAME))
error = EROFS;
- /*
- * Rely only on the PDIRUNLOCK flag which should be carefully
- * tracked by underlying filesystem.
- */
- if (cnp->cn_flags & PDIRUNLOCK)
- VOP_UNLOCK(dvp, LK_THISLAYER, p);
if ((error == 0 || error == EJUSTRETURN) && lvp != NULL) {
if (ldvp == lvp) {
*ap->a_vpp = dvp;
--
Brian Buchanan, CISSP b...@holo.org
--------------------------------------------------------------------------
FreeBSD - The Power to Serve http://www.freebsd.org
------------------------------
Message: 8
Date: Sun, 1 Aug 2004 21:38:06 +0200
From: Holger Kipp <Holge...@alogis.com>
Subject: Re: How does CAM determine the order to number drives ?
To: Michael Grant <mg-f...@grant.org>
Cc: freebsd...@freebsd.org
Message-ID: <2004080121...@intserv.int1.b.intern>
Content-Type: text/plain; charset=us-ascii
On Sun, Aug 01, 2004 at 11:36:10AM -0400, Michael Grant wrote:
> I just want to toss this out as an idea: Would it be possible to put
> some identifier somewhere on the drive itself when you partition the
> drive such that the drive itself could always come up in the same
> place regardless even if it moved to a different controller?
iirc vinum provides something like this :-)
Regards,
Holger Kipp
------------------------------
Message: 9
Date: Sun, 1 Aug 2004 17:02:54 -0400
From: Eric Brown <bro...@locust.cns.vt.edu>
Subject: Re: Looking for ntp/PPS setup guide
To: Kevin Oberman <obe...@es.net>
Cc: FreeBSD...@FreeBSD.org
Message-ID: <20040801210...@locust.cns.vt.edu>
Content-Type: text/plain; charset=us-ascii
On Sat, Jul 31, 2004 at 09:38:04PM -0700, Kevin Oberman wrote:
> > Date: Sat, 31 Jul 2004 21:25:04 -0400 (EDT)
> > From: Garrett Wollman <wol...@khavrinen.lcs.mit.edu>
> >
> > In article <200407312319...@ptavv.es.net> you write:
> >
> > >From what I have seen, the non-kernel PPS software handles jitter more
> > >gracefully than the kernel version.
> >
> > Which CDMA receiver do you have? I'm using one from EndRun
> > Technologies which emulates a Trimble Palisade and it seems to perform
> > fairly well:
> >
> > remote refid st t when poll reach delay offset jitter
> > ==============================================================================
> > *GPS_PALISADE(0) .CDMA. 0 l 1 32 377 0.000 -0.016 0.008
> > +NAVOBS1.MIT.EDU .PSC. 1 u 37 64 377 0.836 0.027 0.025
> > xtime-b.nist.gov .ACTS. 1 u 49 64 377 15.477 -6.898 9.374
> > +ntp2.usno.navy. .USNO. 1 u 10 64 337 40.016 -2.378 83.946
> > -gps.freebsd.dk .GPS. 1 u 51 64 377 115.848 4.315 1.408
> >
> > This receiver was recommended to me by Dave Andersen (dga@).
> > (Actually, I stole it from him.) This is using the host-triggered
> > timestamp mode of this device rather than PPS.
>
> I am running the EndRun Proesis Ct. It can emulate many different clocks
> and, if you don't have PPS, the Palisade is probably the best
> choice. Unlike others which send out the time in ASCII every second, the
> Palisade sends out the time in binary when polled. But this is not as
> accurate as PPS which the unit also provides.
>
> The problem is that polling mode and PPS don't work properly together,
> so I have found that the TrueTime format provides the best results.
> ctime=off
> emul=truetime
> ctime=on
>
> I am still looking at the best choice for PPS setup...PPS_SYNC (flag3 1"
> or software PPL "flag3 1". Both do very well. I suspect that the absolute
> time is closest with PPS_SYNC but the stability is often better with PLL
> discipline. If PLL proves the more stable, I will use a fudge of time1
> to correct for the offset. (In my last message bnl-owamp was running
> PPS_SYNC and the system I queries was running PLL. Note the 51
> microsecond offset. I don't know yet if that's real or an artifact of
> network delays.
> --
> R. Kevin Oberman, Network Engineer
> Energy Sciences Network (ESnet)
> Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
> E-mail: obe...@es.net Phone: +1 510 486-8634
You reminded me of the other item I wanted to try to figure out.
Up until now I have been using PPS as my reference time. I also
use a IRIG input so that the stratum of PPS will be 1. I use a
time offset on the IRIG based on PPS.
It would be nice to find the true offset of the PPS signal given any
interrupt latency through kernel code. I attempted to interface the
PPS output on the parallel port back to my DATUM receiver. The receiver
will gladly measure the event time for me. I couldn't make it work.
I was unable to verify if in fact the PPS output signal was configured.
If it was, then the remaining possibility is that I need some sort of
buffer between PPS output and event input on the receiver.
Any clues?
--Eric Brown
------------------------------
Message: 10
Date: Sun, 1 Aug 2004 22:27:46 +0100
From: Mike Bristow <mi...@urgle.com>
Subject: Re: How does CAM determine the order to number drives ?
To: Michael Grant <mg-f...@grant.org>
Cc: freebsd...@freebsd.org
Message-ID: <20040801212...@urgle.com>
Content-Type: text/plain; charset=us-ascii
On Sun, Aug 01, 2004 at 11:36:10AM -0400, Michael Grant wrote:
> I just want to toss this out as an idea: Would it be possible to put
> some identifier somewhere on the drive itself when you partition the
> drive such that the drive itself could always come up in the same
> place regardless even if it moved to a different controller?
Yes. GEOM in FreeBSD 5.x does this (with geom_vol_ffs). Basically,
devices with a FFS filesystem with a volume name of "foo" appear
in '/dev/vol/foo'.
I use this for my Soekris box; I don't know if an image I'm building
will appear on /dev/ad0s1 or /dev/ad0s2; so I label the filesystem
with a timestamp and stuff /dev/vol/<timestamp> in /etc/fstab. That
way I can dd the image onto either slice. The same principle applies
to disk locations.
--
You dont have to be illiterate to use the Internet, but it help's.
------------------------------
Message: 11
Date: Mon, 02 Aug 2004 10:15:45 +1000
From: Gregory Bond <g...@itga.com.au>
Subject: Re: Looking for ntp/PPS setup guide
To: freebsd...@freebsd.org
Message-ID: <2004080200...@lightning.itga.com.au>
Content-Type: text/plain; charset=us-ascii
Thanks all for the enlightening response. There is one bit I still don't get:
How do I tell the kernel which port to look for the PPS signal on with
PPS_SYNC?
------------------------------
Message: 12
Date: Mon, 2 Aug 2004 11:54:04 +1000
From: Richard Perini <r...@ci.com.au>
Subject: Re: Looking for ntp/PPS setup guide
To: freebsd...@freebsd.org
Message-ID: <2004080211...@mippet.ci.com.au>
Content-Type: text/plain; charset=us-ascii
On Mon, Aug 02, 2004 at 10:15:45AM +1000, Gregory Bond wrote:
>
> Thanks all for the enlightening response. There is one bit I still don't get:
> How do I tell the kernel which port to look for the PPS signal on with
> PPS_SYNC?
Greg,
(FreeBSD-4-Stable)
I use "device pps" in the kernel config file, and a MAKEDEV pps
makes the /dev node. The pps signal is taken to a pin on the parallel
port. (I think the ACK pin, but I'm guessing without pulling the
connector apart which I can't do at the moment).
--
Richard Perini Internet: r...@ci.com.au
Corinthian Engineering Pty Ltd PHONE: +61 2 9906 4333
Sydney, Australia FAX: +61 2 9906 2464
------------------------------
Message: 13
Date: Mon, 02 Aug 2004 04:22:31 +0200
From: M.San...@t-online.de (Marc Santhoff)
Subject: Re: Unable to use a PS/2 keyboard after a boot without it.
To: "FreeBSD stable (Liste)" <freebsd...@freebsd.org>
Message-ID: <1091413350....@zaphod.das.netz>
Content-Type: text/plain
Am So, den 01.08.2004 schrieb Jesper Wallin um 15:53:
> Heya again..
>
> Once again, thanks for all the replies.. Sure, you can "hotplug" the keyboards as long
> as you booted the machine with a keyboard connected..
Be careful, I don't know much about the Hardware of PS/2 port, but I
have killed at least one Keyboard controller by hotpluggin a AT-Keyboard
(out of approx. 10 machines where i did it, before i stopped doing so).
Bye,
Marc
------------------------------
Message: 14
Date: Sun, 01 Aug 2004 22:13:58 -0700
From: "Kevin Oberman" <obe...@es.net>
Subject: Re: Looking for ntp/PPS setup guide
To: Gregory Bond <g...@itga.com.au>
Cc: freebsd...@freebsd.org
Message-ID: <200408020513...@ptavv.es.net>
> From: Gregory Bond <g...@itga.com.au>
> Date: Mon, 02 Aug 2004 10:15:45 +1000
> Sender: owner-free...@freebsd.org
>
>
> Thanks all for the enlightening response. There is one bit I still don't get:
> How do I tell the kernel which port to look for the PPS signal on with
> PPS_SYNC?
You create a softlink from the physical device (parallel or serial).
ln -s /dev/ttyd0 /dev/pps0 (127.127.22.0 in ntp.conf)
or
ln -s /dev/ttyd1 /dev/pps0 (127.127.22.0 in ntp.conf)
or
ln -s /dev/ttyd1 /dev/pps1 (127.127.22.1 in ntp.conf)
or
ln -s /dev/ppc0 /dev/ppc2 (127.127.22.2 in ntp.conf)
--
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: obe...@es.net Phone: +1 510 486-8634
------------------------------
Message: 15
Date: Mon, 2 Aug 2004 08:13:28 +0200
From: Zoran Kolic <kol...@eunet.yu>
Subject: Re: XFree86 i830M unresolved symbols
To: freebsd...@freebsd.org
Message-ID: <2004080206...@kolic.net>
Content-Type: text/plain; charset=us-ascii
> XFree86 Version 4.2.1 / X Window System
> (protocol Version 11, revision 0, vendor release 6600)
> Release Date: 3 September 2002
> If the server is older than 6-12 months, or if your card is
> newer than the above date, look for a newer version before
> reporting problems. (See http://www.XFree86.Org/)
> Build Operating System: FreeBSD 5.0-CURRENT i386 [ELF]
Hi Ethan!
If you've installed 4.10, this
looks strange to me. Will you
chack again your CD and every-
thing? See date and version in
the text!
ZK
------------------------------
Message: 16
Date: Mon, 02 Aug 2004 16:49:03 +1000
From: Joel Hatton <jo...@auscert.org.au>
Subject: How to successfully mount /root if not in fstab?
To: freebsd...@freebsd.org
Message-ID: <200408020649....@app.auscert.org.au>
Hi,
I've been caught out (twice now - yes, I was stupid enough to do this
twice!) with mounting /root after removing a hard disk from my system and
would like to know why it wasn't trivial to recover. I had FreeBSD 4.10
installed on ad1, with ad0 hosting the boot manager and ad1 containing an
msdos boot sector; I removed ad0 from the system and forgot to edit
/etc/fstab before rebooting. With ad1 now ad0, the system booted the kernel
ok, but /root couldn't be found (of course) - at this point all was as
expected. However, when prompted for a mount command, I don't know why
"ufs:/dev/ad0s1a" didn't work - error 6 and the prompt was redisplayed.
To recover, I booted from a fixit CD, mounted /root and edited fstab to
reflect the new arrangement and all was well. However, I discovered another
anomaly while doing this - when I entered the fixit shell I first tried
"fsck -y /dev/ad0s1a" but it returned an error about /etc/fstab. I was
able to "mount -o rw /dev/ad0s1a /mydir", however. Why?
Sorry for the lack of detail, but I'll bet both of these things have been
encountered by many others before...
cheers,
-- Joel Hatton --
Security Analyst | Hotline: +61 7 3365 4417
AusCERT - Australia's national CERT | Fax: +61 7 3365 7031
The University of Queensland | WWW: www.auscert.org.au
Qld 4072 Australia | Email: aus...@auscert.org.au
------------------------------
_______________________________________________
freebsd...@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stabl...@freebsd.org"
End of freebsd-stable Digest, Vol 72, Issue 1
*********************************************