Date: Wed, 29 Jul 1998 07:11:22 +0900 From: Tetsuro FURUYA <ht5t-fry@asahi-net.or.jp> To: smarzloff@carif-idf.org Cc: freebsd-stable@FreeBSD.ORG, Tetsuro FURUYA <tfu@ff.iij4u.or.jp> Subject: Re: Disk problem. Message-ID: <199807282211.HAA25371@galois.tf.or.jp> In-Reply-To: Your message of "Mon, 27 Jul 1998 09:54:56 %2B0200" References: <19980727095456.A26476@rafiki.intranet.carif.asso.fr>
next in thread | previous in thread | raw e-mail | index | archive | help
Sorry, because of too long mail header, quoted-printable might not be decoded.
So, resend this mail.
Oh, you have returned.
In Message-ID: <19980727095456.A26476@rafiki.intranet.carif.asso.fr>
Stephane Marzloff <smarzloff@carif-idf.org> wrote:
> Dans un message du 09 Jul à 06:49, Tetsuro FURUYA écrivait :
> >
> >
> > Your ide disk sector is broken.
> > Try
> > bad144 -s -v /dev/wd0
> > or
> > badsect & fsck (This is rather difficult. So, please read man).
>
> Hello..
> I'm back.
>
> So I launch the 'bad144 -s -v wd0, it run about 48H for a 4,8Go.
We have to pay attention to drive assignment, or we will destroy
disk. Block device name is required.
We have to read man bad144(8) carefully.
So it took 48 Hours, but finally bad144 finished.
It seems too long. Something may be abnormal except disk sectors.
>
> This is the report :
>
> [18:22] root@rafiki:~ # bad144 -s -v wd0
> cyl: 621, tracks: 255, secs: 63, sec/cyl: 16065, start: 0, end: 9992178
> 9976365 of 9992178 blocks ( 99%)
> bad block information at sector 9992304 in /dev/rwd0c:
> cartridge serial number: -1108550333(10)
> bt_flag=a70a(16)?
> bad144: /dev/rwd0c: bad flag in bad-sector table
> bad144: /dev/rwd0c: bad magic number
???
> bad144: cyl/trk/sect out of range in existing entry: sn=656226137,
cn=40848, tn=44, sn=245
This seems to be a gabage data on the bad sector information which is
located in the first 5 even numbered sectors of the last track of the
disk pack.
If there exists bad sectors, they are processed before these data.
> bad144: cyl/trk/sect out of range in existing entry: sn=429083853, cn=26709, tn=58, sn=114
............
> bad144: cyl/trk/sect out of range in existing entry: sn=577960943, cn=35976, tn=102, sn=77
> Exit 1
If there are bad sectors, bad144 displays those blocks when executing
access to those blocks, such as,
Block: 1308727 will be marked BAD.
Did you have messages like this ?
Unfortunately, bad144 leaves no data of bad blocks on
/var/log/messages.
So, I get this data doing, ============================
#/usr/local/bin/bash (or /bin/sh)
#bad144 -s -v 2>1& | tee bad144.log
#echo "#include <iostream.h>">cr.cc;echo "using namespace std;\
main(){char cc;while(cin.get(cc)){if(cc==0x0d)cc=0x0a;cout << cc;}}" >>cr.cc;
#g++ cr.cc;./a.out < bad144.log | egrep -i BAD
=======================================================
And, it says "/dev/rwd0c: bad flag in bad-sector table".
It is difficult to know the meaning of this statement.
Does "bad flag" mean that bad-sector table is broken ?
Or, is "bad sector" registered as a flag in bad-sector table ?
If there are bad sectors and the case is the latter,
you can know these are fixed or not by achieving some simple tests.
For example, =====================================
/usr/local/bin/bash (or /bin/sh)
su root
sync
cat /dev/zero >/usr/ttt 2> catzero1.log
sync
cat /usr/ttt > /dev/null 2> catzero2.log
rm /usr/ttt
sync
find /usr -name "peanuts" > /dev/null 2> find.log
sync
egrep -R "peanuts" /usr > /dev/null 2> egrep.log
less /var/log/messages
==================================================
And if wd driver has hung up, bad144 could not end and display
these gabage data.
So, there seems to be no need for a patch to wd.c.
In this case, fdisk do not show whether there is bad sectors or not,
because, there are bad sectors in my disk, but my fdsik shows the same
list.
>
> # fdisk
> ******* Working on device /dev/rwd0 *******
> parameters extracted from in-core disklabel are:
> cylinders=622 heads=255 sectors/track=63 (16065 blks/cyl)
>
> parameters to be used for BIOS calculations are:
> cylinders=622 heads=255 sectors/track=63 (16065 blks/cyl)
>
> Media sector size is 512
> Warning: BIOS sector numbering starts with sector 1
> Information from DOS bootblock is:
> The data for partition 1 is:
> sysid 165,(FreeBSD/NetBSD/386BSD)
> start 63, size 9992367 (4879 Meg), flag 80 (active)
> beg: cyl 0/ sector 1/ head 1;
> end: cyl 621/ sector 63/ head 254
> The data for partition 2 is:
> <UNUSED>
> The data for partition 3 is:
> <UNUSED>
> The data for partition 4 is:
> <UNUSED>
>
> --
> Stéphane Marzloff -> smarzloff@carif-idf.org
> (setq viper-mode t)
>
If you cannot find bad sectors, and the defects continues,
that was not for bad sectors.
Tetsuro Furuya <ht5t-fry@asahi-net.or.jp>
These are my logs. =====================================
#bad144 -s -v wd0 2>1& | tee bad144.log
#/usr/local/bin/bash (or /bin/sh)
#bad144 -s -v 2>1& | tee bad144.log
#echo "#include <iostream.h>">cr.cc;echo "using namespace std;\
main(){char cc;while(cin.get(cc)){if(cc==0x0d)cc=0x0a;cout << cc;}}" >>cr.cc;
#g++ cr.cc;./a.out < bad144.log
cyl: 657, tracks: 64, secs: 63, sec/cyl: 4032, start: 0, end: 2652804
0 of 2652804 blocks ( 0%)
4032 of 2652804 blocks ( 0%)
8064 of 2652804 blocks ( 0%)
......
1278144 of 2652804 blocks ( 48%)
Block: 1278972 will be marked BAD.
Block: 1279172 will be marked BAD.
Block: 1279173 will be marked BAD.
......
Block: 1310765 will be marked BAD.
Too many bad sectors, can only handle 126 per slice.
#egrep -R "peanuts" /usr/ports > /dev/null 2> egrep.log
#cat egrep.log
egrep: /usr/ports/archivers/gshar+gunshar/patches/patch-aa: Input/output error
egrep: /usr/ports/comms/mgetty+sendfax/work/mgetty-1.0.0/tools/g3.o: Input/output error
egrep: /usr/ports/japanese/mew/work/mew-1.54/mew-header.el: Input/output error
egrep: /usr/ports/japanese/mew/work/mew-1.54/Makefile.orig: Input/output error
......
#tail -5 /var/log/messages
Jul 28 22:45:01 dilemma /kernel: wd0s1f: hard error reading fsbn
1215944 of 1215944-1215951 (wd0s1 bn 1342920; cn 333 tn 4 sn 12)wd0:
status 59<rdy,seekdone,drq,err> error 10<no_id>
Jul 28 22:45:20 dilemma /kernel: wd0: interrupt timeout:
Jul 28 22:45:21 dilemma /kernel: wd0: status 58<rdy,seekdone,drq>
error 10<no_id>
Jul 28 22:47:35 dilemma /kernel: wd0s1f: hard error reading fsbn
1215936 of 1215936-1215939 (wd0s1 bn 1342912; cn 333 tn 4 sn 4)wd0:
status 59<rdy,seekdone,drq,err> error 10<no_id>
Jul 28 22:48:17 dilemma /kernel: wd0s1f: hard error reading fsbn
1215940 of 1215940-1215943 (wd0s1 bn 1342916; cn 333 tn 4 sn 8)wd0:
status 59<rdy,seekdone,drq,err> error 40<uncorr>
Tetsuro Furuya <ht5t-fry@asahi-net.or.jp>
========================================================================
TEL: 048-852-3520 FAX: 048-858-1597 ||
E-Mail: 8==------
ht5t-fry@asahi-net.or.jp , tfu@ff.iij4u.or.jp * ||
pgp-fingerprint: \|/
pub Tetsuro FURUYA <ht5t-fry@asahi-net.or.jp>
Key fingerprint = F1 BA 5F C1 C2 48 1D C7 AE 5F 16 ED 12 17 75 38
=========================================================================
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199807282211.HAA25371>