Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 26 Oct 1998 04:23:38 +0300
From:      "Mikhail A. Sokolov" <mishania@demos.net>
To:        freebsd-current@FreeBSD.ORG
Subject:   fsck inoperability/CAM?
Message-ID:  <19981026042338.40761@demos.su>

next in thread | raw e-mail | index | archive | help
Hello,

when doing fsck -y (since -p fails) on just rebuilt RAID-5 (due to disk
loss), I come out to way strange results.. The FS is obviously broken, but
I never thought it will come up that way with RAID-5, and even then,
why would it break fsck?

Actually, the OS is 2 days old, ELF -current with aout kernel, the device
is DPT-4333-UW with 5 seagate barracuda's 19171 RAID-5. The loss of the FS
is probably caused by cp -R from other disks, after which one of the discs
was announced to not to be able to work by RAID, which is itself scarry, -
the system sometimes freezes tight when more or less big disk activity 
(like, make world) starts. I've pulled these and other disks out one by one
and checked, all of those seem to be ok according to multi-verifications on
low level, other machines and so on. Different controllers were used as
well, so, that might be software problem. One time I was lucky to find 
console shouting about:

--
infinete wait buffer: device 0x20401
SEYQADDR = 0x9
STAT1 = 0xa
da2:(ahc0:0:2:0) Bus device RESET message sent
--
Unfortunately, I can't reproduce the whole messages.

CAM and SMP is enabled, no softupdates.


the fsck part of my message:

Program terminated with signal 11, Segmentation fault.
#0  0x8049b68 in iblock ()
(gdb) where
#0  0x8049b68 in iblock ()
#1  0x804980e in ckinode ()
#2  0x804a301 in clri ()
#3  0x804cc2d in pass4 ()
#4  0x804ac7b in checkfilesys ()
#5  0x804a9e1 in main ()
#6  0x80480c9 in _start ()


UNKNOWN FILE TYPE I=367275
CLEAR? yes
137699129 BAD I=367276
571373860 BAD I=367276
-618920435 BAD I=367276
EXCESSIVE BAD BLKS I=367276
CONTINUE? yes
INCORRECT BLOCK COUNT I=367276 (-659171574 should be 80)
CORRECT? yes
UNKNOWN FILE TYPE I=367277
CLEAR? yes
56583624 BAD I=367278
-2131332271 BAD I=367278
-303071672 BAD I=367278
-1759035936 BAD I=367300
CONTINUE? yes
CLEAR? yes
UNKNOWN FILE TYPE I=367314
CLEAR? yes
INCORRECT BLOCK COUNT I=367334 (582170220 should be 80)
CORRECT? yes
HOLD BAD BLOCK? yes
UNKNOWN FILE TYPE I=367335
CLEAR? yes
UNKNOWN FILE TYPE I=367336
CLEAR? yes
PARTIALLY ALLOCATED INODE I=367337
CLEAR? yes
....
DUP/BAD  I=250882  OWNER=ms MODE=100644
SIZE=1359981 MTIME=Apr  1 00:00 1997
FILE=/local/vh/ftp.demos.net/pub/ripe/dbase/others/radb.db.gz
...
nice mtime:
BAD/DUP FILE I=366762  OWNER=620986308 MODE=14133
SIZE=15034536333323123082 MTIME=Jan 20 23:14 1904
CLEAR? yes
BAD/DUP FILE I=366766  OWNER=3356080835 MODE=142347
SIZE=6917460694118870189 MTIME=Nov 25 03:32 1956
CLEAR? yes
BAD/DUP FILE I=366768  OWNER=3727934015 MODE=121546
SIZE=2981631057524013585 MTIME=Dec  4 09:39 2012
CLEAR? yes
BAD/DUP FILE I=366777  OWNER=325407338 MODE=126433
SIZE=4111564472578667750 MTIME=Feb  3 13:00 1974
CLEAR? yes
BAD/DUP FILE I=366779  OWNER=611695138 MODE=140257
SIZE=11195809982008702958 MTIME=Feb 13 13:51 1967
CLEAR? yes
BAD/DUP FILE I=366781  OWNER=119711167 MODE=144570
SIZE=5364355988561614553 MTIME=Jan 29 09:53 2029
CLEAR? yes
BAD/DUP FILE I=366783  OWNER=2012960013 MODE=15317
SIZE=7763439426879057780 MTIME=May  3 04:45 1982
CLEAR? yes
BAD/DUP FILE I=366791  OWNER=4058534106 MODE=103417
SIZE=15639337409666208743 MTIME=Mar  6 19:23 1965
CLEAR? yes
BAD/DUP FILE I=366806  OWNER=4235278789 MODE=12543
SIZE=5585102475081859373 MTIME=Sep  9 13:56 1917
CLEAR? yes
BAD/DUP FILE I=366807  OWNER=2295127275 MODE=126642
BAD/DUP FILE I=366969  OWNER=359404662 MODE=142653
SIZE=5489296613212808044 MTIME=Oct 14 13:17 2029
CLEAR? yes

Segmentation fault
{skraldespand}/home/mishania#

Thanks for hints,

-- 
-mishania

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19981026042338.40761>