lkml.org 
[lkml]   [1999]   [Oct]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: access beyond end of device errors in 2.2.13pre18 AND 2.2.5
Mark Hagger wrote:

> On Thu, 21 Oct 1999, Tommy van Leeuwen wrote:
> > 6 identical machines and only 1 gives the same errors. I think (and i'm
> > pretty sure) that it's caused by reusing inodes or such (i'm not that
> > much into source).
> >
> > Try newfs'ing a disk. No matter how hard you try, the corruption will only
> > occure once the disk is 'used'. I've never seen this corruption on a new
> > disk, only on heavy used disks (i.e. news and mail data).
>
> And Manfred wrote:
> > Could you post the complete error message from 2.2.13pre18?There should be a
> > third line below, and the exact numbers might be intresting.
>
> As it happens I have just re-formated all the disks that were being somewhat
> thrashed (as I have just put redhat 6.1 on our system and decided to start
> from scratch, and thanks to redhat for their interesting NIS (ypserv) server
> RPM). So far I have not seen any access errors, but I will continue to hit the
> system hard.
>
> Prior to reformat the disks were very heavily fragmented, due to considerable
> hammering over the last 9 months or so, some up to around 20% fragmented as
> well (on a 16 gig partition on a 18 gig drive). Typically large files were
> created during a run and then deleted completely.
>
> So perhaps the "re-used inodes" is a factor.
>
> Here is some interesting log file entries from the pre-reinstall:
>
> Oct 21 19:22:57 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:22:57 nc10 kernel: 08:0d: rw=0, want=1029576352, limit=15639246
> Oct 21 19:22:57 nc10 kernel: dev 08:0d blksize=1024 blocknr=1029576351 sector=2059152702 size=10
> 24 count=1
> Oct 21 19:22:58 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:22:58 nc10 kernel: 08:0d: rw=0, want=1073509243, limit=15639246
> Oct 21 19:22:58 nc10 kernel: dev 08:0d blksize=1024 blocknr=-1073974406 sector=2147018484 size=1
> 024 count=1
> Oct 21 19:22:58 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:22:58 nc10 kernel: 08:0d: rw=0, want=1029576352, limit=15639246
> Oct 21 19:22:58 nc10 kernel: dev 08:0d blksize=1024 blocknr=1029576351 sector=2059152702 size=10
> 24 count=1
> Oct 21 19:22:58 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:22:58 nc10 kernel: 08:0d: rw=0, want=1073509243, limit=15639246
> Oct 21 19:22:58 nc10 kernel: dev 08:0d blksize=1024 blocknr=-1073974406 sector=2147018484 size=1
> 024 count=1
> Oct 21 19:22:58 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:22:58 nc10 kernel: 08:0d: rw=0, want=1029576352, limit=15639246
> Oct 21 19:22:58 nc10 kernel: dev 08:0d blksize=1024 blocknr=1029576351 sector=2059152702 size=10
> 24 count=1
> Oct 21 19:22:58 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:22:58 nc10 kernel: 08:0d: rw=0, want=1073509243, limit=15639246
> Oct 21 19:22:58 nc10 kernel: dev 08:0d blksize=1024 blocknr=-1073974406 sector=2147018484 size=1
> 024 count=1
> Oct 21 19:22:58 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:22:58 nc10 kernel: 08:0d: rw=0, want=1029576352, limit=15639246
> Oct 21 19:22:58 nc10 kernel: dev 08:0d blksize=1024 blocknr=1029576351 sector=2059152702 size=10
> 24 count=1
> Oct 21 19:22:59 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:22:59 nc10 kernel: 08:0d: rw=0, want=1073509243, limit=15639246
> Oct 21 19:22:59 nc10 kernel: dev 08:0d blksize=1024 blocknr=-1073974406 sector=2147018484 size=1
> 024 count=1
> Oct 21 19:22:59 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:22:59 nc10 kernel: 08:0d: rw=0, want=1029576352, limit=15639246
> Oct 21 19:22:59 nc10 kernel: dev 08:0d blksize=1024 blocknr=1029576351 sector=2059152702 size=10
> 24 count=1
> Oct 21 19:22:59 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:22:59 nc10 kernel: 08:0d: rw=0, want=1073509243, limit=15639246
> Oct 21 19:22:59 nc10 kernel: dev 08:0d blksize=1024 blocknr=-1073974406 sector=2147018484 size=1
> 024 count=1
> Oct 21 19:22:59 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:22:59 nc10 kernel: 08:0d: rw=0, want=1029576352, limit=15639246
> Oct 21 19:22:59 nc10 kernel: dev 08:0d blksize=1024 blocknr=1029576351 sector=2059152702 size=10
> 24 count=1
> Oct 21 19:23:00 nc10 kernel: attempt to access beyond end of device
> Oct 21 19:23:00 nc10 kernel: 08:0d: rw=0, want=1073509243, limit=15639246
> Oct 21 19:23:00 nc10 kernel: dev 08:0d blksize=1024 blocknr=-1073974406 sector=2147018484 size=1
> 024 count=1
>

Turning OFF DMA on my WDC __IMMEDIATELY__ removed this problem (attempting to access beyond end of
device). However, I now suffer from slower disk access. I'm using 2.2.6 - 2.2.12. Never had this
problem with 2.0.3x.

/dev/hda:

Model=WDC AC28400R, FwRev=17.01J17, SerialNo=WD-WM6281727434
Config={ HardSect NotMFM HdSw>15uSec SpinMotCtl Fixed DTR>5Mbs FmtGapReq }
RawCHS=16383/16/63, TrkSize=57600, SectSize=600, ECCbytes=40
BuffType=3(DualPortCache), BuffSize=512kB, MaxMultSect=16, MultSect=off
DblWordIO=no, maxPIO=2(fast), DMA=yes, maxDMA=0(slow)
CurCHS=1027/255/63, CurSects=16514064, LBA=yes
LBA CHS=513/510/63 Remapping, LBA=yes, LBAsects=16514064
tDMA={min:120,rec:120}, DMA modes: mword0 mword1 *mword2
IORDY=on/off, tPIO={min:160,w/IORDY:120}, PIO modes: mode3 mode4
UDMA modes: mode0 mode1 mode2


/dev/hda:
multcount = 0 (off)
I/O support = 0 (default 16-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 1027/255/63, sectors = 16514064, start = 0


--
------------------------------
Jim A. Horning
jhorning@nps.navy.mil
Space Systems Academic Group
U.S. Naval Postgraduate School



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:54    [W:0.051 / U:0.032 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site