After burning a CD, when trying to read I am getting the following
behaviour:
~> dd if=/dev/cdr of=/dev/null bs=2048
dd: reading `/dev/cdr': Input/output error
176996+0 records in
176996+0 records out
Whereas, the same disk read on the same drive under the normal ide-driver
works ok:
~> dd if=/dev/hdc of=/dev/null bs=2048
177039+0 records in
177039+0 records out
Other disks can be read on this drive also under ide-scsi ok, as well as
this disk can be read on a SCSI DVD-drive fine. Is it a known behaviour,
or a problem in the drive / driver?
2.4.20, drive is Benq CRW 4012A
Thanks
Guennadi
On Tue, 31 Dec 2002, Guennadi Liakhovetski wrote:
This is caused by either the way you burn it, or the inability of cdrecord
to create a CD which doesn't do this. It's related to read-ahead, but I
don't remember the exact detail. In any case, you *may* be able to get rid
of the problem by using -pad in mkisofs, or in cdrecord.
> After burning a CD, when trying to read I am getting the following
> behaviour:
>
> ~> dd if=/dev/cdr of=/dev/null bs=2048
> dd: reading `/dev/cdr': Input/output error
> 176996+0 records in
> 176996+0 records out
>
> Whereas, the same disk read on the same drive under the normal ide-driver
> works ok:
>
> ~> dd if=/dev/hdc of=/dev/null bs=2048
> 177039+0 records in
> 177039+0 records out
>
> Other disks can be read on this drive also under ide-scsi ok, as well as
> this disk can be read on a SCSI DVD-drive fine. Is it a known behaviour,
> or a problem in the drive / driver?
>
> 2.4.20, drive is Benq CRW 4012A
I would say a known behaviour, but one which is only triggered by certain
CDs. Not all CDs (in my experience) do this.
--
bill davidsen <[email protected]>
CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.
On Wed, 2003-01-01 at 15:56, Bill Davidsen wrote:
> On Tue, 31 Dec 2002, Guennadi Liakhovetski wrote:
>
> This is caused by either the way you burn it, or the inability of cdrecord
> to create a CD which doesn't do this. It's related to read-ahead, but I
> don't remember the exact detail. In any case, you *may* be able to get rid
> of the problem by using -pad in mkisofs, or in cdrecord.
Sounds like a scsi error handling funny more than anything else. The end
of disk case for cd-r/cd-rw burned disks are a bit fuzzier than normal
disks. The ide-cd layer knows about this.
Guennadi Liakhovetski wrote:
>
> After burning a CD, when trying to read I am getting the following
> behaviour:
>
> ~> dd if=/dev/cdr of=/dev/null bs=2048
> dd: reading `/dev/cdr': Input/output error
> 176996+0 records in
> 176996+0 records out
I have seen similar problems before I started using the -dao option for
cdrecord.
--
Kasper Dupont -- der bruger for meget tid p? usenet.
For sending spam use mailto:[email protected]
for(_=52;_;(_%5)||(_/=5),(_%5)&&(_-=2))putchar(_);
> > This is caused by either the way you burn it, or the inability of cdrecord
> > to create a CD which doesn't do this. It's related to read-ahead, but I
> > don't remember the exact detail. In any case, you *may* be able to get rid
> > of the problem by using -pad in mkisofs, or in cdrecord.
I used -isosize (copying from another CD), which has an effect similar to
-pad. Just tried with -pad - same. -dao, as suggested by Kasper Dupont
<[email protected]>, doesn't help either. On the contrary, the latter
makes the error appear in all 3 reads - ide-scsi, ide, scsi (I used it
together with -isosize). Also strange, the number of blocks read by dd
bs=2048 on ide and scsi from the same disk differ...
> Sounds like a scsi error handling funny more than anything else. The end
> of disk case for cd-r/cd-rw burned disks are a bit fuzzier than normal
> disks. The ide-cd layer knows about this.
But, as I described, a native SCSI-CD/DVD-drive reads this disk without
problems.
Thanks
---
Guennadi Liakhovetski
* Guennadi Liakhovetski writes:
>I used -isosize (copying from another CD), which has an effect similar
>to -pad. Just tried with -pad - same. -dao, as suggested by Kasper
>Dupont <[email protected]>, doesn't help either. On the contrary, the
>latter makes the error appear in all 3 reads - ide-scsi, ide, scsi (I
>used it together with -isosize). Also strange, the number of blocks
>read by dd bs=2048 on ide and scsi from the same disk differ...
Try -raw. Some drives (e.g. my acer 2010A) have - according to Joerg
Schilling - broken firmwares, that have problems in SAO mode.
regards
Markus