Subject: corruption on compression, or bitrot?
To: NetBSD Current Users <current-users@netbsd.org>
From: The Grey Wolf <greywolf@starwolf.com>
List: current-users
Date: 09/09/2004 17:10:03
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hey, not sure where to send this.

here's my situation.

I have a large bunch of data -- about 2.3G worth.
I tar it up.  OK.
I gzip it.  OK.
I gzip -d it.  No go:  crc error.

Fine.  Split it into 64MB chunks using dd if=file of=file.$n bs=1024k \
	count=64 skip=`expr $n \* 64`.  OK.
Gzip each separate one using gzip -cv1 $file.$n > $file.$n.gz.  OK.
As a sanity check, run gzip -dvc $file.$n.gz > /dev/null.  OK.
Go back at the end of the run and try again.  *BOOM*  First one gets a
crc error.

Is this filesystem error, disk error/bitrot or what?  What would cause
the corruption to occur after the file had already been checked for
integrity?

Filesystem in question is NOT mounted with softdeps.
I have no errors via dmesg with regard to the disk itself, nor to the
mounted partition.

Disk is:
	wd1 at atabus0 drive 1: <ST320420A>
	wd1: drive supports 16-sector PIO transfers, LBA addressing
	wd1: 19458 MB, 39535 cyl, 16 head, 63 sec, 512 bytes/sect x \
		39851760 sectors
	wd1: 32-bit data port
	wd1: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 4 (Ultra/66)
	wd1(viaide0:0:1): using PIO mode 4, Ultra-DMA mode 4 (Ultra/66) \
		(using DMA data transfers)

wd0, wd2 and wd3 are not currently in use.

All are connected to onboard controllers.

rivendell 5# atactl wd1 smart status
SMART supported, SMART status unknown
id value thresh crit collect reliability description			raw
  1  66   25     no  online  positive    Raw read error rate     	12195457
  3  70    0     no  online  positive    Spin-up time            	0
  4 100   20     yes online  positive    Start/stop count        	55
  5 100   36     yes online  positive    Reallocated sector count	4
  7  60   30     yes online  positive    Seek error rate         	610007117151
  9  78    0     no  online  positive    Power-on hours count    	19294
 10 100   97     yes online  positive    Spin retry count        	0
 12 100   20     yes online  positive    Device power cycle count	417
194  53    0     no  online  positive    Temperature             	56
195  74    0     no  online  positive    Hardware ECC Recovered  	79324759
197 100    0     no  online  positive    Current pending sector  	0
198 100    0     no  offline positive    Offline uncorrectable   	0
199 200    0     no  online  positive    Ultra DMA CRC error count	3
200  95    0     no  offline positive    Unknown                 	67436544
202 100    0     no  online  positive    Unknown                 	0
rivendell 6#

				--*greywolf;
- --
Support Open Operating Systems -- subvert the Microsoft paradigm.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (NetBSD)

iD8DBQFBQPDfDOGizqAnVRERAny7AKCNog4fLM+IgqyNmPKToEChmtJ3+wCfXd4f
Qv9PoV+LkE0atX5LwtpEEBY=
=fwU7
-----END PGP SIGNATURE-----