1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
|
.TH DISKSCAN 1 2013-10-25 "DiskScan @PACKAGE_VERSION@" "User Commands"
.ad l
.nh
.SH NAME
diskscan - scan a disk for failed and near failure sectors
.SH SYNOPSIS
\fBdiskscan\fR [options...] \fIblock_device\fR
.SH DESCRIPTION
\fBdiskscan\fR is intended to check a disk and find any bad sectors already present
and assess it for any possible sectors that are in the process of going bad.
The operation is all read-only and can cause no direct damage to the data on
the disk.
.PP
diskscan reads the entire block device and notes the time it took to read a
block. When there is an error it is immediately noted and also when there is a
higher latency to read a block. A histogram of the block latency times is also
given to assess the health of the disk.
.PP
The output of diskscan will show any serious errors or very high latency and
will also emit an histogram at the end of the run in the form:
.RS +4n
.nf
.PP
I: Validating path /dev/sdg
I: Opened disk /dev/sdg
I: Scanning disk /dev/sdg
Access time histogram:
1: 0
10: 0
100: 0
500: 120
1000: 0
2000: 1
3000: 0
4000: 0
5000: 0
6000: 0
7000: 0
8000: 0
9000: 0
10000: 0
15000: 0
20000: 0
25000: 0
30000: 0
above that: 0
I: Closed disk /dev/sdg
.fi
.RE
.PP
This means that all I/Os in this case were between 100 and 600 msec and there
were 120 chunks being read. Current these chunks are 1MB in size.
.SH OPTIONS
\fB-v\fR, \fB--verbose\fR
display verbose information from the workings of the scan
use multiple times for increased verbosity.
.PP
\fB-f\fR, \fB--fix\fR
Attempt to fix areas that are nearing failure. This should only be
attempted on an unmounted block device and never on an inuse filesystem or
corruption is likely.
.PP
\fB-s <mode>\fR, \fB--scan <mode>\fR
Scan mode can be either \fBseq\fR or \fBrandom\fR, random reduces the chance that the
disk will be able spend time to recover data before we try to access a sector
but the seeks add noise to the latency measurement. Sequential test is the
default and random test is still experimental with regard to its usefulness.
.PP
\fB-e <size>\fR, \fB--size <size>\fR
Set the size in which the scan will be done, this must be a multiple of the sector size
which is normally 512 bytes.
.PP
\fB-o <file>\fR, \fB--output <file>\fR
Set the output file that the scan will generate. This is a JSON file with the
summary and details about the exceptional events found during the scan.
.PP
\fB-r <file>\fR, \fB--raw-log <file>\fR
Set the output file for the raw log which logs everything done and seen during
the scan. This is a rather large file but it can help get the finer details of
the scan progress and the disk behavior during the scan. This is too a JSON file.
.SH "SEE ALSO"
\fBbadblocks\fR(1), \fBfsck\fR(1)
.SH AUTHOR
Baruch Even <baruch@ev-en.org>
|
