1TAPESTAT(1) Linux User's Manual TAPESTAT(1)
2
6 tapestat - Report tape statistics.
7
9 tapestat [ -k | -m ] [ -t ] [ -V ] [ -y ] [ -z ] [ --human ] [ interval
10 [ count ] ]
11
13 The tapestat command is used for monitoring the activity of tape drives
14 connected to a system.
15 The first report generated by the tapestat command provides statistics
17 concerning the time since the system was booted, unless the -y option
18 is used, when this first report is omitted. Each subsequent report
19 covers the time since the previous report.
20 The interval parameter specifies the amount of time in seconds between
22 each report. The count parameter can be specified in conjunction with
23 the interval parameter. If the count parameter is specified, the value
24 of count determines the number of reports generated at interval seconds
25 apart. If the interval parameter is specified without the count parame‐
26 ter, the tapestat command generates reports continuously.
27
30 The tapestat report provides statistics for each tape drive connected
31 to the system. The following data are displayed:
32 r/s
34 The number of reads issued expressed as the number per second
35 averaged over the interval.
36 w/s
38 The number of writes issued expressed as the number per second
39 averaged over the interval.
40 kB_read/s | MB_read/s
42 The amount of data read expressed in kilobytes (by default or if
43 option -k used) or megabytes (if option -m used) per second
44 averaged over the interval.
45 kB_wrtn/s | MB_wrtn/s
47 The amount of data written expressed in kilobytes (by default or
48 if option -k used) or megabytes (if option -m used) per second
49 averaged over the interval.
50 %Rd
52 Read percentage wait - The percentage of time over the interval
53 spent waiting for read requests to complete. The time is mea‐
54 sured from when the request is dispatched to the SCSI mid-layer
55 until it signals that it completed.
56 %Wr
58 Write percentage wait - The percentage of time over the interval
59 spent waiting for write requests to complete. The time is mea‐
60 sured from when the request is dispatched to the SCSI mid-layer
61 until it signals that it completed.
62 %Oa
64 Overall percentage wait - The percentage of time over the inter‐
65 val spent waiting for any I/O request to complete (read, write,
66 and other).
67 Rs/s
69 The number of I/Os, expressed as the number per second averaged
70 over the interval, where a non-zero residual value was encoun‐
71 tered.
72 Ot/s
74 The number of I/Os, expressed as the number per second averaged
75 over the interval, that were included as "other". Other I/O
76 includes ioctl calls made to the tape driver and implicit opera‐
77 tions performed by the tape driver such as rewind on close (for
78 tape devices that implement rewind on close). It does not
79 include any I/O performed using methods outside of the tape
80 driver (e.g. via sg ioctls).
81
83 --human
84 Print sizes in human readable format (e.g. 1.0k, 1.2M, etc.)
85 The units displayed with this option supersede any other default
86 units (e.g. kilobytes, sectors...) associated with the metrics.
87 -k Show the amount of data written or read in kilobytes per second
89 instead of megabytes. This option is mutually exclusive with
90 -m.
91 -m Show the amount of data written or read in megabytes per second
93 instead of kilobytes. This option is mutually exclusive with
94 -k.
95 -t Display time stamps. The time stamp format may depend on the
97 value of the S_TIME_FORMAT environment variable (see below).
98 -V Print version and exit.
100 -y Omit the initial statistic showing values since boot.
102 -z Tell tapestat to omit output for any tapes for which there was
104 no activity during the sample period.
105
108 It is possible for a percentage value (read, write, or other) to be
109 greater than 100 percent (the tapestat command will never show a per‐
110 centage value more than 999). If rewinding a tape takes 40 seconds
111 where the interval time is 5 seconds the %Oa value would show as 0 in
112 the intervals before the rewind completed and then show as approxi‐
113 mately 800 percent when the rewind completes.
114 Similar values will be observed for %Rd and %Wr if a tape drive stops
116 reading or writing and then restarts (that is it stopped streaming). In
117 such a case you may see the r/s or w/s drop to zero and the %Rd/%Wr
118 value could be higher than 100 when reading or writing continues
119 (depending on how long it takes to restart writing or reading). This
120 is only an issue if it happens a lot as it may cause tape wear and will
121 impact on the backup times.
122 For fast tape drives you may see low percentage wait times. This does
124 not indicate an issue with the tape drive. For a slower tape drive
125 (e.g. an older generation DDS drive) the speed of the tape (and tape
126 drive) is much slower than filesystem I/O, percent wait times are
127 likely to be higher. For faster tape drives (e.g. LTO) the percentage
128 wait times are likely to be lower as program writing to or reading from
129 tape is going to be doing a lot more filesystem I/O because of the
130 higher throughput.
131 Although tape statistics are implemented in the kernel using atomic
133 variables they cannot be read atomically as a group. All of the statis‐
134 tics values are read from different files under /sys, because of this
135 there may be I/O completions while reading the different files for the
136 one tape drive. This may result in a set of statistics for a device
137 that contain some values before an I/O completed and some after.
138 This command uses rounding down as the rounding method when calculating
140 per second statistics. If, for example, you are using dd to copy one
141 tape to another and running tapestat with an interval of 5 seconds and
142 over the interval there were 3210 writes and 3209 reads then w/s would
143 show 642 and r/s 641 (641.8 rounded down to 641). In such a case if it
144 was a tar archive being copied (with a 10k block size) you would also
145 see a difference between the kB_read/s and kB_wrtn/s of 2 (one I/O 10k
146 in size divided by the interval period of 5 seconds). If instead there
147 were 3210 writes and 3211 reads both w/s and r/s would both show 642
148 but you would still see a difference between the kB_read/s and
149 kB_wrtn/s values of 2 kB/s.
150 This command is provided with an interval in seconds. However inter‐
152 nally the interval is tracked per device and can potentially have an
153 effect on the per second statistics reported. The time each set of
154 statistics is captured is kept with those statistics. The difference
155 between the current and previous time is converted to milliseconds for
156 use in calculations. We can look at how this can impact the statistics
157 reported if we use an example of a tar archive being copied between two
158 tape drives using dd. If both devices reported 28900 kilobytes trans‐
159 ferred and the reading tape drive had an interval of 5001 milliseconds
160 and the writing tape drive 5000 milliseconds that would calculate out
161 as 5778 kB_read/s and 5780 kB_wrtn/s.
162 The impact of some retrieving statistics during an I/O completion,
164 rounding down, and small differences in the interval period on the sta‐
165 tistics calculated should be minimal but may be non-zero.
166
168 The tapestat command takes into account the following environment vari‐
169 ables:
170 S_COLORS
173 When this variable is set, display statistics in color on the
174 terminal. Possible values for this variable are never, always
175 or auto (the latter is the default).
176 Please note that the color (being red, yellow, or some other
178 color) used to display a value is not indicative of any kind of
179 issue simply because of the color. It only indicates different
180 ranges of values.
181 S_COLORS_SGR
184 Specify the colors and other attributes used to display statis‐
185 tics on the terminal. Its value is a colon-separated list of
186 capabilities that defaults to
187 H=31;1:I=32;22:M=35;1:N=34;1:Z=34;22. Supported capabilities
188 are:
189 H= SGR (Select Graphic Rendition) substring for percentage
192 values greater than or equal to 75%.
193 I= SGR substring for tape names.
196 M= SGR substring for percentage values in the range from 50%
199 to 75%.
200 N= SGR substring for non-zero statistics values.
203 Z= SGR substring for zero values.
206 S_TIME_FORMAT
209 If this variable exists and its value is ISO then the current
210 locale will be ignored when printing the date in the report
211 header. The tapestat command will use the ISO 8601 format (YYYY-
212 MM-DD) instead. The timestamp displayed with option -t will
213 also be compliant with ISO 8601 format.
214
217 /sys filesystem must be mounted for tapestat to work. It will not work
218 on kernels that do not have sysfs support
219 This command requires kernel version 4.2 or later (or tape statistics
221 support backported for an earlier kernel version).
222
225 /sys/class/scsi_tape/st<num>/stats/* Statistics files for tape devices.
226 /proc/uptime contains system uptime.
228
230 Initial revision by Shane M. SEYMOUR (shane.seymour <at> hpe.com)
231 Modified for sysstat by Sebastien Godard (sysstat <at> orange.fr)
232
234 iostat(1), mpstat(1)
235 http://pagesperso-orange.fr/sebastien.godard/
237Linux JANUARY 2018 TAPESTAT(1)