1PCP-TAPESTAT(1) General Commands Manual PCP-TAPESTAT(1)
2
6 pcp-tapestat - report tape I/O statistics
7
9 pcp [pcp options] tapestat [-u] [-G method] [-P precision] [-R pattern]
10 [-x [t][,h][,noidle]]
11
13 pcp-tapestat reports I/O statistics for tape devices.
14
16 When invoked via the pcp(1) command, the -a/--archive, -h/--host,
17 -O/--origin, -s/--samples, -t/--interval, -Z/--timezone and several
18 other pcp options become indirectly available; refer to PCPIntro(1) for
19 a complete description of these options.
20 The additional command line options available for pcp-tapestat are:
22 -G method, --aggregate=method
24 Specifies that statistics for device names matching the regular
25 expression specified with the -R regex option should be aggregated
26 according to method. Note this is aggregation based on matching
27 device names (not temporal aggregation). When -G is used, the
28 device name column is reported as method(regex), e.g. if -G sum
29 -R 'st(0|1)$' is specified, the device column will be
30 sum(st(0|1)$) and summed statistics for st0 and st1 will be
31 reported in the remaining columns. If -G is specified but -R is
32 not specified, then the default regex is .*, i.e. matching all
33 device names. If method is sum then the statistics are summed.
34 If method is avg then the statistics are summed and then averaged
35 by dividing by the number of matching device names. If method is
36 min or max, the minimum or maximum statistics for matching devices
37 are reported, respectively.
38 -P N, --precision=N
40 This indicates the precision (number of decimal places) to report.
41 The default precision N may be set to something other than the
42 default (2). Note that the avgrq-sz and avgqu-sz fields are
43 always reported with N+1 decimals of precision. These fields typ‐
44 ically have values less than 1.
45 -R pattern, --regex=pattern
47 This restricts the report to device names matching a regular
48 expression pattern. The given pattern is searched as a perl style
49 regular expression, and will match any portion of a device name.
50 e.g. '^st[0-9]+' will match all device names starting with 'st'
51 followed by one or more numbers. e.g. '^st(0|1)$' will only match
52 'st0' and 'st1'. e.g. 'st0$' will match 'st0' but not 'st1'. See
53 also the -G option for aggregation options.
54 -u, --no-interpolation
56 When replaying a set of archives, by default values are reported
57 according to the requested sample interval (-t option), not
58 according to the actual interval recorded in the archive(s).
59 Without this option PCP interpolates the values to be reported
60 based on the records in the set of archives, which is particularly
61 useful when the -t option is used to replay a set of archives with
62 a longer sampling interval than that with which the archive(s) was
63 originally recorded with. With the -u option, uninterpolated
64 reporting is enabled - every value is reported according to the
65 native recording interval in the set of archives. When the -u
66 option is specified, the -t option makes no sense and is incompat‐
67 ible because the replay interval is always the same as the record‐
68 ing interval in the set of archive. In addition, -u only makes
69 sense when replaying archives, see the -a option on PCPIntro(1),
70 and so if -u is specified then -a must also be specified.
71 -x comma-separated-options
73 Specifies a comma-separated list of one or more extended reporting
74 options as follows:
75 t - prefix every line in the report with a timestamp in ctime(3)
76 format,
77 h - omit the heading, which is otherwise reported every 24 sam‐
78 ples,
79 noidle - Do not display statistics for idle devices.
80
82 The columns in the pcp-tapestat report have the following interpreta‐
83 tion:
84 Timestamp
86 When the -x t option is specified, this column is the timestamp
87 in ctime(3) format.
88 Device Specifies the tape device name. When -G is specified, this is
90 replaced by the aggregation method and regular expression - see
91 the -G and -R options above.
92 r/s The number of reads issued expressed as the number per second
94 averaged over the interval.
95 w/s The number of writes issued expressed as the number per second
97 averaged over the interval.
98 kb_r/s The amount of data read expressed in kilobytes per second aver‐
100 aged over the interval.
101 kb_w/s The amount of data written expressed in kilobytes per second
103 averaged over the interval.
104 r_pct Read percentage wait - the percentage of time over the interval
106 spent waiting for read requests to complete. The time is mea‐
107 sured from when the request is dispatched to the SCSI mid-layer
108 until it signals that it completed.
109 w_pct Write percentage wait - the percentage of time over the interval
111 spent waiting for write requests to complete. The time is mea‐
112 sured from when the request is dispatched to the SCSI mid-layer
113 until it signals that it completed.
114 o_pct Overall percentage wait - the percentage of time over the inter‐
116 val spent waiting for any I/O request to complete (read, write,
117 and other).
118 Rs/s The number of I/Os, expressed as the number per second averaged
120 over the interval, where a non-zero residual value was encoun‐
121 tered.
122 o_cnt The number of I/Os, expressed as the number per second averaged
124 over the interval, that were included as "other". Other I/O
125 includes ioctl calls made to the tape driver and implicit opera‐
126 tions performed by the tape driver such as rewind on close (for
127 tape devices that implement rewind on close). It does not
128 include any I/O performed using methods outside of the tape
129 driver (e.g. via sg ioctls).
130
132 All are generated on standard error and are intended to be self-
133 explanatory.
134
136 Environment variables with the prefix PCP_ are used to parameterize the
137 file and directory names used by PCP. On each installation, the file
138 /etc/pcp.conf contains the local values for these variables. The
139 $PCP_CONF variable may be used to specify an alternative configuration
140 file, as described in pcp.conf(5).
141 For environment variables affecting PCP tools, see pmGetOptions(3).
143
145 PCPIntro(1), pcp(1), pmcd(1), pmchart(1), pmlogger(1), pcp.conf(5) and
146 pcp.env(5).
147Performance Co-Pilot PCP PCP-TAPESTAT(1)