1PCP-IOSTAT(1) General Commands Manual PCP-IOSTAT(1)
2
3
4
6 pmiostat, pcp-iostat - report block I/O statistics
7
9 pcp [pcp options] iostat [-u] [-G method] [-P precision] [-R pattern]
10 [-x [dm][,t][,h][,noidle]]
11
13 pcp-iostat reports I/O statistics for SCSI (by default) or other
14 devices (if the -x option is specified).
15
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
22 The options which control the timing and layout of the information
23 reported by pcp-iostat are as follows:
24
25 -G method, --aggregate=method
26 Specifies that statistics for device names matching the regular
27 expression specified with the -R regex option should be aggregated
28 according to method. Note this is aggregation based on matching
29 device names (not temporal aggregation). When -G is used, the
30 device name column is reported as method(regex), e.g. if -G sum
31 -R 'sd(a|b)$' is specified, the device column will be
32 sum(sd(a|b)$) and summed statistics for sda and sdb will be
33 reported in the remaining columns. If -G is specified but -R is
34 not specified, then the default regex is .*, i.e. matching all
35 device names. If method is sum then the statistics are summed.
36 This includes the %util column, which may therefore exceed 100% if
37 more than one device name matches. If method is avg then the sta‐
38 tistics are summed and then averaged by dividing by the number of
39 matching device names. If method is min or max, the minimum or
40 maximum statistics for matching devices are reported, respec‐
41 tively.
42
43 -P N, --precision=N
44 This indicates the precision (number of decimal places) to report.
45 The default precision N may be set to something other than the
46 default (2). Note that the avgrq-sz and avgqu-sz fields are
47 always reported with N+1 decimals of precision. These fields typ‐
48 ically have values less than 1.
49
50 -R pattern, --regex=pattern
51 This restricts the report to device names matching a regular
52 expression pattern. The given pattern is searched as a perl style
53 regular expression, and will match any portion of a device name.
54 e.g. '^sd[a-zA-Z]+' will match all device names starting with 'sd'
55 followed by one or more alphabetic characters. e.g. '^sd(a|b)$'
56 will only match 'sda' and 'sdb'. e.g. 'sda$' will match 'sda' but
57 not 'sdab'. See also the -G option for aggregation options.
58
59 -u, --no-interpolation
60 When replaying a set of archives, by default values are reported
61 according to the requested sample interval (-t option), not
62 according to the actual interval recorded in the archive(s).
63 Without this option PCP interpolates the values to be reported
64 based on the records in the set of archives, which is particularly
65 useful when the -t option is used to replay a set of archives with
66 a longer sampling interval than that with which the archive(s) was
67 originally recorded with. With the -u option, uninterpolated
68 reporting is enabled - every value is reported according to the
69 native recording interval in the set of archives. When the -u
70 option is specified, the -t option makes no sense and is incompat‐
71 ible because the replay interval is always the same as the record‐
72 ing interval in the set of archive. In addition, -u only makes
73 sense when replaying archives, see the -a option on PCPIntro(1),
74 and so if -u is specified then -a must also be specified.
75
76 -x comma-separated-options
77 Specifies a comma-separated list of one or more extended reporting
78 options as follows:
79 dm - report statistics for device-mapper logical devices instead
80 of SCSI devices,
81 t - prefix every line in the report with a timestamp in ctime(3)
82 format,
83 h - omit the heading, which is otherwise reported every 24 sam‐
84 ples,
85 noidle - Do not display statistics for idle devices.
86
88 The columns in the pcp-iostat report have the following interpretation:
89
90 Timestamp
91 When the -x t option is specified, this column is the timestamp
92 in ctime(3) format.
93
94 Device Specifies the scsi device name, or if -x dm is specified, the
95 device-mapper logical device name. When -G is specified, this
96 is replaced by the aggregation method and regular expression -
97 see the -G and -R options above.
98
99 rrqm/s The number of read requests expressed as a rate per-second that
100 were merged during the reporting interval by the I/O scheduler.
101
102 wrqm/s The number of write requests expressed as a rate per-second that
103 were merged during the reporting interval by the I/O scheduler.
104
105 r/s The number of read requests completed by the device (after
106 merges), expressed as a rate per second during the reporting
107 interval.
108
109 w/s The number of write requests completed by the device (after
110 merges), expressed as a rate per second during the reporting
111 interval.
112
113 rkB/s The average volume of data read from the device expressed as
114 KBytes/second during the reporting interval.
115
116 wkB/s The average volume of data written to the device expressed as
117 KBytes/second during the reporting interval.
118
119 avgrq-sz
120 The average I/O request size for both reads and writes to the
121 device expressed as Kbytes during the reporting interval.
122
123 avgqu-sz
124 The average queue length of read and write requests to the
125 device during the reporting interval.
126
127 await The average time in milliseconds that read and write requests
128 were queued (and serviced) to the device during the reporting
129 interval.
130
131 r_await
132 The average time in milliseconds that read requests were queued
133 (and serviced) to the device during the reporting interval.
134
135 w_await
136 The average time in milliseconds that write requests were queued
137 (and serviced) to the device during the reporting interval.
138
139 %util The percentage of time during the reporting interval that the
140 device was busy processing requests. A value of 100% indicates
141 device saturation.
142
144 All are generated on standard error and are intended to be self-
145 explanatory.
146
148 Environment variables with the prefix PCP_ are used to parameterize the
149 file and directory names used by PCP. On each installation, the file
150 /etc/pcp.conf contains the local values for these variables. The
151 $PCP_CONF variable may be used to specify an alternative configuration
152 file, as described in pcp.conf(5).
153
154 For environment variables affecting PCP tools, see pmGetOptions(3).
155
157 pcp(1), PCPIntro(1), iostat2pcp(1), pmcd(1), pmchart(1), pmlogger(1),
158 pcp.conf(5) and pcp.env(5).
159
160
161
162Performance Co-Pilot PCP PCP-IOSTAT(1)