1PCP-IOSTAT(1)               General Commands Manual              PCP-IOSTAT(1)
2
3
4

NAME

6       pmiostat, pcp-iostat - report block I/O statistics
7

SYNOPSIS

9       pcp [pcp options] iostat [-u?]  [-G method] [-P precision] [-R pattern]
10       [-x [dm][,t][,h][,noidle]]
11

DESCRIPTION

13       pcp-iostat reports I/O statistics for SCSI (by default)  or  other  de‐
14       vices (if the -x option is specified).
15

OPTIONS

17       When  invoked  via  the  pcp(1)  command,  the -a/--archive, -h/--host,
18       -O/--origin, -s/--samples,  -t/--interval,  -Z/--timezone  and  several
19       other pcp options become indirectly available; refer to PCPIntro(1) for
20       a complete description of these options.
21
22       The additional command line options available for pcp-iostat are:
23
24       -G method, --aggregate=method
25            Specifies that statistics for device names  matching  the  regular
26            expression specified with the -R regex option should be aggregated
27            according to method.  Note this is aggregation based  on  matching
28            device names (not temporal aggregation).  When -G is used, the de‐
29            vice name column is reported as method(regex), e.g.  if -G sum  -R
30            'sd(a|b)$'  is  specified, the device column will be sum(sd(a|b)$)
31            and summed statistics for sda and sdb will be reported in the  re‐
32            maining columns.  If -G is specified but -R is not specified, then
33            the default regex is .*,  i.e.  matching  all  device  names.   If
34            method  is  sum then the statistics are summed.  This includes the
35            %util column, which may therefore exceed 100% if more than one de‐
36            vice  name  matches.   If  method  is  avg then the statistics are
37            summed and then averaged by dividing by the number of matching de‐
38            vice  names.  If method is min or max, the minimum or maximum sta‐
39            tistics for matching devices are reported, respectively.
40
41       -P N, --precision=N
42            This indicates the precision (number of decimal places) to report.
43            The default precision N may be set to something other than the de‐
44            fault (2).  Note that the avgrq-sz and avgqu-sz fields are  always
45            reported  with  N+1 decimals of precision.  These fields typically
46            have values less than 1.
47
48       -R pattern, --regex=pattern
49            This restricts the report to device names matching a  regular  ex‐
50            pression  pattern.   The given pattern is searched as a perl style
51            regular expression, and will match any portion of a  device  name.
52            e.g. '^sd[a-zA-Z]+' will match all device names starting with 'sd'
53            followed by one or more alphabetic characters.   e.g.  '^sd(a|b)$'
54            will only match 'sda' and 'sdb'.  e.g. 'sda$' will match 'sda' but
55            not 'sdab'.  See also the -G option for aggregation options.
56
57       -u, --no-interpolation
58            When replaying a set of archives, by default values  are  reported
59            according  to  the  requested sample interval (-t option), not ac‐
60            cording to the actual interval recorded in the archive(s).   With‐
61            out  this  option PCP interpolates the values to be reported based
62            on the records in the set of archives, which is particularly  use‐
63            ful  when the -t option is used to replay a set of archives with a
64            longer sampling interval than that with which the  archive(s)  was
65            originally  recorded with.  With the -u option, uninterpolated re‐
66            porting is enabled - every value is reported according to the  na‐
67            tive  recording  interval in the set of archives.  When the -u op‐
68            tion is specified, the -t option makes no sense and is  incompati‐
69            ble  because the replay interval is always the same as the record‐
70            ing interval in the set of archive.  In addition,  -u  only  makes
71            sense  when  replaying archives, see the -a option on PCPIntro(1),
72            and so if -u is specified then -a must also be specified.
73
74       -x comma-separated-options
75            Specifies a comma-separated list of one or more extended reporting
76            options as follows:
77            dm  -  report statistics for device-mapper logical devices instead
78            of SCSI devices,
79            t - prefix every line in the report with a timestamp  in  ctime(3)
80            format,
81            h  -  omit  the heading, which is otherwise reported every 24 sam‐
82            ples,
83            noidle - Do not display statistics for idle devices.
84
85       -?, --help
86            Display usage message and exit.
87

REPORT

89       The columns in the pcp-iostat report have the following interpretation:
90
91       Timestamp
92              When the -x t option is specified, this column is the  timestamp
93              in ctime(3) format.
94
95       Device Specifies  the  scsi  device name, or if -x dm is specified, the
96              device-mapper logical device name.  When -G is  specified,  this
97              is  replaced  by the aggregation method and regular expression -
98              see the -G and -R options above.
99
100       rrqm/s The number of read requests expressed as a rate per-second  that
101              were merged during the reporting interval by the I/O scheduler.
102
103       wrqm/s The number of write requests expressed as a rate per-second that
104              were merged during the reporting interval by the I/O scheduler.
105
106       r/s    The number of read  requests  completed  by  the  device  (after
107              merges), expressed as a rate per second during the reporting in‐
108              terval.
109
110       w/s    The number of write requests  completed  by  the  device  (after
111              merges), expressed as a rate per second during the reporting in‐
112              terval.
113
114       rkB/s  The average volume of data read from  the  device  expressed  as
115              KBytes/second during the reporting interval.
116
117       wkB/s  The  average  volume  of data written to the device expressed as
118              KBytes/second during the reporting interval.
119
120       avgrq-sz
121              The average I/O request size for both reads and  writes  to  the
122              device expressed as Kbytes during the reporting interval.
123
124       avgqu-sz
125              The  average  queue length of read and write requests to the de‐
126              vice during the reporting interval.
127
128       await  The average time in milliseconds that read  and  write  requests
129              were  queued  (and  serviced) to the device during the reporting
130              interval.
131
132       r_await
133              The average time in milliseconds that read requests were  queued
134              (and serviced) to the device during the reporting interval.
135
136       w_await
137              The average time in milliseconds that write requests were queued
138              (and serviced) to the device during the reporting interval.
139
140       %util  The percentage of time during the reporting  interval  that  the
141              device  was busy processing requests.  A value of 100% indicates
142              device saturation.
143

PCP ENVIRONMENT

145       Environment variables with the prefix PCP_ are used to parameterize the
146       file  and  directory names used by PCP.  On each installation, the file
147       /etc/pcp.conf contains the  local  values  for  these  variables.   The
148       $PCP_CONF  variable may be used to specify an alternative configuration
149       file, as described in pcp.conf(5).
150
151       For environment variables affecting PCP tools, see pmGetOptions(3).
152

SEE ALSO

154       PCPIntro(1), pcp(1), iostat2pcp(1), pmcd(1),  pmchart(1),  pmlogger(1),
155       pcp.conf(5) and pcp.env(5).
156
157
158
159Performance Co-Pilot                  PCP                        PCP-IOSTAT(1)
Impressum