mirror of
https://github.com/ibm-s390-linux/s390-tools.git
synced 2026-08-05 02:14:52 +00:00
This commit is based on the s390-tools-1.39.0 version. Changes on top of s390-tools-1.39.0: - Add MIT license to all source files - Add LICENSE file - Transform REAMDE to README.md (markdown) - Add AUTHORS.md file - Add CONTRIBUTING.md file - Move changelog from README to CHANGELOG.md file Reviewed-by: Stefan Haberland <sth@linux.vnet.ibm.com> Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
222 lines
5.9 KiB
Groff
222 lines
5.9 KiB
Groff
.\" Copyright 2017 IBM Corp.
|
|
.\" s390-tools is free software; you can redistribute it and/or modify
|
|
.\" it under the terms of the MIT license. See LICENSE for details.
|
|
.\"
|
|
.TH ZIOMON_MGR 8 "Jul 2008" "s390-tools"
|
|
|
|
.SH NAME
|
|
ziorep_traffic \- I/O traffic report for FCP adapters.
|
|
|
|
.SH SYNOPSIS
|
|
.B ziorep_traffic
|
|
[-V] [-v] [-h] [-b <begin>] [-e <end>] [-i <time>] [-s] [-c <chpid>] [-u <id>] [-t <num>] [-p <port>] [-l <lun>] [-d <fdev> ] [-m <mdev> ] [-x] [-D] [-C a|u|p|m|A] <filename>
|
|
|
|
|
|
|
|
.SH DESCRIPTION
|
|
.B ziorep_traffic
|
|
Prints a report from the specified data.
|
|
|
|
.SH OPTIONS
|
|
.TP
|
|
.BR "\-h" " or " "\-\-help"
|
|
Print help information, then exit.
|
|
|
|
.TP
|
|
.BR "\-v" " or " "\-\-version"
|
|
Print version information, then exit.
|
|
|
|
.TP
|
|
.BR "\-V" " or " "\-\-verbose"
|
|
Be verbose.
|
|
|
|
.TP
|
|
.BR "\-b" " or " "\-\-begin"
|
|
Limit the timeframe to consider to data beginning with the specified date.
|
|
.br
|
|
Dates must be specified in the following format: YYYY-MM-DD HH:MM[:SS].
|
|
.br
|
|
E.g. 2008-03-21 09:08 is 9:08 on March 21, 2008.
|
|
|
|
.TP
|
|
.BR "\-e" " or " "\-\-end"
|
|
Limit the timeframe to consider to data ending with the specified date.
|
|
.br
|
|
Dates must be specified in the following format: YYYY-MM-DD HH:MM[:SS].
|
|
.br
|
|
E.g. 2008-03-21 09:08 is 9:08 on March 21, 2008.
|
|
|
|
.TP
|
|
.BR "\-i" " or " "\-\-interval"
|
|
Specify an aggregation interval. The interval is given in seconds, and must be a multiple
|
|
of the interval as found in the source data.
|
|
|
|
.TP
|
|
.BR "\-s" " or " "\-\-summary"
|
|
Print a summary of the data, then exit.
|
|
|
|
.TP
|
|
.BR "\-c" " or " "\-\-chpid"
|
|
Consider the specified physical adapter. Adapters must be specified in hex.
|
|
.br
|
|
If multiple adapters should be specified, specify each one separately.
|
|
.br
|
|
E.g. '-c 7a'.
|
|
|
|
.TP
|
|
.BR "\-u" " or " "\-\-bus-id"
|
|
Consider the specified bus-IDs. bus-IDs must be specified in hex.
|
|
.br
|
|
If multiple bus-IDs should be specified, specify each one separately.
|
|
.br
|
|
E.g. '-u 0.0.4c78'.
|
|
|
|
.TP
|
|
.BR "\-p" " or " "\-\-port"
|
|
Consider the specified target port. Target ports must be specified in hex.
|
|
.br
|
|
If multiple ports should be specified, specify each one separately.
|
|
.br
|
|
E.g. '-p 0x500507630040710b'.
|
|
|
|
.TP
|
|
.BR "\-l" " or " "\-\-lun"
|
|
Consider the specified lun. luns must be specified in hex.
|
|
.br
|
|
If multiple luns should be specified, specify each one separately.
|
|
.br
|
|
E.g. '-l 0x4021402d00000000'.
|
|
|
|
.TP
|
|
.BR "\-d" " or " "\-\-device"
|
|
Consider the specified device.
|
|
.br
|
|
If multiple devices should be specified, specify each one separately.
|
|
.br
|
|
E.g. '-d sda'.
|
|
|
|
.TP
|
|
.BR "\-m" " or " "\-\-mdev"
|
|
Consider the specified multipath device.
|
|
.br
|
|
If multiple devices should be specified, specify each one separately.
|
|
.br
|
|
E.g. '-m 36005076303ffc1040002120'.
|
|
|
|
.TP
|
|
.BR "\-x" " or " "\-\-export-csv"
|
|
Write data to file(s) in CSV format. Output filenames will be based on the data filename.
|
|
|
|
.TP
|
|
.BR "\-t" " or " "\-\-topline"
|
|
Repeat topline after specified number of frames.
|
|
.br
|
|
0 for no repeat (default).
|
|
|
|
.TP
|
|
.BR "\-D" " or " "\-\-detailed"
|
|
Print histograms.
|
|
|
|
.TP
|
|
.BR "\-C" " or " "\-\-collapse"
|
|
Collapse the data by the specified criterion:
|
|
.br
|
|
.BR "a"
|
|
collapse by physical adapter.
|
|
.br
|
|
.BR "u"
|
|
collapse by bus-ID.
|
|
.br
|
|
.BR "p"
|
|
collapse by target port.
|
|
.br
|
|
.BR "m"
|
|
collapse by multipath device.
|
|
.br
|
|
.BR "A"
|
|
collapse all data into a single dataset.
|
|
|
|
|
|
.SH OUTPUT
|
|
Here is a list of the columns and their descriptions.
|
|
Timestamps of the frames printed depict the ending of the respective timeframe.
|
|
|
|
.TP
|
|
.BR "I/O rt MB/s"
|
|
I/O rate of the respective device in megabytes per second, considering individual requests and their total processing time, including channel latency, etc.
|
|
.br
|
|
.IR "min" " and " "max"
|
|
give the minimum and maximum respectively.
|
|
|
|
.TP
|
|
.BR "thrp in MB/s"
|
|
Throughput of the respective device in megabytes per second, considering times of activity only.
|
|
.br
|
|
.IR "avg" " and " "stdev"
|
|
give the average utilization as well as its standard deviation respectively.
|
|
|
|
.TP
|
|
.BR "I/O requests"
|
|
Statistics on the numbers and kinds of I/O requests seen.
|
|
.br
|
|
.IR "#reqs"
|
|
is the total number of requests,
|
|
.br
|
|
.IR "rd"
|
|
is the number of read requests,
|
|
.br
|
|
.IR "wrt"
|
|
is the number of write requests and
|
|
.br
|
|
.IR "bidi"
|
|
is the number of bi-directional requests.
|
|
|
|
.TP
|
|
.BR "I/O subs. lat. in us"
|
|
Latency of the I/O subsystem in microseconds. This is the time needed to execute a request, from the I/O stack down to the hardware and back up again. It excludes channel and fabric latencies.
|
|
.br
|
|
.IR "min" ", " "max" ", " "avg" " and " "stdev"
|
|
give the minimum, maximum and average latency as well as its standard deviation respectively.
|
|
|
|
.TP
|
|
.BR "channel lat. in us"
|
|
Channel latency in microseconds. This is the time needed to execute a request on the FCP adapter. It excludes I/O subsystem and fabric latencies.
|
|
.br
|
|
.IR "min" ", " "max" ", " "avg" " and " "stdev"
|
|
give the minimum, maximum and average latency as well as its standard deviation respectively.
|
|
|
|
.TP
|
|
.BR "fabric lat. in us"
|
|
Fabric latency in microseconds. This is the time needed for the requests within the fabric, from the FCP adapter to the target device and back again. It excludes I/O subsystem and channel latencies.
|
|
.br
|
|
.IR "min" ", " "max" ", " "avg" " and " "stdev"
|
|
give the minimum, maximum and average latency as well as its standard deviation respectively.
|
|
|
|
|
|
.SH EXAMPLES
|
|
.B Example
|
|
.br
|
|
Print a traffic report for all devices.
|
|
Only data after 8:57 April 5, 2008 should be considered.
|
|
Data should be aggregated to 60 second intervals.
|
|
|
|
ziorep_traffic -i 60 -b "2008-04-05 08:57" sample.log
|
|
|
|
.B Example
|
|
.br
|
|
Print a detailed traffic report for all devices connected to target port 0x500507630313c562.
|
|
Data should be aggregated over the whole data range.
|
|
|
|
ziorep_traffic -i 0 -p 0x500507630313c562 sample.log
|
|
|
|
.B Example
|
|
.br
|
|
Print a detailed traffic report for all devices that are either connected to target port 0x500507630313c562 or part of multipath device 36005076303ffc5620000000000001314.
|
|
The result should be collapsed on the bus-ID level.
|
|
|
|
ziorep_traffic -C u -p 0x500507630313c562 -m 36005076303ffc5620000000000001314 sample.log
|
|
|
|
.SH "SEE ALSO"
|
|
.BR ziorep_config (8),
|
|
.BR ziorep_utilization (8)
|