cpet/openbsd-ports
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Enhancements from jmc@

Browse Source
This commit is contained in:
marco 2005-12-07 16:09:39 +00:00
parent 0f09687db1
commit c7a37cdf3c
1 changed files with 77 additions and 57 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

134
sysutils/iogen/src/iogen.8
View File

@ -1,4 +1,4 @@
.\" $OpenBSD: iogen.8,v 1.1 2005/12/06 20:54:59 marco Exp $
.\" $OpenBSD: iogen.8,v 1.2 2005/12/07 16:09:39 marco Exp $
.\"
.\" Copyright (c) 2005 Marco Peereboom <marco@peereboom.us>
.\"
@ -14,103 +14,123 @@
.\" ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
.\" OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
.\"
.Dd Dec 6, 2005
.Dd December 6, 2005
.Dt IOGEN 8
.Os
.Sh NAME
.Nm iogen
.Nd i/o generator.
.Nd I/O generator
.Sh SYNOPSIS
.Nm iogen
.Op Fl r
.Op Fl k
.Op Fl s Ar max-file-size
.Bk -words
.Op Fl kr
.Op Fl b Ar max-io-size
.Op Fl p Ar read-percentage
.Op Fl d Ar target-directory
.Op Fl f Ar result-directory
.Op Fl n Ar nr-forks
.Op Fl p Ar read-percentage
.Op Fl s Ar max-file-size
.Op Fl t Ar update-time
.Ek
.Sh DESCRIPTION
.Nm
is a lightweight tool that generates heavily fragmented i/o.
It accomplishes this by forking an N number of children that run io to a
is a lightweight tool that generates heavily fragmented I/O.
It accomplishes this by forking a number of children that run I/O to a
filesystem.
.Pp
This tool is intended to test storage stacks under stress and worst case
scenarios.
However due to heavy fragmentation of the io files it tends to bypass caching
algorithms in storage stacks.
However due to heavy fragmentation of the I/O files,
it tends to bypass caching algorithms in storage stacks.
.Pp
The options are as follows:
.Pp
.Bl -tag -width xlxltarget-directory
.It Fl r
Randomize io size between 1 and max-io-size.
Enabling this flag will disable data verification.
Default is disabled.
.It Fl k
Kill all running iogen processes.
.It Fl s Ar max-file-size
The file where the io is run to and from will grow sequentially until it is
bigger or equal to this value.
At that point all write io will also become random.
Default is 1GB.
.Bl -tag -width Ds
.It Fl b Ar max-io-size
This is the fixed io size unless the -r flag is set.
Default is 64KB.
This is the fixed I/O size unless the -r flag is set.
The default is 64KB.
.It Fl d Ar target-directory
This is the directory where the io file will be written to.
Default is current directory.
This is the directory where the I/O file will be written to.
The default is the current working directory.
.It Fl f Ar result-directory
This is the directory where the result file will be written to.
The result file is update every update-time seconds with statistics.
Default is current directory.
The result file is updated every
.Ar update-time
seconds with statistics.
The default is the current working directory.
.It Fl k
Kill all running
.Nm
processes.
.It Fl n Ar nr-forks
This will determine how many identical processes will be forked to run io.
Default is 1.
This will determine how many identical processes will be forked to run I/O.
The default is 1.
.It Fl r
Randomize I/O size between 1 and
.Ar max-io-size .
Enabling this flag will disable data verification.
The default is disabled.
.It Fl s Ar max-file-size
The file where the I/O is run to and from will grow sequentially until it is
bigger or equal to this value.
At that point all write I/O will also become random.
The default is 1GB.
.It Fl t Ar update-time
This determines the minimal amount of time between updates.
Under heavy io this value can be skewed due to asynchronous nature of alarm(3).
Default is 60 seconds.
Under heavy I/O this value can be skewed due to
the asynchronous nature of
.Xr alarm 3 .
The default is 60 seconds.
.El
.Pp
Examples:
.Bd -literal -offset indent
Run iogen with all defaults in the current directory:
$ iogen
.Pp
Run iogen with all defaults with a 1 second result file update:
$ iogen -t 1
.Pp
Run iogen with a 2GB max file, 128KB io size and result in /tmp:
$ iogen -s 2g -b 128k -t 1 -f /tmp
.Ed
.Pp
The algorithm for io generation is incredibly simple however it has proven to
be very effective at bringing out issues in storage stacks.
Although the algorithm for I/O generation is incredibly simple,
it has proven to be very effective at
bringing out issues in storage stacks.
It first grows the initial file a minimal amount to be able to start running
io in it.
After the initial growth it reads randomly within the current file size.
I/O in it.
After the initial growth,
it reads randomly within the current file size.
Every run is a distribution between reads and writes which is governed by
the read percentage value.
The file is grown sequentially until it reaches maximum file size.
Whenever this happens a message is logged to syslogd(8) and all writes become
random.
.Pp
To monitor progress one can tail(1) the result file which is updated every
update-time interval or send the process a HUP signal.
Whenever an io processes receives a HUP signal it prints statistical values
to stderr(4) at its earliest convenience.
To monitor progress one can
.Xr tail 1
the result file which is updated every
.Ar update-time
interval or send the process a
.Dv HUP
signal.
Whenever an I/O process receives a
.Dv HUP
signal, it prints statistical values to
.Xr stderr 4
at its earliest convenience.
.Sh EXAMPLES
Run iogen with all defaults in the current working directory:
.Pp
.Dl $ iogen
.Pp
Run
.Nm
with all defaults and a 1 second result file update:
.Pp
.Dl $ iogen -t 1
.Pp
Run
.Nm
with a 2GB max file, 128KB I/O size, and result file in
.Pa /tmp :
.Pp
.Dl $ iogen -s 2g -b 128k -t 1 -f /tmp
.Sh HISTORY
The first version of
.Nm
was written in 2005.
.Sh AUTHORS
Marco Peereboom
.Aq marco@peereboom.us
.An Marco Peereboom Aq marco@peereboom.us
.Sh CAVEATS
This tool is capable of running extremely heavy io.
This tool is capable of running extremely heavy I/O.
It is known to have broken hardware before so please use caution and don't
complain if something bad happens.