mirror of
https://github.com/ceph/ceph
synced 2024-12-16 00:15:35 +00:00
76 lines
2.5 KiB
Groff
76 lines
2.5 KiB
Groff
.TH CSYN 8
|
|
.SH NAME
|
|
csyn \- ceph synthetic workload generator
|
|
.SH SYNOPSIS
|
|
.B csyn
|
|
[ \fB\-m monaddr:port\fP ]
|
|
\fB--syn\fI command ...\fP
|
|
.SH DESCRIPTION
|
|
.B csyn
|
|
is a simple synthetic workload generator for the Ceph distributed file system.
|
|
It uses the userspace client library to generate simple workloads against
|
|
a currently running file system. The file system need not be mounted via
|
|
.BR cfuse (8)
|
|
or the kernel client.
|
|
.PP
|
|
One or more \fB--syn\fI command\fR arguments specify the particular workload,
|
|
as documented below.
|
|
.SH OPTIONS
|
|
.TP
|
|
\fB\-d\fP
|
|
Detach from console and daemonize after startup.
|
|
.TP
|
|
\fB\-c\fI ceph.conf\fR, \fB\-\-conf=\fIceph.conf\fR
|
|
Use \fIceph.conf\fP configuration file instead of the default \fI/etc/ceph/ceph.conf\fP
|
|
for runtime configuration options.
|
|
.TP
|
|
\fB\-C\fI cluster.conf\fR, \fB\-\-cluster-conf=\fIcluster.conf\fR
|
|
Use \fIcluster.conf\fP cluster configuration file instead of the default \fI/etc/ceph/cluster.conf\fP
|
|
to determine monitor addresses during startup.
|
|
.TP
|
|
\fB\-m\fI monaddress[:port]\fR
|
|
Connect to specified monitor (instead of looking through \fIcluster.conf\fR).
|
|
.TP
|
|
\fB\-\-num_client\fI num\fR
|
|
Run \fInum\fR different clients, each in a separate thread.
|
|
.TP
|
|
\fB\-\-syn\fI workloadspec\fR
|
|
Run the given workload. May be specified as many times as needed. Workloads will
|
|
normally run sequentially.
|
|
.SH WORKLOADS
|
|
Each workload should be preceeded by \fB--syn\fP on the command line. This is not a
|
|
complete list.
|
|
.TP
|
|
\fBmknap\fI path snapname\fP
|
|
Create a snapshot called \fIsnapname\fP on \fIpath\fP.
|
|
.TP
|
|
\fBrmsnap\fI path snapname\fP
|
|
Delete snapshot called \fIsnapname\fP on \fIpath\fP.
|
|
.TP
|
|
\fBrmfile\fI path\fP
|
|
Delete/unlink \fIpath\fP.
|
|
.TP
|
|
\fBwritefile\fI sizeinmb blocksize\fP
|
|
Create a file, named after our client id, that is \fIsizeinmb\fP MB by writing \fIblocksize\fP chunks.
|
|
.TP
|
|
\fBreadfile\fI sizeinmb blocksize\fP
|
|
Read file, named after our client id, that is \fIsizeinmb\fP MB by writing \fIblocksize\fP chunks.
|
|
.TP
|
|
\fBrw\fI sizeinmb blocksize\fP
|
|
Write file, then read it back, as above.
|
|
.TP
|
|
\fBmakedirs\fI numsubdirs numfiles depth\fP
|
|
Create a hierarchy of directories that is \fIdepth\fP levels deep. Give each
|
|
directory \fInumsubdirs\fP subdirectories and \fInumfiles\fP files.
|
|
.TP
|
|
\fBwalk\fP
|
|
Recursively walk the file system (like \fBfind\fP).
|
|
|
|
.SH AVAILABILITY
|
|
.B csyn
|
|
is part of the Ceph distributed file system. Please refer to the Ceph wiki at
|
|
http://ceph.newdream.net/wiki for more information.
|
|
.SH SEE ALSO
|
|
.BR ceph (8),
|
|
.BR cfuse (8)
|