.\" -*- mode: troff; coding: utf-8 -*-
.\" Automatically generated by Podwrapper::Man 1.38.2 (Pod::Simple 3.43)
.\"
.\" Standard preamble:
.\" ========================================================================
.de Sp \" Vertical space (when we can't use .PP)
.if t .sp .5v
.if n .sp
..
.de Vb \" Begin verbatim text
.ft CW
.nf
.ne \\$1
..
.de Ve \" End verbatim text
.ft R
.fi
..
.\" \*(C` and \*(C' are quotes in nroff, nothing in troff, for use with C<>.
.ie n \{\
.    ds C` ""
.    ds C' ""
'br\}
.el\{\
.    ds C`
.    ds C'
'br\}
.\"
.\" Escape single quotes in literal strings from groff's Unicode transform.
.ie \n(.g .ds Aq \(aq
.el       .ds Aq '
.\"
.\" If the F register is >0, we'll generate index entries on stderr for
.\" titles (.TH), headers (.SH), subsections (.SS), items (.Ip), and index
.\" entries marked with X<> in POD.  Of course, you'll have to process the
.\" output yourself in some meaningful fashion.
.\"
.\" Avoid warning from groff about undefined register 'F'.
.de IX
..
.nr rF 0
.if \n(.g .if rF .nr rF 1
.if (\n(rF:(\n(.g==0)) \{\
.    if \nF \{\
.        de IX
.        tm Index:\\$1\t\\n%\t"\\$2"
..
.        if !\nF==2 \{\
.            nr % 0
.            nr F 2
.        \}
.    \}
.\}
.rr rF
.\" ========================================================================
.\"
.IX Title "nbdkit-scan-filter 1"
.TH nbdkit-scan-filter 1 2024-05-10 nbdkit-1.38.2 NBDKIT
.\" For nroff, turn off justification.  Always turn off hyphenation; it makes
.\" way too many mistakes in technical documents.
.if n .ad l
.nh
.SH NAME
nbdkit\-scan\-filter \- scan disk prefetching data ahead of sequential reads
.SH SYNOPSIS
.IX Header "SYNOPSIS"
.Vb 2
\& nbdkit \-\-filter=scan PLUGIN [scan\-ahead=false] [scan\-clock=false]
\&                             [scan\-forever=true] [scan\-size=]NN
.Ve
.PP
.Vb 1
\& nbdkit \-\-filter=scan \-\-filter=cache PLUGIN
.Ve
.PP
.Vb 1
\& nbdkit \-\-filter=scan \-\-filter=cow PLUGIN cow\-on\-cache=true
.Ve
.SH DESCRIPTION
.IX Header "DESCRIPTION"
\&\f(CW\*(C`nbdkit\-scan\-filter\*(C'\fR is a filter that scans the disk prefetching
data.  It is sometimes useful if you expect that the client will read
the disk sequentially.
.PP
The basic operation of the filter is that when a client connects, the
filter will start issuing \f(CW\*(C`.cache\*(C'\fR (prefetch) requests to the plugin
across the whole disk.  Plugins which support this command will
prefetch the data, making subsequent reads faster.  For plugins which
do not support this command, you can inject \fBnbdkit\-cache\-filter\fR\|(1)
below (after) this filter, giving approximately the same effect.
\&\fBnbdkit\-cow\-filter\fR\|(1) can be used instead of nbdkit-cache-filter, if
you add the \f(CW\*(C`cow\-on\-cache=true\*(C'\fR option.
.PP
Various parameters can be used to tune scanning, although the defaults
should be suitable in most cases.
.PP
A similar filter is \fBnbdkit\-readahead\-filter\fR\|(1).
.SS Limitations
.IX Subsection "Limitations"
In a number of significant cases this filter will do nothing.  The
filter will print a warning message if this happens.
.IP "Thread model must be parallel*" 4
.IX Item "Thread model must be parallel*"
For example \fBnbdkit\-curl\-plugin\fR\|(1) only supports
\&\f(CW\*(C`serialize_requests\*(C'\fR, and so this filter cannot perform prefetches in
parallel with the read requests.
.IP "Only scans while clients are connected*" 4
.IX Item "Only scans while clients are connected*"
The current filter only scans while there is at least one client
connected.
.IP "Only scans the default export*" 4
.IX Item "Only scans the default export*"
The current filter only scans the default export and ignores all
clients connecting to the non-default export name.
.Sp
*We may be able to lift these restrictions in future.
.ie n .IP "Underlying filters or plugin must support "".cache"" (prefetch)" 4
.el .IP "Underlying filters or plugin must support \f(CW.cache\fR (prefetch)" 4
.IX Item "Underlying filters or plugin must support .cache (prefetch)"
Very many plugins do not have the concept of prefetching and/or
do not implement the \f(CW\*(C`.cache\*(C'\fR callback, and so there is no
way for this filter to issue prefetches.
.Sp
You can usually get around this by adding \fI\-\-filter=cache\fR after this
filter as explained above.
.IP "Prefetching the whole disk may load it all into cache" 4
.IX Item "Prefetching the whole disk may load it all into cache"
In particular if you use this filter together with
\&\fBnbdkit\-cache\-filter\fR\|(1) or \fBnbdkit\-cow\-filter\fR\|(1), they will cache
the whole content of the plugin into a temporary file.  This may be
many gigabytes of data, consuming all space in \fI/var/tmp\fR.  Of course
this is the whole point of using this filter, but you should be aware
of it.
.Sp
If using the cache filter, the total size of the cache can be limited
(see "CACHE MAXIMUM SIZE" in \fBnbdkit\-cache\-filter\fR\|(1)).
.SH PARAMETERS
.IX Header "PARAMETERS"
.IP \fBscan\-ahead=false\fR 4
.IX Item "scan-ahead=false"
By default the filter tries to stay ahead of incoming read requests.
That is to say, it starts prefetching at the beginning of the disk and
continues incrementally, but if the client issues a read beyond the
current prefetch point then the filter skips forward and begins
prefetching after the read.
.Sp
However if you set this parameter to false, then this behaviour is
disabled.  The filter simply prefetches sequentially regardless of
client requests.
.IP \fBscan\-clock=false\fR 4
.IX Item "scan-clock=false"
By default, if all clients disconnect and then another client
connects, prefetching resumes at the same place in the disk.  (Like
stopping and starting a clock.)
.Sp
If you set this parameter to false, then the filter starts prefetching
from the beginning of the disk again.
.IP \fBscan\-forever=true\fR 4
.IX Item "scan-forever=true"
By default the filter scans over the disk once and then stops.
.Sp
If you set this parameter to true, then after the disk has been
prefetched completely, the filter goes back to the beginning and
starts over, repeating this for as long as nbdkit is running and there
are clients connected.
.IP \fBscan\-size=\fRNN 4
.IX Item "scan-size=NN"
This parameter controls the prefetch block size.  The default is
\&\f(CW\*(C`2M\*(C'\fR.  This must be a power of 2 and most plugins will have their own
limits on the amount of data they can prefetch in a single request.
.SH FILES
.IX Header "FILES"
.ie n .IP \fR\fI$filterdir\fR\fI/nbdkit\-scan\-filter.so\fR 4
.el .IP \fR\f(CI$filterdir\fR\fI/nbdkit\-scan\-filter.so\fR 4
.IX Item "$filterdir/nbdkit-scan-filter.so"
The filter.
.Sp
Use \f(CW\*(C`nbdkit \-\-dump\-config\*(C'\fR to find the location of \f(CW$filterdir\fR.
.SH VERSION
.IX Header "VERSION"
\&\f(CW\*(C`nbdkit\-scan\-filter\*(C'\fR first appeared in nbdkit 1.32.
.SH "SEE ALSO"
.IX Header "SEE ALSO"
\&\fBnbdkit\fR\|(1),
\&\fBnbdkit\-cache\-filter\fR\|(1),
\&\fBnbdkit\-cow\-filter\fR\|(1),
\&\fBnbdkit\-file\-plugin\fR\|(1),
\&\fBnbdkit\-readahead\-filter\fR\|(1),
\&\fBnbdkit\-retry\-filter\fR\|(1),
\&\fBnbdkit\-torrent\-plugin\fR\|(1),
\&\fBnbdkit\-vddk\-plugin\fR\|(1),
\&\fBnbdkit\-filter\fR\|(3),
\&\fBqemu\-img\fR\|(1).
.SH AUTHORS
.IX Header "AUTHORS"
Richard W.M. Jones
.SH COPYRIGHT
.IX Header "COPYRIGHT"
Copyright Red Hat
.SH LICENSE
.IX Header "LICENSE"
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
.IP \(bu 4
Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
.IP \(bu 4
Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
.IP \(bu 4
Neither the name of Red Hat nor the names of its contributors may be
used to endorse or promote products derived from this software without
specific prior written permission.
.PP
THIS SOFTWARE IS PROVIDED BY RED HAT AND CONTRIBUTORS ''AS IS'' AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL RED HAT OR
CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
SUCH DAMAGE.