NAME

       nbdkit-log-filter - nbdkit log filter

SYNOPSIS

        nbdkit --filter=log PLUGIN
                            [logfile=FILE | logscript=SCRIPT] [logappend=BOOL]
                            [PLUGIN-ARGS...]

DESCRIPTION

       "nbdkit-log-filter" is a filter that logs all transactions to a file or external script.

       When used as the first filter, it can show the original client requests.  As a later
       filter, it can show how earlier filters have modified the original request.

       When using "logfile=FILE", logs are written to a log file with the format described in
       "LOG FILE FORMAT" below.

       When using "logscript=SCRIPT", logs invoke the external script.  See "LOG SCRIPT" below.

       An alternative to this filter is simply to run nbdkit with the -f and -v flags which
       enable verbose debugging to stderr.  This logs many aspects of nbdkit operation, but
       requires running nbdkit in the foreground.  The log filter uses a more parsimonious and
       more easily parsable format and works when nbdkit runs in the background.

PARAMETERS

       "logfile" or "logscript" or both can be given.  If neither then the filter is inactive.

       logfile=FILE
           The file where the log is written.  See "LOG FILE FORMAT" below.

       logscript=SCRIPT
           (nbdkit ≥ 1.24)

           Log lines invoke an external script.  See "LOG SCRIPT" below.

       logappend=true
       logappend=false
           (nbdkit ≥ 1.8)

           This only affects "logfile".  If "false" (the default), if the file already exists it
           will be truncated.  If "true", the filter appends to the existing log file.

EXAMPLES

       Serve the file disk.img, and log each client transaction in the file disk.log:

        nbdkit --filter=log file disk.img logfile=disk.log

       Repeat the task, but with the cow (copy-on-write) filter to perform local caching of data
       served from the original plugin:

        nbdkit --filter=cow --filter=log file disk.img logfile=disk.log2

       After running a client that performs the same operations under each of the two servers,
       you can compare disk.log and disk.log2 to see the impact of the caching.

LOG FILE FORMAT

       An example logging session of a client that requests an export list before performing a
       single successful read is:

        2020-08-06 02:07:23.080415 ListExports id=1 readonly=0 tls=0 ...
        2020-08-06 02:07:23.080502 ...ListExports id=1 exports=("") return=0
        2020-08-06 02:07:23.080712 connection=1 Connect export="" tls=0 size=0x400 minsize=0x1 prefsize=0x200 maxsize=0xffffffff write=1 flush=1 rotational=0 trim=1 zero=2 fua=2 extents=1 cache=2 fast_zero=1
        2020-08-06 02:07:23.080907 connection=1 Read id=1 offset=0x0 count=0x200 ...
        2020-08-06 02:07:23.080927 connection=1 ...Read id=1 return=0
        2020-08-06 02:07:23.081255 connection=1 Disconnect transactions=1

       All lines start with a timestamp in "YYYY-MM-DD HH:MM:ZZ.MS" format.

       For connected calls, "connection=N" is present to distinguish between clients.

       The action follows.  Currently the following actions are logged: ListExports, Ready, Fork,
       Preconnect, Connect, Read, Write, Zero, Trim, Extents, Cache, Flush and Disconnect.

       Some actions are logged across two lines showing the call and return value.  Because
       nbdkit handles requests in parallel different requests may be intermingled.  Use the
       "id=N" field for correlation, it is unique per connection.

       Strings and lists are shell-quoted.

LOG SCRIPT

       If "logscript=SCRIPT" is given on the command line then log entries are passed to the
       external script.

       The script is passed several shell variables:

       $act
           The action name, like "Read", "Write" etc.

       $connection
           The connection ID identifying the client, only for connected calls like "Read".

       $error
           For messages of type "LEAVE" which fail ("$return = -1"), this contains the errno as a
           string, for example "EIO".

       $id The transaction ID, used to correlate actions which are split into two messages
           "ENTER" and "LEAVE".

       $return
           For messages of type "LEAVE" this is the return code, usually 0 for success and "-1"
           if there was an error.

       $type
           The message type: "ENTER", "LEAVE" or "PRINT".

       other shell variables
           Other parameters like "offset=N" are turned into shell variables $offset etc.

       Note the return value of the script is ignored.  Log scripts cannot modify or interrupt
       request processing.

   Log script examples
       The script:

        nbdkit -f --filter=log null 10M \
               logscript='echo $connection $type $id $act $offset >&2'

       might print lines like:

        PRINT Ready
        1 ENTER 1 Read 0x0
        1 ENTER 2 Write 0x200
        1 LEAVE 2 Write
        1 LEAVE 1 Read

       corresponding to log file lines:

        Ready thread_model=3
        connection=1 Read id=1 offset=0x0 count=0x200 ...
        connection=1 Write id=2 offset=0x200 count=0x200 ...
        connection=1 ...Write id=2
        connection=1 ...Read id=1

       This script will trigger a message when any client reads:

        nbdkit -f --filter=log memory 10M \
               logscript='
                   if [ "$act" = "Read" -a "$type" = "ENTER" ]; then
                       echo Client is reading $count bytes from $offset >&2
                   fi
               '

FILES

       "logfile=FILE" parameter
           This filter writes to the file specified by the "logfile=FILE" parameter.

       $filterdir/nbdkit-log-filter.so
           The filter.

           Use "nbdkit --dump-config" to find the location of $filterdir.

VERSION

       "nbdkit-log-filter" first appeared in nbdkit 1.4.

SEE ALSO

       nbdkit(1), nbdkit-file-plugin(1), nbdkit-cow-filter(1), nbdkit-filter(3),
       nbdkit-stats-filter(1).

AUTHORS

       Eric Blake

COPYRIGHT

       Copyright (C) 2018 Red Hat Inc.

LICENSE

       Redistribution and use in source and binary forms, with or without modification, are
       permitted provided that the following conditions are met:

       •   Redistributions of source code must retain the above copyright notice, this list of
           conditions and the following disclaimer.

       •   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.

       •   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.

       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.