<screen>&prompt.root; <userinput>tunefs -n enable /filesystem</userinput>

      <screen>&prompt.root; <userinput>tunefs -n enable /filesystem</userinput>

&prompt.root; <userinput>tunefs -n disable /filesystem</userinput></screen>

&prompt.root; <userinput>tunefs -n disable /filesystem</userinput></screen>

	it is mounted.  A good time to enable Soft Updates is before any

	it is mounted.  A good time to enable Soft Updates is before any

	partitions have been mounted, in single-user mode.</para>

	partitions have been mounted, in single-user mode.</para>

        file creation and deletion, through the use of a memory cache.  We

        file creation and deletion, through the use of a memory cache.  We

        recommend to use Soft Updates on all of your file systems.  There

        recommend to use Soft Updates on all of your file systems.  There

        are two downsides to Soft Updates that you should be aware of:  First,

        are two downsides to Soft Updates that you should be aware of:  First,

        but could very easily be several seconds (even a minute!) behind

        but could very easily be several seconds (even a minute!) behind

        updating the physical disk.  If your system crashes you may lose more

        updating the physical disk.  If your system crashes you may lose more

        work than otherwise.  Secondly, Soft Updates delays the freeing of

        work than otherwise.  Secondly, Soft Updates delays the freeing of

	out of space and the update to fail.</para>

	out of space and the update to fail.</para>

      <sect3>

      <sect3>

	  or not at all.  If the data blocks of a file did not find

	  or not at all.  If the data blocks of a file did not find

	  their way out of the buffer cache onto the disk by the time

	  their way out of the buffer cache onto the disk by the time

	  of the crash, &man.fsck.8; is able to recognize this and

	  of the crash, &man.fsck.8; is able to recognize this and

	  0.  Additionally, the implementation is clear and simple.

	  0.  Additionally, the implementation is clear and simple.

	  The disadvantage is that meta-data changes are slow.  An

	  The disadvantage is that meta-data changes are slow.  An

	  <command>rm -r</command>, for instance, touches all the files

	  <command>rm -r</command>, for instance, touches all the files

  	  implementation is still clear and simple, so there is a low

  	  implementation is still clear and simple, so there is a low

  	  risk for bugs creeping into the code.  The disadvantage is

  	  risk for bugs creeping into the code.  The disadvantage is

  	  that there is no guarantee at all for a consistent state of

  	  that there is no guarantee at all for a consistent state of

  	  that updated large amounts of meta-data (like a power

  	  that updated large amounts of meta-data (like a power

  	  failure, or someone pressing the reset button),

  	  failure, or someone pressing the reset button),

  	  will be left in an unpredictable state.  There is no opportunity

  	  will be left in an unpredictable state.  There is no opportunity

  	  comes up again; the data blocks of a file could already have

  	  comes up again; the data blocks of a file could already have

  	  been written to the disk while the updates of the inode

  	  been written to the disk while the updates of the inode

  	  table or the associated directory were not.  It is actually

  	  table or the associated directory were not.  It is actually

  	  impossible to implement a <command>fsck</command> which is

  	  impossible to implement a <command>fsck</command> which is

  	  able to clean up the resulting chaos (because the necessary

  	  able to clean up the resulting chaos (because the necessary

  	  information is not available on the disk).  If the

  	  information is not available on the disk).  If the

	  is to use &man.newfs.8; on it and restore it from backup.

	  is to use &man.newfs.8; on it and restore it from backup.

	  </para>

	  </para>

	  might result.  On the other hand, in case of a crash, all

	  might result.  On the other hand, in case of a crash, all

	  pending meta-data operations can be quickly either rolled-back

	  pending meta-data operations can be quickly either rolled-back

	  or completed from the logging area after the system comes

	  or completed from the logging area after the system comes

	<para>Kirk McKusick, the developer of Berkeley FFS,

	<para>Kirk McKusick, the developer of Berkeley FFS,

	   solved this problem with Soft Updates: all pending

	   solved this problem with Soft Updates: all pending

	   If the system crashes, this causes an implicit <quote>log

	   If the system crashes, this causes an implicit <quote>log

	   rewind</quote>: all operations which did not find their way

	   rewind</quote>: all operations which did not find their way

	   to the disk appear as if they had never happened.  A

	   to the disk appear as if they had never happened.  A

	   be the one of 30 to 60 seconds earlier.  The

	   be the one of 30 to 60 seconds earlier.  The

	   algorithm used guarantees that all resources in use

	   algorithm used guarantees that all resources in use

	   are marked as such in their appropriate bitmaps: blocks and inodes.

	   are marked as such in their appropriate bitmaps: blocks and inodes.

	   marked as <quote>used</quote> which are actually <quote>free</quote>.

	   marked as <quote>used</quote> which are actually <quote>free</quote>.

	   &man.fsck.8; recognizes this situation,

	   &man.fsck.8; recognizes this situation,

	   and frees the resources that are no longer used.  It is safe to

	   and frees the resources that are no longer used.  It is safe to

	   forcibly mounting it with <command>mount -f</command>.  In

	   forcibly mounting it with <command>mount -f</command>.  In

	   order to free resources that may be unused, &man.fsck.8;

	   order to free resources that may be unused, &man.fsck.8;

	   needs to be run at a later time.  This is the idea behind

	   needs to be run at a later time.  This is the idea behind

	   the <emphasis>background fsck</emphasis>: at system startup

	   the <emphasis>background fsck</emphasis>: at system startup

	   time, only a <emphasis>snapshot</emphasis> of the

	   time, only a <emphasis>snapshot</emphasis> of the

	   run later on.  All file systems can then be mounted

	   run later on.  All file systems can then be mounted

	   <quote>dirty</quote>, so the system startup proceeds in

	   <quote>dirty</quote>, so the system startup proceeds in

	   multiuser mode.  Then, background <command>fsck</command>s

	   multiuser mode.  Then, background <command>fsck</command>s

	   is highly sensitive regarding loss of user data), and a

	   is highly sensitive regarding loss of user data), and a

	   higher memory consumption.  Additionally there are some

	   higher memory consumption.  Additionally there are some

	   idiosyncrasies one has to get used to.

	   idiosyncrasies one has to get used to.

	   somewhat <quote>older</quote>.  In situations where

	   somewhat <quote>older</quote>.  In situations where

	   the standard synchronous approach would have caused some

	   the standard synchronous approach would have caused some

	   zero-length files to remain after the

	   zero-length files to remain after the

	   <command>fsck</command>, these files do not exist at all

	   <command>fsck</command>, these files do not exist at all

	   nor the file contents have ever been written to disk.

	   nor the file contents have ever been written to disk.

	   Disk space is not released until the updates have been

	   Disk space is not released until the updates have been

	   written to disk, which may take place some time after

	   written to disk, which may take place some time after

	   running <command>rm</command>.  This may cause problems

	   running <command>rm</command>.  This may cause problems

	   that does not have enough free space to hold all the files

	   that does not have enough free space to hold all the files

	   twice.</para>

	   twice.</para>

      </sect3>

      </sect3>