]> git.wh0rd.org - dump.git/blobdiff - dump/dump.8.in
Added -F script option to restore.
[dump.git] / dump / dump.8.in
index 5d16fc77d53182fc3d6152c1b85cea63ecf0ecce..5a1dc751b9a7a72758e70fcb924ee4e928ce3e3d 100644 (file)
@@ -30,7 +30,7 @@
 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 .\" SUCH DAMAGE.
 .\"
-.\"    $Id: dump.8.in,v 1.26 2001/04/10 13:42:22 stelian Exp $
+.\"    $Id: dump.8.in,v 1.28 2001/07/18 12:54:06 stelian Exp $
 .\"
 .Dd __DATE__
 .Dt DUMP 8
@@ -117,7 +117,8 @@ The default level is 9.
 .It Fl B Ar records
 The number of 1 kB blocks per volume.
 This option overrides the end-of-media detection, and calculation 
-of tape size based on length and density.
+of tape size based on length and density. If compression is on this
+limits the size of the compressed output per volume.
 .It Fl a
 .Dq auto-size .
 Bypass all tape length calculations, and write
@@ -201,7 +202,7 @@ The script must return 0 if
 .Nm
 should continue without asking the user to change the tape, 1 if 
 .Nm
-dump should continue but ask the user to change the tape. 
+should continue but ask the user to change the tape. 
 Any other exit code will cause
 .Nm
 to abort.
@@ -483,6 +484,8 @@ Each reel requires a new process, so parent processes for
 reels already written just hang around until the entire tape
 is written.
 .Pp
+The estimated number of tapes is not correct if compression is on.
+.Pp
 It would be nice if
 .Nm
 knew about the dump sequence,