FOSSIL(4) FOSSIL(4)
NAME
fossil, flchk, flfmt - archival file server
SYNOPSIS
fossil/fossil [ -Dt ] [ -c cmd ]... [ -f file ] [ -m free-
memory-percent ]
fossil/flchk [ -f ] [ -c ncache ] [ -h host ] file
fossil/flfmt [ -y ] [ -b blocksize ] [ -h host ] [ -l label
] [ -v score ] file
fossil/conf [ -w ] file [ config ]
fossil/last file
DESCRIPTION
Fossil is the main file system for Plan 9. Unlike the Plan
9 file servers of old, fossil is a collection of user-space
programs that run on a standard Plan 9 kernel. The name of
the main fossil file server at Murray Hill is pie. The Plan
9 distribution file server, sources, is also a fossil
server.
Fossil is structured as a magnetic disk write buffer option-
ally backed by a Venti server for archival storage. It
serves the Plan 9 protocol via TCP. A fossil file server
conventionally presents three trees in the root directory of
each file system: active, archive, and snapshot. /active is
the root of a conventional file system whose blocks are
stored in a disk file. In a typical configuration, the file
server periodically marks the entire file system copy-on-
write, effectively taking a snapshot of the file system at
that moment. This snapshot is made available in a name cre-
ated from the date and time of the snapshot:
/snapshot/yyyy/mmdd/hhmm where yyyy is the full year, mm is
the month number, dd is the day number, hh is the hour, and
mm is the minute. The snapshots in /snapshot are ephemeral:
eventually they are deleted to reclaim the disk space they
occupy. Long-lasting snapshots stored on a Venti server are
kept in /archive and also named from the date (though not
the time) of the snapshot: /archive/yyyy/mmdds, where yyyy,
mm, and dd are year, month, and day as before, and s is a
sequence number if more than one archival snapshot is done
in a day. For the first snapshot, s is null. For the sub-
sequent snapshots, s is .1, .2, .3, etc. The root of the
main file system that is frozen for the first archival snap-
shot of December 15, 2002 will be named /archive/2002/1215/.
The attach name used in mount (see bind(1), bind(2) and
Page 1 Plan 9 (printed 12/22/24)
FOSSIL(4) FOSSIL(4)
attach(5)) selects a file system to be served and optionally
a subtree, in the format fs[/dir]. An empty attach name
selects main/active.
Fossil normally requires all users except `none' to provide
authentication tickets on each attach(5). To keep just any-
one from connecting, `none' is only allowed to attach after
another user has successfully attached on the same connec-
tion. The other user effectively acts as a chaperone for
`none'. Authentication can be disabled using the -A flag to
open or srv (see fossilcons(8)).
The groups called noworld and write are special on the file
server. Any user belonging to noworld has attenuated access
privileges. Specifically, when checking such a user's
access to files, the file's permission bits are first ANDed
with 0770 for normal files and 0771 for directories. The
effect is to deny world access permissions to noworld users,
except when walking into directories. If the write group
exists, then the file system appears read-only to users not
in the group. This is used to make the Plan 9 distribution
file server (sources.cs.bell-labs.com) readable by the world
but writable only to the developers.
Fossil starts a new instance of the fossil file server. It
is configured mainly through console commands, documented in
fossilcons(8).
The options are:
-D Toggle the debugging flag, which is initially off.
When the flag is set, information about authentica-
tion and all protocol messages are written to stan-
dard error.
-t Start a file server console on /dev/cons. If this
option is given, fossil does not fork itself into
the background.
-c cmd Execute the console command cmd. This option may be
repeated to give multiple commands. Typically the
only commands given on the command line are ``.
file,'' which executes a file containing commands,
and ``srv -pcons,'' which starts a file server con-
sole on /srv/cons. See fossilcons(8) for more
information.
-f file Read and execute console commands stored in the Fos-
sil disk file. Conf (q.v.) reads and writes the
command set stored in the disk.
Page 2 Plan 9 (printed 12/22/24)
FOSSIL(4) FOSSIL(4)
-m Allocate free-memory-percent percent of the avail-
able free RAM for buffers. This overrides all other
memory sizing parameters, notably the -c option to
open. 30% is a reasonable choice.
Flchk checks the fossil file system stored in file for
inconsistencies. Flchk is deprecated in favor of the con-
sole check command (see fossilcons(8)). Flchk prints fossil
console commands that may be executed to take care of bad
pointers (clrp), bad entries (clre), bad directory entries
(clri), unreachable blocks (bfree). Console commands are
interspersed with more detailed commentary on the file sys-
tem. The commands are distinguished by being prefixed with
sharp signs. Note that all proposed fixes are rather dras-
tic: offending pieces of file system are simply chopped off.
Flchk does not modify the file system, so it is safe to run
concurrently with fossil, though in this case the list of
unreachable blocks and any inconsistencies involving the
active file system should be taken with a grain of salt.
The options are:
-f Fast mode. By default, flchk checks the entire
file system image for consistency, which includes
all the archives to Venti and can take a very long
time. In fast mode, flchk avoids walking in Venti
blocks whenever possible.
-c ncache
Keep a cache of ncache (by default, 1000) file sys-
tem blocks in memory during the check.
-h host Use host as the Venti server.
Flfmt prepares file as a new fossil file system. The file
system is initialized with three empty directories active,
archive, and snapshot, as described above. The options are:
-y Yes mode. By default, flfmt will prompt for
confirmation before formatting a file that
already contains a fossil file system, and
before formatting a file that is not served
directly by a kernel device. If the -y flag
is given, no such checks are made.
-b blocksize Set the file system block size (by default,
8192).
-h host Use host as the Venti server.
Page 3 Plan 9 (printed 12/22/24)
FOSSIL(4) FOSSIL(4)
-l label Set the textual label on the file system to
label. The label is only a comment.
-v score Initialize the file system using the vac file
system stored on Venti at score. The score
should have been generated by fossil rather
than by vac(1), so that the appropriate snap-
shot metadata is present.
Conf reads or writes the configuration branded on the Fossil
disk file. By default, it reads the configuration from the
disk and prints it to standard output. If the -w flag is
given, conf reads a new configuration from config (or else
from standard input) and writes it to the disk. Inside the
configuration file, the argument `*' may be used to stand in
for the name of the disk holding the configuration. The
Plan 9 kernel boot process runs ``fossil -f disk'' to start
a Fossil file server. The disk is just a convenient place
to store configuration information.
Last prints the vac score that resulted after the most
recent archival snapshot of the fossil in file.
EXAMPLES
Place the root of the archive file system on /n/dump and
show the modified times of the MIPS C compiler over all
dumps in December 2002:
9fs dump
ls -l /n/dump/2002/12*/mips/bin/vc
To get only one line of output for each version of the com-
piler:
ls -lp /n/dump/2002/12*/mips/bin/vc | uniq
Initialize a new file system, start the server with permis-
sion checking turned off, create a users file, and mount the
server:
fossil/flfmt /dev/sdC0/fossil
fossil/conf -w /dev/sdC0/fossil <<EOF
fsys main config
fsys main open -AWP
fsys main
create /active/adm adm sys d775
create /active/adm/users adm sys 664
users -w
srv -p fscons
srv fossil
EOF
fossil/fossil -f /dev/sdC0/fossil
Page 4 Plan 9 (printed 12/22/24)
FOSSIL(4) FOSSIL(4)
mount /srv/fossil /n/fossil
See the discussion of the users and uname commands in
fossilcons(8) for more about the user table.
Perhaps because the disk has been corrupted or replaced,
format a new file system using the last archive score
printed on the console:
fossil/flfmt -v b9b3...5559 /dev/sdC0/fossil
Note that while /snapshot will be lost, /active and /archive
will be restored to their contents at the time of the last
archival snapshot.
Blindly accept the changes prescribed by flchk (not recom-
mended):
fossil/flchk /dev/sdC0/fossil | sed -n 's/^# //p' >>/srv/fscons
A better strategy is to vet the output, filter out any sug-
gestions you're not comfortable with, and then use the sed
command to prepare the script.
SOURCE
/usr/local/plan9/src/cmd/fossil
SEE ALSO
yesterday(1), fs(3), fs(4), srv(4), fossilcons(8),
loadfossil(8), venti(8)
BUGS
It is possible that the disk format (but not the Venti for-
mat) will change in the future, to make the disk a full
cache rather than just a write buffer. Changing to the new
format will require reformatting the disk as in the example
above, but note that this will preserve most of the file
system (all but /snapshot) with little effort.
The -m option currently assumes a block size of 8K bytes,
and a single file system per fossil instance.
Page 5 Plan 9 (printed 12/22/24)