mdmon: Document startup/shutdown requirements in man page.

Signed-off-by: NeilBrown <neilb@suse.de>
This commit is contained in:
NeilBrown 2010-02-08 14:43:51 +11:00
parent 319767b85c
commit cd9a8b5cb4
1 changed files with 58 additions and 1 deletions

59
mdmon.8
View File

@ -21,7 +21,7 @@ To service metadata update requests a daemon,
is introduced.
.I Mdmon
is tasked with polling the sysfs namespace looking for changes in
.BR array_state ,
.BR array_state ,
.BR sync_action ,
and per disk
.BR state
@ -169,6 +169,63 @@ RAID arrays. The only times it is run other that by
is when the boot scripts need to restart it after mounting the new
root filesystem.
.SH START UP AND SHUTDOWN
As
.I mdmon
needs to be running whenever any filesystem on the monitored device is
mounted there are special considerations when the root filesystem is
mounted from an
.I mdmon
monitored device.
When the array is assembled by the
.B initramfs
code, mdadm will automatically start
.I mdmon
as required. This means that
.I mdmon
must be installed on the
.B initramfs
and there must be a writable filesystem (typically tmpfs) which
.B mdmon
can create a
.B .pid
and
.B .sock
file on. The particular filesystem to use is given to mdmon at compile
time and defaults to
.BR /lib/init/rw .
This filesystem must persist through to the end of the boot sequence.
After the final root filesystem has be instantiated (usually with
.BR pivot_root )
and after
.B /var
is mounted writable,
.I mdmon
should be run with
.I "\-\-all \-\-takeover"
so that the
.I mdmon
running from the
.B initramfs
can be replaced with one running in the main root.
At shutdown time,
.I mdmon
should not be killed along with other processes. Also as it holds a
file (socket actually) open in
.B /var
it will not be possible to unmount
.B /var
if it is a separate filesystem. Rather the
.B /var
filesystem, like the root filesystem, should be remounted read-only.
.SH EXAMPLES
.B " mdmon \-\-all \-\-takeover"