| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
- MAN[1-9] -> MAN.
|
| |
|
|
|
|
| |
Reviewed by: grog
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
userland tool:
Use the vfs.devfs.generation sysctl to test for devfs presense
(thanks phk!) when devfs is active it will not try to create the
device nodes in /dev and therefore will not complain about the
failure to do so.
Revert the change in the #define for VINUM_DIR in the kernel
header so that vinum can find its device nodes.
Replace perror() with vinum_perror() to print file/line when
DEVBUG is defined (not defined by default).
kernel:
Don't use the #define names for the "superdev" creation since
they will be prepended by "/dev/" (based on VINUM_DIR), instead
use string constants.
Create both debug and non-debug "superdev" nodes in the devfs.
Problem noticed and fix tested by: Martin Blapp <mblapp@fuchur.lan.attic.ch>
|
|
|
|
|
|
|
|
|
| |
Submitted by: Jimmy Olgeni <olgeni@uli.it>
PR: 24233
Fix typos.
Submitted by: "Jeroen C. van Gelderen" <jeroen@vangelderen.org>
|
|
|
|
| |
Reported by: dougb
|
| |
|
|
|
|
|
|
|
|
| |
Correct location of history file.
Change references from /dev/wd to /dev/ad.
Use more appropriate defaults for stripe size in examples.
|
| |
|
|
|
|
|
|
| |
vinum_attach: Get correct names (were being overwritten).
Submitted by: Terry Glanfield <Terry.Glanfield@program-products.co.uk>
|
| |
|
|
|
|
| |
Reported by: Crist J . Clark <cjclark@alum.mit.edu>
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
multi-plex volume.
Confusion reported by: many
Clarify recommendations for default plex stripe size.
|
| |
|
| |
|
|
|
|
| |
Tripped-over-by: Nicole Harrington <nicole@picturetrail.com>
|
|
|
|
| |
Reported by: bde
|
|
|
|
| |
Submitted by: brian
|
|
|
|
|
|
| |
This is still terrible. I need to finally implement variables.
Reported-by: jmg
|
| |
|
| |
|
|
|
|
| |
checkparity: Allow the -v flag to display progress.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
/var/log/vinum_history. Also check that any existing history file is
a regular file.
Admonished-by: imp
Allow stopping long-running commands with ^C (well, SIGINTR).
Previously this would stop the program.
|
|
|
|
|
|
|
|
|
| |
directly off the drives.
This command shows a minor problem: it requires that the device nodes
exist, whereas kernel Vinum does not. Thus dumpconfig may show no
configuration for certain drives which are up and happily running.
This is documented in the man page.
|
|
|
|
| |
option is given. This will break the data in the plex.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Describe -i option for start.
Update description of start command (will now work on plexes).
Update discussion of choice of stripe size (don't use power of 2).
Remove references to raw devices.
|
| |
|
|
|
|
|
|
|
|
|
| |
are read or write.
Bug-introduced-in: Revision 1.26
Show percentage complete in brief status of reviving and initializing
subdisks.
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add -i option to delay between blocks copied, to reduce performance
impact.
Remove -v option, things work now.
Rewrite parity maintenance functions.
Rebuildparity: Add -v option to check whether it was broken before
rebuilding each block.
|
| |
|
|
|
|
| |
Fixed some misformattings.
|
|
|
|
| |
Not tested.
|
|
|
|
|
|
|
|
|
| |
Describe stop -f to stop when config updates are disabled.
Explain in more detail why Vinum needs its own partition type, and why
you can't make it the c partition.
Implicitly-sanctioned-by: jkh
|
|
|
|
| |
Implicitly-sanctioned-by: jkh
|
|
|
|
|
|
| |
be marked as stalled, from 1 second to 5 seconds.
Implicitly-sanctioned-by: jkh
|
|
|
|
|
|
|
| |
process, move the body of vinum_rebuildparity and vinum_checkparity to
a new function, parityops.
Implicitly-sanctioned-by: jkh
|
|
|
|
| |
Approved-by: jkh
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
replaces an older attempt to silence vinum(8) when started in
single-user mode.
Add entries for vinum_raid[45].
Replace the preprocessor variable name CDEV_MAJOR with
VINUM_CDEV_MAJOR.
continue_revive: Set the proc title to 'reviving <sdname>' so that it
can be recognized in a ps list.
Approved-by: jkh
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
drive if it is set.
vinum_lvi: Change line spacing. This is still not right.
Change checks for striped or RAID-5 plexes to a macro 'isstriped',
which now also includes RAID-4.
Change checks for RAID-5 plexes to a macro 'isparity', which now also
includes RAID-4.
vinum_lsi: If the subdisk is initializing or reviving, don't wait a
whole second to check whether it's advancing; wait in increments of 50
ms up to 1 second. This makes for a barely perceptible delay.
Approved-by: jkh
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
enabled and print a warning message in varous places if they are not.
Fix typos in comments.
Change some indents to approach style(9).
initvol: If init fails, place the subdisks in a "down" state.
vinum_start(): If we're starting a plex, try doing it directly. It's
possible that the state of the plex differs from that of the subdisks,
for example if somebody has used 'setupstate'.
New functions vinum_raid4 and vinum_raid5 to create RAID-4 and RAID-5
volumes, analagous to vinum_mirror and vinum_stripe.
vinum_checkparity: Don't try to check the parity of a non-parity plex,
print a rude remark instead.
Approved-by: jkh
|
|
|
|
|
|
|
| |
Without this change, you cannot do the required init on a new raid5 volume.
Reviewed by: grog
Approved by: jkh
|
|
|
|
| |
Pointed-out-by: "Rodney W. Grimes" <freebsd@gndrsh.dnsmgr.net>
|
|
|
|
|
|
|
| |
Submitted-by: Marius Bendiksen <marius@marius.scancall.no>
Describe state transitions in more detail, in particular how to revive
subdisks.
|
|
|
|
| |
Add declaration for vinum_mv.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Suggested-by: Bernd Walter (ticso@cicely.de)
Add key pairs for 'mv' and 'move' (a synonym for 'mv'). Required for
the move command code submitted by Marius Bendiksen
<marius@marius.scancall.no>
make_devices: Don't create symlinks for drives if they are only
referenced. Previously, spurious symlinks appeared in the current
directory.
Problem-reported-by: Bernd Walter (ticso@cicely.de)
No longer create character devices, now that there is no difference.
Make the devices as character devices, not block devices.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
on alpha.
Submitted-by: Bernd Walter <ticso@cicely.de>
Replace %q formats with %lld.
Desired-by: bde
Remove #ifdef RAID5
vinum_lsi: If a subdisk is in 'reviving' state, check the status of
the revive and report the status of the revive. In verbose mode,
display the pid of the reviver.
|