zfsd(8), the ZFS fault management daemon

Add zfsd, which deals with hard drive faults in ZFS pools. It manages
hotspares and replements in drive slots that publish physical paths.

cddl/usr.sbin/zfsd
	Add zfsd(8) and its unit tests

cddl/usr.sbin/Makefile
	Add zfsd to the build

lib/libdevdctl
	A C++ library that helps devd clients process events

lib/Makefile
share/mk/bsd.libnames.mk
share/mk/src.libnames.mk
	Add libdevdctl to the build. It's a private library, unusable by
	out-of-tree software.

etc/defaults/rc.conf
	By default, set zfsd_enable to NO

etc/mtree/BSD.include.dist
	Add a directory for libdevdctl's include files

etc/mtree/BSD.tests.dist
	Add a directory for zfsd's unit tests

etc/mtree/BSD.var.dist
	Add /var/db/zfsd/cases, where zfsd stores case files while it's shut
	down.

etc/rc.d/Makefile
etc/rc.d/zfsd
	Add zfsd's rc script

sys/cddl/contrib/opensolaris/uts/common/fs/zfs/vdev.c
	Fix the resource.fs.zfs.statechange message. It had a number of
	problems:

	It was only being emitted on a transition to the HEALTHY state.
	That made it impossible for zfsd to take actions based on drives
	getting sicker.

	It compared the new state to vdev_prevstate, which is the state that
	the vdev had the last time it was opened.  That doesn't make sense,
	because a vdev can change state multiple times without being
	reopened.

	vdev_set_state contains logic that will change the device's new
	state based on various conditions.  However, the statechange event
	was being posted _before_ that logic took effect.  Now it's being
	posted after.

Submitted by:	gibbs, asomers, mav, allanjude
Reviewed by:	mav, delphij
Relnotes:	yes
Sponsored by:	Spectra Logic Corp, iX Systems
Differential Revision:	https://reviews.freebsd.org/D6564

1 files changed, 10 insertions, 13 deletions

diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/vdev.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/vdev.c
index 9b08e35..17e2f02 100644
--- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/vdev.c
+++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/vdev.c
@@ -3366,19 +3366,6 @@ vdev_set_state(vdev_t *vd, boolean_t isopen, vdev_state_t state, vdev_aux_t aux)
 	    vd->vdev_ops->vdev_op_leaf)
 		vd->vdev_ops->vdev_op_close(vd);
 
-	/*
-	 * If we have brought this vdev back into service, we need
-	 * to notify fmd so that it can gracefully repair any outstanding
-	 * cases due to a missing device.  We do this in all cases, even those
-	 * that probably don't correlate to a repaired fault.  This is sure to
-	 * catch all cases, and we let the zfs-retire agent sort it out.  If
-	 * this is a transient state it's OK, as the retire agent will
-	 * double-check the state of the vdev before repairing it.
-	 */
-	if (state == VDEV_STATE_HEALTHY && vd->vdev_ops->vdev_op_leaf &&
-	    vd->vdev_prevstate != state)
-		zfs_post_state_change(spa, vd);
-
 	if (vd->vdev_removed &&
 	    state == VDEV_STATE_CANT_OPEN &&
 	    (aux == VDEV_AUX_OPEN_FAILED || vd->vdev_checkremove)) {
@@ -3459,6 +3446,16 @@ vdev_set_state(vdev_t *vd, boolean_t isopen, vdev_state_t state, vdev_aux_t aux)
 		vd->vdev_removed = B_FALSE;
 	}
 
+	/*
+	* Notify the fmd of the state change.  Be verbose and post
+	* notifications even for stuff that's not important; the fmd agent can
+	* sort it out.  Don't emit state change events for non-leaf vdevs since
+	* they can't change state on their own.  The FMD can check their state
+	* if it wants to when it sees that a leaf vdev had a state change.
+	*/
+	if (vd->vdev_ops->vdev_op_leaf)
+		zfs_post_state_change(spa, vd);
+
 	if (!isopen && vd->vdev_parent)
 		vdev_propagate_state(vd->vdev_parent);
 }