Multiple DVA Scrubbing Fix

Currently, there is an issue in the sequential scrub code which
prevents self healing from working in some cases. The scrub code
will split up all DVA copies of a bp and issue each of them
separately. The problem is that, since each of the DVAs is no
longer associated with the others, the self healing code doesn't
have the opportunity to repair problems that show up in one of the
DVAs with the data from the others.

This patch fixes this issue by ensuring that all IOs issued by the
sequential scrub code include all DVAs. Initially, only the first
DVA of each is attempted. If an issue arises, the IO is retried
with all available copies, giving the self healing code a chance
to correct the issue.

To test this change, this patch also adds the ability for zinject
to specify individual DVAs to inject read errors into. We then
add a new test case that utilizes this functionality to ensure
scrubs and self-healing reads can handle and transparently fix
issues with individual copies of blocks.

Reviewed-by: Brian Behlendorf <[email protected]>
Reviewed-by: Matt Ahrens <[email protected]>
Signed-off-by: Tom Caputi <[email protected]>
Closes #8453 

1 files changed, 39 insertions, 6 deletions

diff --git a/module/zfs/zio_inject.c b/module/zfs/zio_inject.c
index 7a7401ecd..78896d3dc 100644
--- a/module/zfs/zio_inject.c
+++ b/module/zfs/zio_inject.c
@@ -124,7 +124,7 @@ freq_triggered(uint32_t frequency)
  * Returns true if the given record matches the I/O in progress.
  */
 static boolean_t
-zio_match_handler(const zbookmark_phys_t *zb, uint64_t type,
+zio_match_handler(const zbookmark_phys_t *zb, uint64_t type, int dva,
     zinject_record_t *record, int error)
 {
 	/*
@@ -148,8 +148,10 @@ zio_match_handler(const zbookmark_phys_t *zb, uint64_t type,
 	    zb->zb_level == record->zi_level &&
 	    zb->zb_blkid >= record->zi_start &&
 	    zb->zb_blkid <= record->zi_end &&
-	    error == record->zi_error)
+	    (record->zi_dvas == 0 || (record->zi_dvas & (1ULL << dva))) &&
+	    error == record->zi_error) {
 		return (freq_triggered(record->zi_freq));
+	}
 
 	return (B_FALSE);
 }
@@ -199,7 +201,8 @@ zio_handle_decrypt_injection(spa_t *spa, const zbookmark_phys_t *zb,
 		    handler->zi_record.zi_cmd != ZINJECT_DECRYPT_FAULT)
 			continue;
 
-		if (zio_match_handler(zb, type, &handler->zi_record, error)) {
+		if (zio_match_handler(zb, type, ZI_NO_DVA,
+		    &handler->zi_record, error)) {
 			ret = error;
 			break;
 		}
@@ -210,6 +213,37 @@ zio_handle_decrypt_injection(spa_t *spa, const zbookmark_phys_t *zb,
 }
 
 /*
+ * If this is a physical I/O for a vdev child determine which DVA it is
+ * for. We iterate backwards through the DVAs matching on the offset so
+ * that we end up with ZI_NO_DVA (-1) if we don't find a match.
+ */
+static int
+zio_match_dva(zio_t *zio)
+{
+	int i = ZI_NO_DVA;
+
+	if (zio->io_bp != NULL && zio->io_vd != NULL &&
+	    zio->io_child_type == ZIO_CHILD_VDEV) {
+		for (i = BP_GET_NDVAS(zio->io_bp) - 1; i >= 0; i--) {
+			dva_t *dva = &zio->io_bp->blk_dva[i];
+			uint64_t off = DVA_GET_OFFSET(dva);
+			vdev_t *vd = vdev_lookup_top(zio->io_spa,
+			    DVA_GET_VDEV(dva));
+
+			/* Compensate for vdev label added to leaves */
+			if (zio->io_vd->vdev_ops->vdev_op_leaf)
+				off += VDEV_LABEL_START_SIZE;
+
+			if (zio->io_vd == vd && zio->io_offset == off)
+				break;
+		}
+	}
+
+	return (i);
+}
+
+
+/*
  * Determine if the I/O in question should return failure.  Returns the errno
  * to be returned to the caller.
  */
@@ -235,15 +269,14 @@ zio_handle_fault_injection(zio_t *zio, int error)
 
 	for (handler = list_head(&inject_handlers); handler != NULL;
 	    handler = list_next(&inject_handlers, handler)) {
-
 		if (zio->io_spa != handler->zi_spa ||
 		    handler->zi_record.zi_cmd != ZINJECT_DATA_FAULT)
 			continue;
 
-		/* If this handler matches, return EIO */
+		/* If this handler matches, return the specified error */
 		if (zio_match_handler(&zio->io_logical->io_bookmark,
 		    zio->io_bp ? BP_GET_TYPE(zio->io_bp) : DMU_OT_NONE,
-		    &handler->zi_record, error)) {
+		    zio_match_dva(zio), &handler->zi_record, error)) {
 			ret = error;
 			break;
 		}