It's easy to suspect that there might be some underlying hardware
failures or similar issues when userspace receives a Buffer I/O error
from a raid device.
In order to hopefully send more sysadmins on the right track, lets
report that a read failed at least in part due to bad blocks in the bad
block list on device metadata.
There are real world examples where bad block lists accidentally get
propagated or copied around, so having this warning helps mitigate the
consequences
Signed-off-by: Doug V Johnson <dougvj@xxxxxxxxxx>
---
drivers/md/raid5.c | 10 +++++++++-
drivers/md/raid5.h | 2 +-
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 0ae9ac695d8e..5d80e9bcbd6f 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3671,7 +3671,14 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh,
sh->dev[i].sector + RAID5_STRIPE_SECTORS(conf)) {
struct bio *nextbi =
r5_next_bio(conf, bi, sh->dev[i].sector);
-
+ /* If we recorded bad blocks from the metadata
+ * on any of the devices then report this to
+ * userspace in case anyone might suspect
+ * something more fundamental instead
+ */
+ if (s->bad_blocks)
+ pr_warn_ratelimited("%s: read encountered block in device bad block list.",
+ mdname(conf->mddev));
bio_io_error(bi);
bi = nextbi;
}
@@ -4682,6 +4689,7 @@ static void analyse_stripe(struct stripe_head *sh, struct stripe_head_state *s)
if (rdev) {
is_bad = rdev_has_badblock(rdev, sh->sector,
RAID5_STRIPE_SECTORS(conf));
+ s->bad_blocks++;
if (s->blocked_rdev == NULL) {
if (is_bad < 0)
set_bit(BlockedBadBlocks, &rdev->flags);
diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
index eafc6e9ed6ee..c755c321ae36 100644
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -282,7 +282,7 @@ struct stripe_head_state {
* read all devices, just the replacement targets.
*/
int syncing, expanding, expanded, replacing;
- int locked, uptodate, to_read, to_write, failed, written;
+ int locked, uptodate, to_read, to_write, failed, written, bad_blocks;
int to_fill, compute, req_compute, non_overwrite;
int injournal, just_cached;
int failed_num[2];
--
2.48.1
