]> git.baikalelectronics.ru Git - kernel.git/commitdiff
md/raid10: print more useful messages on device failure.
authorRobert Becker <Rob.Becker@riverbed.com>
Mon, 14 Dec 2009 01:49:57 +0000 (12:49 +1100)
committerNeilBrown <neilb@suse.de>
Mon, 14 Dec 2009 01:51:41 +0000 (12:51 +1100)
When we get a read error on a device in a RAID10, and attempting to
repair the error fails, print more useful messages about why it
failed.

Signed-off-by: Robert Becker <Rob.Becker@riverbed.com>
Signed-off-by: NeilBrown <neilb@suse.de>
drivers/md/raid10.c

index d9e28a65af58cdc5b2d394f1d1b5debf040477f1..670449f7411fcef371aee08ddab50e608a826ae7 100644 (file)
@@ -1488,6 +1488,7 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
                /* write it back and re-read */
                rcu_read_lock();
                while (sl != r10_bio->read_slot) {
+                       char b[BDEVNAME_SIZE];
                        int d;
                        if (sl==0)
                                sl = conf->copies;
@@ -1503,9 +1504,21 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
                                                 r10_bio->devs[sl].addr +
                                                 sect + rdev->data_offset,
                                                 s<<9, conf->tmppage, WRITE)
-                                   == 0)
+                                   == 0) {
                                        /* Well, this device is dead */
+                                       printk(KERN_NOTICE
+                                              "raid10:%s: read correction "
+                                              "write failed"
+                                              " (%d sectors at %llu on %s)\n",
+                                              mdname(mddev), s,
+                                              (unsigned long long)(sect+
+                                              rdev->data_offset),
+                                              bdevname(rdev->bdev, b));
+                                       printk(KERN_NOTICE "raid10:%s: failing "
+                                              "drive\n",
+                                              bdevname(rdev->bdev, b));
                                        md_error(mddev, rdev);
+                               }
                                rdev_dec_pending(rdev, mddev);
                                rcu_read_lock();
                        }
@@ -1526,10 +1539,22 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
                                if (sync_page_io(rdev->bdev,
                                                 r10_bio->devs[sl].addr +
                                                 sect + rdev->data_offset,
-                                                s<<9, conf->tmppage, READ) == 0)
+                                                s<<9, conf->tmppage,
+                                                READ) == 0) {
                                        /* Well, this device is dead */
+                                       printk(KERN_NOTICE
+                                              "raid10:%s: unable to read back "
+                                              "corrected sectors"
+                                              " (%d sectors at %llu on %s)\n",
+                                              mdname(mddev), s,
+                                              (unsigned long long)(sect+
+                                                   rdev->data_offset),
+                                              bdevname(rdev->bdev, b));
+                                       printk(KERN_NOTICE "raid10:%s: failing drive\n",
+                                              bdevname(rdev->bdev, b));
+
                                        md_error(mddev, rdev);
-                               else
+                               } else {
                                        printk(KERN_INFO
                                               "raid10:%s: read error corrected"
                                               " (%d sectors at %llu on %s)\n",
@@ -1537,6 +1562,7 @@ static void fix_read_error(conf_t *conf, mddev_t *mddev, r10bio_t *r10_bio)
                                               (unsigned long long)(sect+
                                                    rdev->data_offset),
                                               bdevname(rdev->bdev, b));
+                               }
 
                                rdev_dec_pending(rdev, mddev);
                                rcu_read_lock();