For non growing migration case, there are several errors in checkpointing:
1. Write offset was computed incorrectly based on old disk number.
This causes checkpointing to stop working
(suspend_lo value was wrong calculated and cannot be saved in sysfs).
2. Checkpointing cannot start due to wrong sync_completed value reported by md
on reshape start. md reports value 2 instead 0.
3. Layout value for raid0 causes that geo_map() returns error.
Signed-off-by: Adam Kwolek <adam.kwolek@xxxxxxxxx>
---
mdadm/mdadm/Grow.c | 2 +-
mdadm/mdadm/monitor.c | 8 +++++++-
mdadm/mdadm/restripe.c | 1 +
3 files changed, 9 insertions(+), 2 deletions(-)
diff --git a/mdadm/mdadm/Grow.c b/mdadm/mdadm/Grow.c
index f56613f..3747f76 100644
--- a/mdadm/mdadm/Grow.c
+++ b/mdadm/mdadm/Grow.c
@@ -2280,7 +2280,7 @@ int grow_backup_ext(struct supertype *st, struct mdinfo *sra,
if (sra->new_level == 6)
new_odata--;
- write_offset = offset * 512 * odata;
+ write_offset = offset * 512 * new_odata;
bytes_per_unit = sra->new_chunk * new_odata;
if (chunk > sra->new_chunk)
bytes_per_unit *= (chunk / sra->new_chunk);
diff --git a/mdadm/mdadm/monitor.c b/mdadm/mdadm/monitor.c
index 4cd634d..a37cd93 100644
--- a/mdadm/mdadm/monitor.c
+++ b/mdadm/mdadm/monitor.c
@@ -248,8 +248,14 @@ static int read_and_act(struct active_array *a)
if (safe_sync_max >= a->info.component_size)
sysfs_set_str(&a->info, NULL, "sync_max", "max");
- else
+ else {
+ /* Workarround:
+ * sometimes md reports sync_completed == 2 but in fact it is 0
+ */
+ if ((new_sync_completed == 2) && (safe_sync_max == 0))
+ safe_sync_max = 2;
sysfs_set_num(&a->info, NULL, "sync_max", safe_sync_max);
+ }
} else {
sysfs_set_num(&a->info, NULL, "sync_max", a->grow_sync_max);
}
diff --git a/mdadm/mdadm/restripe.c b/mdadm/mdadm/restripe.c
index 7add7ce..3b16ad8 100644
--- a/mdadm/mdadm/restripe.c
+++ b/mdadm/mdadm/restripe.c
@@ -45,6 +45,7 @@ static int geo_map(int block, unsigned long long stripe, int raid_disks,
switch(level*100 + layout) {
case 000:
+ case 000 + ALGORITHM_PARITY_N: /* layout has no matter for raid0 */
case 400:
case 400 + ALGORITHM_PARITY_N:
case 500 + ALGORITHM_PARITY_N:
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
