On Sun, Nov 8, 2020 at 6:53 AM Zhao Heming <heming.zhao@xxxxxxxx> wrote:
>
[...]
> How to fix:
>
> The simple & clear solution is block the reshape action in initiator
> side. When node is executing "--grow" and detecting there is ongoing
> resyncing, it should immediately return & report error to user space.
>
> Signed-off-by: Zhao Heming <heming.zhao@xxxxxxxx>
The code looks good to me. But please revise the commit log as something
similar to the following.
========================== 8< ==========================
md/cluster: block reshape requests with resync job initiated from remote node
In cluster env, a node can start resync job when the resync cmd was executed
on a different node. Reshape requests should be blocked for resync job initiated
by any node. Current code only <condition to block reshape requests>.
This results
in a dead lock in <condition> (see repro steps below). Fix this by <adding the
extra check>.
Repro steps:
...
========================== 8< ==========================
In this way, whoever reading the commit log, which could be yourself in 2021,
will understand the primary goal of this change quickly.
Does this make sense?
Thanks,
Song
> ---
> drivers/md/md.c | 8 ++++++--
> 1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/md/md.c b/drivers/md/md.c
> index 98bac4f304ae..74280e353b8f 100644
> --- a/drivers/md/md.c
> +++ b/drivers/md/md.c
> @@ -7278,6 +7278,7 @@ static int update_raid_disks(struct mddev *mddev, int raid_disks)
> return -EINVAL;
> if (mddev->sync_thread ||
> test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) ||
> + test_bit(MD_RESYNCING_REMOTE, &mddev->recovery) ||
> mddev->reshape_position != MaxSector)
> return -EBUSY;
>
> @@ -9662,8 +9663,11 @@ static void check_sb_changes(struct mddev *mddev, struct md_rdev *rdev)
> }
> }
>
> - if (mddev->raid_disks != le32_to_cpu(sb->raid_disks))
> - update_raid_disks(mddev, le32_to_cpu(sb->raid_disks));
> + if (mddev->raid_disks != le32_to_cpu(sb->raid_disks)) {
> + ret = update_raid_disks(mddev, le32_to_cpu(sb->raid_disks));
> + if (ret)
> + pr_warn("md: updating array disks failed. %d\n", ret);
> + }
>
> /*
> * Since mddev->delta_disks has already updated in update_raid_disks,
> --
> 2.27.0
>
