Commit 05a72772 authored by Lars Ellenberg's avatar Lars Ellenberg Committed by Jens Axboe

drbd: drbdsetup detach of an unresponsive local disk should not block IO "forever"

When detaching, we make sure no application IO is in-flight
by internally suspending IO, then trigger the state change,
wait for the result, and finally internally resume IO again.

Once we triggered the stat change to "Failed",
we expect it to change from Failed to Diskless.
(To avoid races, we actually wait for it to leave "Failed").

On an unresponsive local IO backend, this may not happen, ever.
Don't have a "hung" detach block IO "forever", but resume IO
before waiting for the state change to Diskless.

We may well be able to continue IO to and from a healthy peer.
Signed-off-by: default avatarPhilipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: default avatarLars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: default avatarJens Axboe <axboe@fb.com>
parent 92f108b4
...@@ -1929,9 +1929,9 @@ static int adm_detach(struct drbd_device *device, int force) ...@@ -1929,9 +1929,9 @@ static int adm_detach(struct drbd_device *device, int force)
retcode = drbd_request_state(device, NS(disk, D_FAILED)); retcode = drbd_request_state(device, NS(disk, D_FAILED));
drbd_md_put_buffer(device); drbd_md_put_buffer(device);
/* D_FAILED will transition to DISKLESS. */ /* D_FAILED will transition to DISKLESS. */
drbd_resume_io(device);
ret = wait_event_interruptible(device->misc_wait, ret = wait_event_interruptible(device->misc_wait,
device->state.disk != D_FAILED); device->state.disk != D_FAILED);
drbd_resume_io(device);
if ((int)retcode == (int)SS_IS_DISKLESS) if ((int)retcode == (int)SS_IS_DISKLESS)
retcode = SS_NOTHING_TO_DO; retcode = SS_NOTHING_TO_DO;
if (ret) if (ret)
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment