Techwizz/kernel-fxtec-pro1x
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

drbd: fix hang on local read errors while disconnected

Browse source 
"canceled" w_read_retry_remote never completed, if they have been
canceled after drbd_disconnect connection teardown cleanup has already
run (or we are currently not connected anyways).

Fixed by not queueing a remote retry if we already know it won't work
(pdsk not uptodate), and cleanup ourselves on "cancel", in case we hit a
race with drbd_disconnect.

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
This commit is contained in:
Lars Ellenberg 2010-05-27 09:45:45 +02:00 committed by Jens Axboe
parent 32fa7e91f9
commit d255e5ff5f
3 changed files with 16 additions and 13 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

22
drivers/block/drbd/drbd_req.c
View file

@ -452,20 +452,21 @@ void __req_mod(struct drbd_request *req, enum drbd_req_event what,
dev_alert(DEV, "Local READ failed sec=%llus size=%u\n",
(unsigned long long)req->sector, req->size);
/* _req_mod(req,to_be_send); oops, recursion... */
D_ASSERT(!(req->rq_state & RQ_NET_MASK));
req->rq_state |= RQ_NET_PENDING;
inc_ap_pending(mdev);
__drbd_chk_io_error(mdev, FALSE);
put_ldev(mdev);
/* NOTE: if we have no connection,
* or know the peer has no good data either,
* then we don't actually need to "queue_for_net_read",
* but we do so anyways, since the drbd_io_error()
* and the potential state change to "Diskless"
* needs to be done from process context */
/* no point in retrying if there is no good remote data,
* or we have no connection. */
if (mdev->state.pdsk != D_UP_TO_DATE) {
_req_may_be_done(req, m);
break;
}
/* _req_mod(req,to_be_send); oops, recursion... */
req->rq_state |= RQ_NET_PENDING;
inc_ap_pending(mdev);
/* fall through: _req_mod(req,queue_for_net_read); */
case queue_for_net_read:
@ -575,6 +576,9 @@ void __req_mod(struct drbd_request *req, enum drbd_req_event what,
_req_may_be_done(req, m);
break;
case read_retry_remote_canceled:
req->rq_state &= ~RQ_NET_QUEUED;
/* fall through, in case we raced with drbd_disconnect */
case connection_lost_while_pending:
/* transfer log cleanup after connection loss */
/* assert something? */

1
drivers/block/drbd/drbd_req.h
View file

@ -91,6 +91,7 @@ enum drbd_req_event {
send_failed,
handed_over_to_network,
connection_lost_while_pending,
read_retry_remote_canceled,
recv_acked_by_peer,
write_acked_by_peer,
write_acked_by_peer_and_sis, /* and set_in_sync */

6
drivers/block/drbd/drbd_worker.c
View file

@ -266,10 +266,8 @@ int w_read_retry_remote(struct drbd_conf *mdev, struct drbd_work *w, int cancel)
* to give the disk the chance to relocate that block */
spin_lock_irq(&mdev->req_lock);
if (cancel ||
mdev->state.conn < C_CONNECTED ||
mdev->state.pdsk <= D_INCONSISTENT) {
_req_mod(req, send_canceled);
if (cancel || mdev->state.pdsk != D_UP_TO_DATE) {
_req_mod(req, read_retry_remote_canceled);
spin_unlock_irq(&mdev->req_lock);
dev_alert(DEV, "WE ARE LOST. Local IO failure, no peer.\n");
return 1;