diff mbox series

[v5] I/O errors for ALUA state transitions

Message ID 20240514140344.19538-1-mwilck@suse.com (mailing list archive)
State Accepted
Headers show
Series [v5] I/O errors for ALUA state transitions | expand

Commit Message

Martin Wilck May 14, 2024, 2:03 p.m. UTC 
When a host is configured with a few LUNs and IO is running,
injecting FC faults repeatedly leads to path recovery problems.
The LUNs have 4 paths each and 3 of them come back active after
say an FC fault which makes two of the paths go down, instead of
all 4. This happens after several iterations of continuous FC faults.

Reason here is that we're returning an I/O error whenever we're
encountering sense code 06/04/0a (LOGICAL UNIT NOT ACCESSIBLE,
ASYMMETRIC ACCESS STATE TRANSITION) instead of retrying.

mwilck: The original patch was developed by Rajashekhar M A and Hannes
Reinecke. I moved the code to alua_check_sense() as suggested by
Mike Christie [1]. Evan Milne had raised the question whether pg->state
should be set to transitioning in the UA case [2]. I believe that doing
this is correct. SCSI_ACCESS_STATE_TRANSITIONING by itself doesn't cause
I/O errors. Our handler schedules an RTPG, which will only result in
an I/O error condition if the transitioning timeout expires.

[1] https://lore.kernel.org/all/0bc96e82-fdda-4187-148d-5b34f81d4942@oracle.com/
[2] https://lore.kernel.org/all/CAGtn9r=kicnTDE2o7Gt5Y=yoidHYD7tG8XdMHEBJTBraVEoOCw@mail.gmail.com/

Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Martin Wilck <mwilck@suse.com>
Co-authored-by: Rajashekhar M A <rajs@netapp.com>
Reviewed-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Mike Christie <michael.christie@oracle.com>
---
Changes v4->v5:
- changed patch author to myself per discussion with Martin K. Petersen,
  as we can't get Rajashekhar's S-o-b tag easily. Changed commit message
  accordingly. No code changes.
Changes v3->v4:
- fix a whitespace error (Damien Le Moal)
Changes v2->v3:
- drop return value of alua_handle_state_transition() (Christoph Hellwig)
- handle UNIT ATTENTION in alua_tur(), too (Mike Christie)
- restore comment in alua_check_sense() (Damien Le Moal)
---
 drivers/scsi/device_handler/scsi_dh_alua.c | 31 +++++++++++++++-------
 1 file changed, 22 insertions(+), 9 deletions(-)

Comments

Martin K. Petersen May 15, 2024, 1:46 p.m. UTC | #1 
Martin,

> When a host is configured with a few LUNs and IO is running, injecting
> FC faults repeatedly leads to path recovery problems. The LUNs have 4
> paths each and 3 of them come back active after say an FC fault which
> makes two of the paths go down, instead of all 4. This happens after
> several iterations of continuous FC faults.

Applied to 6.10/scsi-staging, thanks!
Martin K. Petersen May 21, 2024, 2:40 a.m. UTC | #2 
On Tue, 14 May 2024 16:03:44 +0200, Martin Wilck wrote:

> When a host is configured with a few LUNs and IO is running,
> injecting FC faults repeatedly leads to path recovery problems.
> The LUNs have 4 paths each and 3 of them come back active after
> say an FC fault which makes two of the paths go down, instead of
> all 4. This happens after several iterations of continuous FC faults.
> 
> Reason here is that we're returning an I/O error whenever we're
> encountering sense code 06/04/0a (LOGICAL UNIT NOT ACCESSIBLE,
> ASYMMETRIC ACCESS STATE TRANSITION) instead of retrying.
> 
> [...]

Applied to 6.10/scsi-queue, thanks!

[1/1] I/O errors for ALUA state transitions
      https://git.kernel.org/mkp/scsi/c/10157b1fc1a7
diff mbox series

Patch

diff --git a/drivers/scsi/device_handler/scsi_dh_alua.c b/drivers/scsi/device_handler/scsi_dh_alua.c
index a226dc1b65d7..4eb0837298d4 100644
--- a/drivers/scsi/device_handler/scsi_dh_alua.c
+++ b/drivers/scsi/device_handler/scsi_dh_alua.c
@@ -414,28 +414,40 @@  static char print_alua_state(unsigned char state)
 	}
 }
 
-static enum scsi_disposition alua_check_sense(struct scsi_device *sdev,
-					      struct scsi_sense_hdr *sense_hdr)
+static void alua_handle_state_transition(struct scsi_device *sdev)
 {
 	struct alua_dh_data *h = sdev->handler_data;
 	struct alua_port_group *pg;
 
+	rcu_read_lock();
+	pg = rcu_dereference(h->pg);
+	if (pg)
+		pg->state = SCSI_ACCESS_STATE_TRANSITIONING;
+	rcu_read_unlock();
+	alua_check(sdev, false);
+}
+
+static enum scsi_disposition alua_check_sense(struct scsi_device *sdev,
+					      struct scsi_sense_hdr *sense_hdr)
+{
 	switch (sense_hdr->sense_key) {
 	case NOT_READY:
 		if (sense_hdr->asc == 0x04 && sense_hdr->ascq == 0x0a) {
 			/*
 			 * LUN Not Accessible - ALUA state transition
 			 */
-			rcu_read_lock();
-			pg = rcu_dereference(h->pg);
-			if (pg)
-				pg->state = SCSI_ACCESS_STATE_TRANSITIONING;
-			rcu_read_unlock();
-			alua_check(sdev, false);
+			alua_handle_state_transition(sdev);
 			return NEEDS_RETRY;
 		}
 		break;
 	case UNIT_ATTENTION:
+		if (sense_hdr->asc == 0x04 && sense_hdr->ascq == 0x0a) {
+			/*
+			 * LUN Not Accessible - ALUA state transition
+			 */
+			alua_handle_state_transition(sdev);
+			return NEEDS_RETRY;
+		}
 		if (sense_hdr->asc == 0x29 && sense_hdr->ascq == 0x00) {
 			/*
 			 * Power On, Reset, or Bus Device Reset.
@@ -502,7 +514,8 @@  static int alua_tur(struct scsi_device *sdev)
 
 	retval = scsi_test_unit_ready(sdev, ALUA_FAILOVER_TIMEOUT * HZ,
 				      ALUA_FAILOVER_RETRIES, &sense_hdr);
-	if (sense_hdr.sense_key == NOT_READY &&
+	if ((sense_hdr.sense_key == NOT_READY ||
+	     sense_hdr.sense_key == UNIT_ATTENTION) &&
 	    sense_hdr.asc == 0x04 && sense_hdr.ascq == 0x0a)
 		return SCSI_DH_RETRY;
 	else if (retval)