Whamcloud - gitweb
LU-16990 kfilnd: Use NETWORK_TIMEOUT for TAG_RX_CANCEL 82/51782/2
authorChris Horn <chris.horn@hpe.com>
Thu, 9 Mar 2023 00:18:41 +0000 (18:18 -0600)
committerOleg Drokin <green@whamcloud.com>
Tue, 22 Aug 2023 06:41:07 +0000 (06:41 +0000)
We can get ECANCELED for some tagged receives which results in
transaction failure with TN_EVENT_TAG_RX_CANCEL. This can occur due
to problems with either the source or the target, so we should
use NETWORK_TIMEOUT message status.

Test-Parameters: trivial
HPE-bug-id: LUS-11520
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Change-Id: Ic3c1910f8a8c43447cbbc28129e23350e726830d
Reviewed-on: https://review.whamcloud.com/c/fs/lustre-release/+/51782
Tested-by: jenkins <devops@whamcloud.com>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Ron Gredvig <ron.gredvig@hpe.com>
Reviewed-by: Ian Ziemba <ian.ziemba@hpe.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
lnet/klnds/kfilnd/kfilnd_tn.c

index d5b0672..2fa0a25 100644 (file)
@@ -1318,7 +1318,7 @@ static int kfilnd_tn_state_wait_timeout_tag_comp(struct kfilnd_transaction *tn,
        switch (event) {
        case TN_EVENT_TAG_RX_CANCEL:
                kfilnd_tn_status_update(tn, -ETIMEDOUT,
-                                       LNET_MSG_STATUS_REMOTE_TIMEOUT);
+                                       LNET_MSG_STATUS_NETWORK_TIMEOUT);
                kfilnd_peer_tn_failed(tn->tn_kp, -ETIMEDOUT);
                break;