nvme/tcp: fail qpair when spdk_sock_flush() fails
If spdk_sock_flush() returns an error, there's no reason not to disconnect the qpair, as it usually means that that socket's connection has been terminated. Signed-off-by: Konrad Sztyber <konrad.sztyber@intel.com> Change-Id: I54e9bebc38e2a24a3baf69eb18ec3c654b210318 Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/16644 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Aleksey Marchuk <alexeymar@nvidia.com>
This commit is contained in:
parent
739c6d7c5a
commit
ac94b60b54
@ -1752,7 +1752,7 @@ nvme_tcp_qpair_process_completions(struct spdk_nvme_qpair *qpair, uint32_t max_c
|
||||
if (spdk_unlikely(tqpair->qpair.ctrlr->timeout_enabled)) {
|
||||
nvme_tcp_qpair_check_timeout(qpair);
|
||||
}
|
||||
return rc;
|
||||
goto fail;
|
||||
}
|
||||
}
|
||||
|
||||
|
Loading…
Reference in New Issue
Block a user