lib/nvmf: raise RDMA acceptor backlog to 100.

10 is kind of unreasonable. You could easily start
seeing failures if you had just 3 intiators trying to
connect with 4 io qpairs at once.

Signed-off-by: Seth Howell <seth.howell@intel.com>
Change-Id: I9985ffa3b03ebb33880eb5934b60eaafab57c82d
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3096
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
Community-CI: Broadcom CI
Community-CI: Mellanox Build Bot
Reviewed-by: Shuhei Matsumoto <shuhei.matsumoto.xt@hitachi.com>
Reviewed-by: Aleksey Marchuk <alexeymar@mellanox.com>
This commit is contained in:
Seth Howell 2020-06-26 09:46:36 -07:00 committed by Tomasz Zawadzki
parent 203ed4f673
commit 27f435bc1d

View File

@ -2625,7 +2625,7 @@ nvmf_rdma_listen(struct spdk_nvmf_transport *transport,
return -1; return -1;
} }
rc = rdma_listen(port->id, 10); /* 10 = backlog */ rc = rdma_listen(port->id, 100); /* 100 = backlog */
if (rc < 0) { if (rc < 0) {
SPDK_ERRLOG("rdma_listen() failed\n"); SPDK_ERRLOG("rdma_listen() failed\n");
rdma_destroy_id(port->id); rdma_destroy_id(port->id);