lustre-devel-lustre.org archive mirror
 help / color / mirror / Atom feed
From: James Simmons <jsimmons@infradead.org>
To: lustre-devel@lists.lustre.org
Subject: [lustre-devel] [PATCH 15/23] lnet: Preferred NI logic breaks MR routing
Date: Tue, 11 Aug 2020 08:20:11 -0400	[thread overview]
Message-ID: <1597148419-20629-16-git-send-email-jsimmons@infradead.org> (raw)
In-Reply-To: <1597148419-20629-1-git-send-email-jsimmons@infradead.org>

From: Chris Horn <hornc@cray.com>

Edge (final-hop) routers typically use the non-multi-rail destination
(NMR_DST) send case. i.e. they treat the destination as
non-multi-rail. The reason for this is that we do not want routers to
modify the destination peer interface selected by the message
originator. As a result of using the NMR_DST send case, edge routers
set a preferred NI, and then continue to use that NI, because it's
preferred, even if the NI goes down and the router has other healthy
interfaces available to it. Routers do not need to use the preferred
NI selection logic when they are forwarding a message, so modify the
NMR_DST algorithm to allow routers to select any suitable local NI.

HPE-bug-id: LUS-9045
WC-bug-id: https://jira.whamcloud.com/browse/LU-13712
Lustre-commit: ef6c35877b96c ("LU-13712 lnet: Preferred NI logic breaks MR routing")
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Reviewed-on: https://review.whamcloud.com/39168
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
---
 net/lnet/lnet/lib-move.c | 20 ++++++++++++++++++--
 1 file changed, 18 insertions(+), 2 deletions(-)

diff --git a/net/lnet/lnet/lib-move.c b/net/lnet/lnet/lib-move.c
index aa6fe37..7c14518 100644
--- a/net/lnet/lnet/lib-move.c
+++ b/net/lnet/lnet/lib-move.c
@@ -2107,7 +2107,7 @@ struct lnet_ni *
 static int
 lnet_handle_any_local_nmr_dst(struct lnet_send_data *sd)
 {
-	int rc;
+	int rc = 0;
 
 	/* sd->sd_best_lpni is already set to the final destination */
 
@@ -2122,7 +2122,23 @@ struct lnet_ni *
 		return -EFAULT;
 	}
 
-	rc = lnet_select_preferred_best_ni(sd);
+	if (sd->sd_msg->msg_routing) {
+		/* If I'm forwarding this message then I can choose any NI
+		 * on the destination peer net
+		 */
+		sd->sd_best_ni = lnet_find_best_ni_on_spec_net(NULL,
+							       sd->sd_peer,
+							       sd->sd_best_lpni->lpni_peer_net,
+							       sd->sd_md_cpt,
+							       true);
+		if (!sd->sd_best_ni) {
+			CERROR("Unable to forward message to %s. No local NI available\n",
+			       libcfs_nid2str(sd->sd_dst_nid));
+			rc = -EHOSTUNREACH;
+		}
+	} else {
+		rc = lnet_select_preferred_best_ni(sd);
+	}
 	if (!rc)
 		rc = lnet_handle_send(sd);
 
-- 
1.8.3.1

  parent reply	other threads:[~2020-08-11 12:20 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-11 12:19 [lustre-devel] [PATCH 00/23] lustre: latest patches landed to OpenSFS 08/11/2020 James Simmons
2020-08-11 12:19 ` [lustre-devel] [PATCH 01/23] lustre: lov: one more fix to write_intent end for trunc James Simmons
2020-08-11 12:19 ` [lustre-devel] [PATCH 02/23] lustre: lov: annotate nested locking of obd_dev_mutex James Simmons
2020-08-11 12:19 ` [lustre-devel] [PATCH 03/23] lustre: ptlrpc: make ptlrpc_connection_put() static inline James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 04/23] lustre: mdc: create mdc_acl.c James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 05/23] lustre: llite: Remove mutex on dio read James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 06/23] lustre: obd: rename lprocfs_ / LPROC_SEQ_ to debugfs name James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 07/23] lustre: sec: atomicity of encryption context getting/setting James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 08/23] lustre: sec: encryption support for DoM files James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 09/23] lustre: sec: check if page is empty with ZERO_PAGE James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 10/23] lustre: uapi: add OBD_CONNECT2_GETATTR_PFID James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 11/23] lustre: update version to 2.13.55 James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 12/23] lustre: sysfs: error-check value stored in jobid_var James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 13/23] lnet: Add param to control response tracking James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 14/23] lnet: Ensure LNet pings and pushes are always tracked James Simmons
2020-08-11 12:20 ` James Simmons [this message]
2020-08-11 12:20 ` [lustre-devel] [PATCH 16/23] lnet: socklnd: remove declarations of missing functions James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 17/23] lnet: discard unused lnet_print_hdr() James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 18/23] lnet: clarify initialization of lpni_refcount James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 19/23] lnet: Allow duplicate nets in ip2nets syntax James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 20/23] lustre: llite: pack parent FID in getattr James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 21/23] lnet: Clear lp_dc_error when discovery completes James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 22/23] lnet: Have LNet routers monitor the ni_fatal flag James Simmons
2020-08-11 12:20 ` [lustre-devel] [PATCH 23/23] lnet: socklnd: NID to interface mapping issues James Simmons

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1597148419-20629-16-git-send-email-jsimmons@infradead.org \
    --to=jsimmons@infradead.org \
    --cc=lustre-devel@lists.lustre.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).