From: James Simmons <jsimmons@infradead.org>
To: Andreas Dilger <adilger@whamcloud.com>,
Oleg Drokin <green@whamcloud.com>, NeilBrown <neilb@suse.de>
Cc: Chris Horn <chris.horn@hpe.com>,
Lustre Development List <lustre-devel@lists.lustre.org>
Subject: [lustre-devel] [PATCH 45/49] lnet: Recover peer NI w/exponential backoff interval
Date: Thu, 15 Apr 2021 00:02:37 -0400 [thread overview]
Message-ID: <1618459361-17909-46-git-send-email-jsimmons@infradead.org> (raw)
In-Reply-To: <1618459361-17909-1-git-send-email-jsimmons@infradead.org>
From: Chris Horn <chris.horn@hpe.com>
Perform LNet recovery pings of peer NIs with an exponential backoff
interval.
- The interval is equal to 2^(number failed pings) up to a maximum
of 900 seconds (15 minutes).
- When a message is received the count of failed pings for the
associated peer NI is reset to 0 so that recovery can happen more
quickly.
HPE-bug-id: LUS-9109
WC-bug-id: https://jira.whamcloud.com/browse/LU-13569
Lustre-commit: 917553c537a8860 ("LU-13569 lnet: Recover peer NI w/exponential backoff interval")
Signed-off-by: Chris Horn <chris.horn@hpe.com>
Reviewed-on: https://review.whamcloud.com/39720
Reviewed-by: Neil Brown <neilb@suse.de>
Reviewed-by: Alexander Boyko <alexander.boyko@hpe.com>
Reviewed-by: Serguei Smirnov <ssmirnov@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
---
include/linux/lnet/lib-lnet.h | 22 ++++++++++++++++++++++
include/linux/lnet/lib-types.h | 6 ++++++
net/lnet/lnet/lib-move.c | 8 ++++++++
net/lnet/lnet/lib-msg.c | 6 +++++-
net/lnet/lnet/peer.c | 11 ++++++++++-
5 files changed, 51 insertions(+), 2 deletions(-)
diff --git a/include/linux/lnet/lib-lnet.h b/include/linux/lnet/lib-lnet.h
index e30d0c4..8b369dd 100644
--- a/include/linux/lnet/lib-lnet.h
+++ b/include/linux/lnet/lib-lnet.h
@@ -910,6 +910,28 @@ int lnet_get_peer_ni_info(u32 peer_index, u64 *nid,
return false;
}
+#define LNET_RECOVERY_INTERVAL_MAX 900
+static inline unsigned int
+lnet_get_next_recovery_ping(unsigned int ping_count, time64_t now)
+{
+ unsigned int interval;
+
+ /* 2^9 = 512, 2^10 = 1024 */
+ if (ping_count > 9)
+ interval = LNET_RECOVERY_INTERVAL_MAX;
+ else
+ interval = 1 << ping_count;
+
+ return now + interval;
+}
+
+static inline void
+lnet_peer_ni_set_next_ping(struct lnet_peer_ni *lpni, time64_t now)
+{
+ lpni->lpni_next_ping =
+ lnet_get_next_recovery_ping(lpni->lpni_ping_count, now);
+}
+
/*
* A peer NI is alive if it satisfies the following two conditions:
* 1. peer NI health >= LNET_MAX_HEALTH_VALUE * router_sensitivity_percentage
diff --git a/include/linux/lnet/lib-types.h b/include/linux/lnet/lib-types.h
index cc451cf..af8f61e 100644
--- a/include/linux/lnet/lib-types.h
+++ b/include/linux/lnet/lib-types.h
@@ -573,6 +573,12 @@ struct lnet_peer_ni {
atomic_t lpni_healthv;
/* recovery ping mdh */
struct lnet_handle_md lpni_recovery_ping_mdh;
+ /* When to send the next recovery ping */
+ time64_t lpni_next_ping;
+ /* How many pings sent during current recovery period did not receive
+ * a reply. NB: reset whenever _any_ message arrives from this peer NI
+ */
+ unsigned int lpni_ping_count;
/* CPT this peer attached on */
int lpni_cpt;
/* state flags -- protected by lpni_lock */
diff --git a/net/lnet/lnet/lib-move.c b/net/lnet/lnet/lib-move.c
index bdcba54..ad1517d 100644
--- a/net/lnet/lnet/lib-move.c
+++ b/net/lnet/lnet/lib-move.c
@@ -3398,6 +3398,12 @@ struct lnet_mt_event_info {
}
spin_unlock(&lpni->lpni_lock);
+
+ if (now < lpni->lpni_next_ping) {
+ lnet_net_unlock(0);
+ continue;
+ }
+
lnet_net_unlock(0);
/* NOTE: we're racing with peer deletion from user space.
@@ -3446,6 +3452,8 @@ struct lnet_mt_event_info {
continue;
}
+ lpni->lpni_ping_count++;
+
lpni->lpni_recovery_ping_mdh = mdh;
lnet_peer_ni_add_to_recoveryq_locked(lpni,
&processed_list,
diff --git a/net/lnet/lnet/lib-msg.c b/net/lnet/lnet/lib-msg.c
index 2e8fea7..0a4a317 100644
--- a/net/lnet/lnet/lib-msg.c
+++ b/net/lnet/lnet/lib-msg.c
@@ -863,8 +863,11 @@
switch (hstatus) {
case LNET_MSG_STATUS_OK:
- /* increment the local ni health weather we successfully
+ /* increment the local ni health whether we successfully
* received or sent a message on it.
+ *
+ * Ping counts are reset to 0 as appropriate to allow for
+ * faster recovery.
*/
lnet_inc_healthv(&ni->ni_healthv, lnet_health_sensitivity);
/* It's possible msg_txpeer is NULL in the LOLND
@@ -875,6 +878,7 @@
* as indication that the router is fully healthy.
*/
if (lpni && msg->msg_rx_committed) {
+ lpni->lpni_ping_count = 0;
/* If we're receiving a message from the router or
* I'm a router, then set that lpni's health to
* maximum so we can commence communication
diff --git a/net/lnet/lnet/peer.c b/net/lnet/lnet/peer.c
index f9af5da..15fcb5e 100644
--- a/net/lnet/lnet/peer.c
+++ b/net/lnet/lnet/peer.c
@@ -4006,14 +4006,23 @@ int lnet_get_peer_info(struct lnet_ioctl_peer_cfg *cfg, void __user *bulk)
CDEBUG(D_NET, "lpni %s aged out last alive %lld\n",
libcfs_nid2str(lpni->lpni_nid),
lpni->lpni_last_alive);
+ /* Reset the ping count so that if this peer NI is added back to
+ * the recovery queue we will send the first ping right away.
+ */
+ lpni->lpni_ping_count = 0;
return;
}
/* This peer NI is going on the recovery queue, so take a ref on it */
lnet_peer_ni_addref_locked(lpni);
- CDEBUG(D_NET, "%s added to recovery queue. last alive: %lld health: %d\n",
+ lnet_peer_ni_set_next_ping(lpni, now);
+
+ CDEBUG(D_NET,
+ "%s added to recovery queue. ping count: %u next ping: %lld last alive: %lld health: %d\n",
libcfs_nid2str(lpni->lpni_nid),
+ lpni->lpni_ping_count,
+ lpni->lpni_next_ping,
lpni->lpni_last_alive,
atomic_read(&lpni->lpni_healthv));
--
1.8.3.1
_______________________________________________
lustre-devel mailing list
lustre-devel@lists.lustre.org
http://lists.lustre.org/listinfo.cgi/lustre-devel-lustre.org
next prev parent reply other threads:[~2021-04-15 4:06 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-04-15 4:01 [lustre-devel] [PATCH 00/49] lustre: sync to OpenSFS as of March 30 2021 James Simmons
2021-04-15 4:01 ` [lustre-devel] [PATCH 01/49] lnet: libcfs: Fix for unconfigured arch_stackwalk James Simmons
2021-04-15 4:01 ` [lustre-devel] [PATCH 02/49] lustre: lmv: iput() can safely be passed NULL James Simmons
2021-04-15 4:01 ` [lustre-devel] [PATCH 03/49] lustre: llite: mark extended attr and inode flags James Simmons
2021-04-15 4:01 ` [lustre-devel] [PATCH 04/49] lnet: lnet_notify sets route aliveness incorrectly James Simmons
2021-04-15 4:01 ` [lustre-devel] [PATCH 05/49] lnet: Prevent discovery on peer marked deletion James Simmons
2021-04-15 4:01 ` [lustre-devel] [PATCH 06/49] lnet: Prevent discovery on deleted peer James Simmons
2021-04-15 4:01 ` [lustre-devel] [PATCH 07/49] lnet: Transfer disc src NID when merging peers James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 08/49] lnet: Lookup lpni after discovery James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 09/49] lustre: llite: update and fix module loading bug in mounting code James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 10/49] lnet: socklnd: change various ints to bool James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 11/49] lnet: Correct asymmetric route detection James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 12/49] lustre: fixup ldlm_pool and lu_object shrinker failure cases James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 13/49] lustre: log: Add ending newline for some messages James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 14/49] lustre: use with_imp_locked() more broadly James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 15/49] lnet: o2iblnd: change some ints to bool James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 16/49] lustre: lmv: striped directory as subdirectory mount James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 17/49] lustre: llite: create file_operations registration function James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 18/49] lustre: osc: fix performance regression in osc_extent_merge() James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 19/49] lustre: mds: add enums for MDS_ATTR flags James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 20/49] lustre: uapi: remove OBD_IOC_LOV_GET_CONFIG James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 21/49] lustre: sec: fix migrate for encrypted dir James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 22/49] lnet: libcfs: restore LNET_DUMP_ON_PANIC functionality James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 23/49] lustre: ptlrpc: fix ASSERTION on scp_rqbd_posted James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 24/49] lustre: ldlm: not freed req on enqueue James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 25/49] lnet: uapi: move userland only nidstr.h handling James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 26/49] lnet: libcfs: don't depend on sysctl support for debugfs James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 27/49] lustre: ptlrpc: Add a binary heap implementation James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 28/49] lustre: ptlrpc: Implement NRS Delay Policy James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 29/49] lustre: ptlrpc: rename cfs_binheap to simply binheap James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 30/49] lustre: ptlrpc: mark some functions as static James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 31/49] lustre: use tgt_pool for lov layer James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 32/49] lustre: quota: make used for pool correct James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 33/49] lustre: quota: call rhashtable_lookup near params decl James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 34/49] lustre: lov: cancel layout lock on replay deadlock James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 35/49] lustre: obdclass: Protect cl_env_percpu[] James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 36/49] lnet: libcfs: discard cfs_trace_console_buffers[] James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 37/49] lnet: libcfs: discard cfs_trace_copyin_string() James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 38/49] lustre: lmv: don't use lqr_alloc spinlock in lmv James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 39/49] lustre: lov: fault page update cp_lov_index James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 40/49] lustre: update version to 2.14.51 James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 41/49] lustre: llite: mirror extend/copy keeps sparseness James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 42/49] lustre: ptlrpc: don't use list_for_each_entry_safe unnecessarily James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 43/49] lnet: Age peer NI out of recovery James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 44/49] lnet: Only recover known good peer NIs James Simmons
2021-04-15 4:02 ` James Simmons [this message]
2021-04-15 4:02 ` [lustre-devel] [PATCH 46/49] lustre: lov: return valid stripe_count/size for PFL files James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 47/49] lnet: convert lpni_refcount to a kref James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 48/49] lustre: lmv: handle default stripe_count=-1 properly James Simmons
2021-04-15 4:02 ` [lustre-devel] [PATCH 49/49] lnet: libcfs: discard cfs_array_alloc() 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=1618459361-17909-46-git-send-email-jsimmons@infradead.org \
--to=jsimmons@infradead.org \
--cc=adilger@whamcloud.com \
--cc=chris.horn@hpe.com \
--cc=green@whamcloud.com \
--cc=lustre-devel@lists.lustre.org \
--cc=neilb@suse.de \
/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).