From: Arun Sharma <asharma@fb.com>
To: Joe Perches <joe@perches.com>
Cc: Arun Sharma <asharma@fb.com>, Bjorn Helgaas <bhelgaas@google.com>,
linux-kernel@vger.kernel.org, netdev@vger.kernel.org,
David Miller <davem@davemloft.net>,
Glauber Costa <glommer@parallels.com>,
Ingo Molnar <mingo@elte.hu>,
christoph.paasch@uclouvain.be
Subject: Re: [PATCH] net: Disambiguate kernel message
Date: Tue, 31 Jan 2012 11:47:18 -0800 [thread overview]
Message-ID: <20120131194718.GA9834@dev3310.snc6.facebook.com> (raw)
In-Reply-To: <1328035857.6911.8.camel@joe2Laptop>
On Tue, Jan 31, 2012 at 10:50:57AM -0800, Joe Perches wrote:
>
> These 2 blocks emit the messages in different order.
>
> It might be useful to use a generic routine.
Refactored slightly differently below.
On 1/31/12 11:44 AM, Ingo Molnar wrote:
> A small detail: please don't line break user-visible strings in
> mid sentence. Just keep the line long. Makes it much easier to
> search for the source of a kernel message:
>
pr_info() on a single line now.
>From 08f36b8fd9f2439001a4192c0d1ff390d13e67e0 Mon Sep 17 00:00:00 2001
From: Arun Sharma <asharma@fb.com>
Date: Mon, 30 Jan 2012 14:16:06 -0800
Subject: [PATCH] net: Disambiguate kernel message
Some of our machines were reporting:
TCP: too many of orphaned sockets
even when the number of orphaned sockets was well below the
limit.
We print a different message depending on whether we're out
of TCP memory or there are too many orphaned sockets.
Signed-off-by: Arun Sharma <asharma@fb.com>
Suggested-by: Mohan Srinivasan <mohan@fb.com>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: David Miller <davem@davemloft.net>
Cc: Glauber Costa <glommer@parallels.com>
Cc: Ingo Molnar <mingo@elte.hu>
---
include/net/tcp.h | 20 ++++++++++++++++----
net/ipv4/tcp.c | 11 +++++++----
net/ipv4/tcp_timer.c | 9 +++++----
3 files changed, 28 insertions(+), 12 deletions(-)
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 0118ea9..0dba9d6 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -273,6 +273,14 @@ static inline int between(__u32 seq1, __u32 seq2, __u32 seq3)
return seq3 - seq2 >= seq1 - seq2;
}
+static inline bool tcp_out_of_memory(struct sock *sk)
+{
+ if (sk->sk_wmem_queued > SOCK_MIN_SNDBUF &&
+ sk_memory_allocated(sk) > sk_prot_mem_limits(sk, 2))
+ return true;
+ return false;
+}
+
static inline bool tcp_too_many_orphans(struct sock *sk, int shift)
{
struct percpu_counter *ocp = sk->sk_prot->orphan_count;
@@ -283,13 +291,17 @@ static inline bool tcp_too_many_orphans(struct sock *sk, int shift)
if (orphans << shift > sysctl_tcp_max_orphans)
return true;
}
-
- if (sk->sk_wmem_queued > SOCK_MIN_SNDBUF &&
- sk_memory_allocated(sk) > sk_prot_mem_limits(sk, 2))
- return true;
return false;
}
+static inline void tcp_log_oom(bool orphans, bool socket_memory)
+{
+ if (orphans && net_ratelimit())
+ pr_info("TCP: too many orphaned sockets\n");
+ if (socket_memory && net_ratelimit())
+ pr_info("TCP: out of memory -- consider tuning tcp_mem\n");
+}
+
/* syncookies: remember time of last synqueue overflow */
static inline void tcp_synq_overflow(struct sock *sk)
{
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 9bcdec3..9c38b3a 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2014,11 +2014,14 @@ adjudge_to_death:
}
}
if (sk->sk_state != TCP_CLOSE) {
+ bool too_many_orphans, out_of_socket_memory;
+
sk_mem_reclaim(sk);
- if (tcp_too_many_orphans(sk, 0)) {
- if (net_ratelimit())
- printk(KERN_INFO "TCP: too many of orphaned "
- "sockets\n");
+ too_many_orphans = tcp_too_many_orphans(sk, 0);
+ out_of_socket_memory = tcp_out_of_memory(sk);
+ tcp_log_oom(too_many_orphans, out_of_socket_memory);
+
+ if (too_many_orphans || out_of_socket_memory) {
tcp_set_state(sk, TCP_CLOSE);
tcp_send_active_reset(sk, GFP_ATOMIC);
NET_INC_STATS_BH(sock_net(sk),
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index a516d1e..c47fcb6 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -67,6 +67,7 @@ static int tcp_out_of_resources(struct sock *sk, int do_reset)
{
struct tcp_sock *tp = tcp_sk(sk);
int shift = 0;
+ bool too_many_orphans, out_of_socket_memory;
/* If peer does not open window for long time, or did not transmit
* anything for long time, penalize it. */
@@ -77,10 +78,10 @@ static int tcp_out_of_resources(struct sock *sk, int do_reset)
if (sk->sk_err_soft)
shift++;
- if (tcp_too_many_orphans(sk, shift)) {
- if (net_ratelimit())
- printk(KERN_INFO "Out of socket memory\n");
-
+ too_many_orphans = tcp_too_many_orphans(sk, shift);
+ out_of_socket_memory = tcp_out_of_memory(sk);
+ tcp_log_oom(too_many_orphans, out_of_socket_memory);
+ if (too_many_orphans || out_of_socket_memory) {
/* Catch exceptional cases, when connection requires reset.
* 1. Last segment was sent recently. */
if ((s32)(tcp_time_stamp - tp->lsndtime) <= TCP_TIMEWAIT_LEN ||
--
1.7.4
next prev parent reply other threads:[~2012-01-31 19:47 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-01-30 22:41 [PATCH] net: Disambiguate kernel message Arun Sharma
2012-01-31 4:22 ` Bjorn Helgaas
2012-01-31 18:15 ` Arun Sharma
2012-01-31 18:50 ` Joe Perches
2012-01-31 19:47 ` Arun Sharma [this message]
2012-01-31 20:09 ` Joe Perches
2012-01-31 20:46 ` Arun Sharma
2012-01-31 20:55 ` Joe Perches
2012-01-31 21:15 ` David Miller
2012-01-31 22:05 ` Arun Sharma
2012-01-31 22:12 ` Joe Perches
2012-02-01 19:42 ` David Miller
2012-01-31 19:44 ` Ingo Molnar
2012-01-31 8:51 ` Christoph Paasch
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=20120131194718.GA9834@dev3310.snc6.facebook.com \
--to=asharma@fb.com \
--cc=bhelgaas@google.com \
--cc=christoph.paasch@uclouvain.be \
--cc=davem@davemloft.net \
--cc=glommer@parallels.com \
--cc=joe@perches.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=netdev@vger.kernel.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