Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1590200 > unrolled thread

Re: [RESEND 2] [PATCH] rlimits: Print more information when limits are exceeded

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [RESEND 2] [PATCH] rlimits: Print more information when limits  are exceeded Thomas Gleixner <tglx@linutronix.de> - 2017-03-01 11:40 +0100
    [PATCH v2] rlimits: Print more information when CPU/RT limits are exceeded Arun Raghavan <arun@arunraghavan.net> - 2017-03-01 16:10 +0100
      [tip:timers/core] rlimits: Print more information when CPU/RT  limits are exceeded tip-bot for Arun Raghavan <tipbot@zytor.com> - 2017-03-13 21:40 +0100

#1590200 — Re: [RESEND 2] [PATCH] rlimits: Print more information when limits are exceeded

On Sat, 18 Feb 2017, Arun Raghavan wrote:

> This dumps some information in logs when a process exceeds its CPU or RT
> limits (soft and hard). Makes debugging easier when userspace triggers
> these limits.

Sigh. This changelog sucks. "dumps some information" is pretty useless and
it does not explain WHY you want to do that. Please structure the changelog
in a way which makes it easy to understand.

1) Problem description

2) Solution

> diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c
> index e9e8c10..6dbcf84 100644
> --- a/kernel/time/posix-cpu-timers.c
> +++ b/kernel/time/posix-cpu-timers.c
> @@ -860,6 +860,9 @@ static void check_thread_timers(struct task_struct *tsk,
>  			 * At the hard limit, we just die.
>  			 * No need to calculate anything else now.
>  			 */
> +			printk(KERN_INFO

			pr_info("CPU Watchdog Timeout (hard): %s[%d]\n",

and no artificial line breaks, please.

Thanks,

	tglx

[toc] | [next] | [standalone]

#1590379 — [PATCH v2] rlimits: Print more information when CPU/RT limits are exceeded

When a process is sent a SIGKILL because it exceeded CPU or RT limits,
the cause may not be obvious in userspace -- daemonised processes just
get killed, and even foreground process just see a 'Killed' message. The
lack of any information on why this might be happening in logs can be
confusing to users who are not aware of this mechanism.

We now add messages which dump the process name and tid in logs when a
process exceeds its CPU or RT limits (soft and hard) in order to make it
clearer to people debugging such issues.

v2:
* Clarified changelog to make the problem and solution clearer
* Switched printk() to pr_info()

Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
---
 kernel/time/posix-cpu-timers.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c
index b4377a5..bd6893d 100644
--- a/kernel/time/posix-cpu-timers.c
+++ b/kernel/time/posix-cpu-timers.c
@@ -824,6 +824,8 @@ static void check_thread_timers(struct task_struct *tsk,
 			 * At the hard limit, we just die.
 			 * No need to calculate anything else now.
 			 */
+			pr_info("CPU Watchdog Timeout (hard): %s[%d]\n",
+				tsk->comm, task_pid_nr(tsk));
 			__group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk);
 			return;
 		}
@@ -835,8 +837,7 @@ static void check_thread_timers(struct task_struct *tsk,
 				soft += USEC_PER_SEC;
 				sig->rlim[RLIMIT_RTTIME].rlim_cur = soft;
 			}
-			printk(KERN_INFO
-				"RT Watchdog Timeout: %s[%d]\n",
+			pr_info("RT Watchdog Timeout (soft): %s[%d]\n",
 				tsk->comm, task_pid_nr(tsk));
 			__group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk);
 		}
@@ -934,6 +935,8 @@ static void check_process_timers(struct task_struct *tsk,
 			 * At the hard limit, we just die.
 			 * No need to calculate anything else now.
 			 */
+			pr_info("RT Watchdog Timeout (hard): %s[%d]\n",
+				tsk->comm, task_pid_nr(tsk));
 			__group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk);
 			return;
 		}
@@ -941,6 +944,8 @@ static void check_process_timers(struct task_struct *tsk,
 			/*
 			 * At the soft limit, send a SIGXCPU every second.
 			 */
+			pr_info("CPU Watchdog Timeout (soft): %s[%d]\n",
+				tsk->comm, task_pid_nr(tsk));
 			__group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk);
 			if (soft < hard) {
 				soft++;
-- 
2.9.3

[toc] | [prev] | [next] | [standalone]

#1599786 — [tip:timers/core] rlimits: Print more information when CPU/RT limits are exceeded

Commit-ID:  e7ea7c9806a2681807257ea89085339d33f7fa0b
Gitweb:     http://git.kernel.org/tip/e7ea7c9806a2681807257ea89085339d33f7fa0b
Author:     Arun Raghavan <arun@arunraghavan.net>
AuthorDate: Wed, 1 Mar 2017 20:23:09 +0530
Committer:  Thomas Gleixner <tglx@linutronix.de>
CommitDate: Mon, 13 Mar 2017 21:32:15 +0100

rlimits: Print more information when CPU/RT limits are exceeded

When a process is sent a SIGKILL because it exceeded CPU or RT limits,
the cause may not be obvious in userspace -- daemonised processes just
get killed, and even foreground process just see a 'Killed' message. The
lack of any information on why this might be happening in logs can be
confusing to users who are not aware of this mechanism.

Add messages which dump the process name and tid in dmesg when a process
exceeds its CPU or RT limits (soft and hard) in order to make it clearer to
people debugging such issues.

Signed-off-by: Arun Raghavan <arun@arunraghavan.net>
Link: http://lkml.kernel.org/r/20170301145309.27214-1-arun@arunraghavan.net
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

---
 kernel/time/posix-cpu-timers.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c
index 4513ad1..76bea3a 100644
--- a/kernel/time/posix-cpu-timers.c
+++ b/kernel/time/posix-cpu-timers.c
@@ -825,6 +825,8 @@ static void check_thread_timers(struct task_struct *tsk,
 			 * At the hard limit, we just die.
 			 * No need to calculate anything else now.
 			 */
+			pr_info("CPU Watchdog Timeout (hard): %s[%d]\n",
+				tsk->comm, task_pid_nr(tsk));
 			__group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk);
 			return;
 		}
@@ -836,8 +838,7 @@ static void check_thread_timers(struct task_struct *tsk,
 				soft += USEC_PER_SEC;
 				sig->rlim[RLIMIT_RTTIME].rlim_cur = soft;
 			}
-			printk(KERN_INFO
-				"RT Watchdog Timeout: %s[%d]\n",
+			pr_info("RT Watchdog Timeout (soft): %s[%d]\n",
 				tsk->comm, task_pid_nr(tsk));
 			__group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk);
 		}
@@ -935,6 +936,8 @@ static void check_process_timers(struct task_struct *tsk,
 			 * At the hard limit, we just die.
 			 * No need to calculate anything else now.
 			 */
+			pr_info("RT Watchdog Timeout (hard): %s[%d]\n",
+				tsk->comm, task_pid_nr(tsk));
 			__group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk);
 			return;
 		}
@@ -942,6 +945,8 @@ static void check_process_timers(struct task_struct *tsk,
 			/*
 			 * At the soft limit, send a SIGXCPU every second.
 			 */
+			pr_info("CPU Watchdog Timeout (soft): %s[%d]\n",
+				tsk->comm, task_pid_nr(tsk));
 			__group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk);
 			if (soft < hard) {
 				soft++;

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web