From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S935653AbcCQK1C (ORCPT <rfc822;w@1wt.eu>);
	Thu, 17 Mar 2016 06:27:02 -0400
Received: from e37.co.us.ibm.com ([32.97.110.158]:42031 "EHLO
	e37.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S935615AbcCQK05 (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Thu, 17 Mar 2016 06:26:57 -0400
X-IBM-Helo: d03dlp03.boulder.ibm.com
X-IBM-MailFrom: hanpt@linux.vnet.ibm.com
X-IBM-RcptTo: linux-kernel@vger.kernel.org
Date: Thu, 17 Mar 2016 18:26:16 +0800
From: Han Pingtian <hanpt@linux.vnet.ibm.com>
To: linux-kernel@vger.kernel.org
Subject: cgroup pids controller kills systemd service?
Message-ID: <20160317102616.GA2986@localhost.localdomain>
Mail-Followup-To: linux-kernel@vger.kernel.org
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
User-Agent: Mutt/1.5.23 (2014-03-12)
X-TM-AS-MML: disable
X-Content-Scanned: Fidelis XPS MAILER
x-cbid: 16031710-0025-0000-0000-000022E34948
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

Hi,

I noticed that if a systemd service forks children fiercely, then the whole
service will be terminated, perhaps by the kernel. Is this designed?

I have this reproducer:

$ cat reproducer/reproducer.service
[Unit]
Description=Reproducer of bz 136929
After=multi-user.target

[Service]
ExecStart=/home/hpt/reproducer/reproducer.sh
Type=forking

[Install]
WantedBy=multi-user.target

% cat reproducer/reproducer.sh     
#!/bin/bash

foo()
{
        echo $1: $(date)
        sleep 30
}

bar()
{
        c=1
        while true
        do
                for ((i=1;i<=2048;i++))
                do
                        foo $c &
                        ((c++))
                done

                wait
                c=1
        done
}

# main
bar &

disown -a

exit 0

The default task limit of systemd is 512. Soon after the starting of the
service, it failed with those messages:

% sudo systemctl status reproducer.service                                                                                                                                                 4:02 hpt@pinelp3 ~ %
● reproducer.service - Reproducer of bz 136929
   Loaded: loaded (/etc/systemd/system/reproducer.service; disabled; vendor preset: enabled)
   Active: failed (Result: exit-code) since Thu 2016-03-17 04:02:09 CDT; 13s ago
  Process: 14859 ExecStart=/home/hpt/reproducer/reproducer.sh (code=exited, status=0/SUCCESS)
 Main PID: 14861 (code=exited, status=254)

Mar 17 04:02:07 pinelp3 reproducer.sh[14859]: /home/hpt/reproducer/reproducer.sh: fork: Resource temporarily unavailable
Mar 17 04:02:07 pinelp3 reproducer.sh[14859]: /home/hpt/reproducer/reproducer.sh: fork: Resource temporarily unavailable
Mar 17 04:02:07 pinelp3 reproducer.sh[14859]: /home/hpt/reproducer/reproducer.sh: fork: Resource temporarily unavailable
Mar 17 04:02:08 pinelp3 reproducer.sh[14859]: /home/hpt/reproducer/reproducer.sh: fork: Resource temporarily unavailable
Mar 17 04:02:08 pinelp3 reproducer.sh[14859]: /home/hpt/reproducer/reproducer.sh: fork: Resource temporarily unavailable
Mar 17 04:02:08 pinelp3 reproducer.sh[14859]: /home/hpt/reproducer/reproducer.sh: fork: Resource temporarily unavailable
Mar 17 04:02:08 pinelp3 systemd[1]: reproducer.service: Main process exited, code=exited, status=254/n/a
Mar 17 04:02:09 pinelp3 systemd[1]: reproducer.service: Unit entered failed state.
Mar 17 04:02:09 pinelp3 systemd[1]: reproducer.service: Failed with result 'exit-code'.
Mar 17 04:02:18 pinelp3 systemd[1]: Stopped Reproducer of bz 136929


Thanks in advance.