AppSuite:ResourceLimits: Difference between revisions

From Open-Xchange
(Add redirect to middleware startup)
 
(3 intermediate revisions by the same user not shown)
Line 1: Line 1:
 
#REDIRECT [[:Appsuite:MiddlewareStartup]]
== Configuration of Resource Limits ==
 
=== Overview ===
 
Several ways exist to restrict resources on a linux system from a global level down to user/groups or even shells and the processes started by them.
 
==== Sysctl ====
Sysctl is used to modify kernel parameters at runtime. E.g. to set the maximum number of files
 
  $ sysctl -w fs.file-max=100000
 
To permanently set them append to the main configuration file and reload the settings
 
  $ echo fs.file-max=100000 >> /etc/sysctl.conf
  $ sysctl -p
 
More infos can be found via '''man sysctl'''
 
==== Limits.conf ====
 
Allows to restrict resources an a global, group or user level. E.g:
 
  $ cat /etc/security/limits.d/90-nproc.conf
  # Default limit for number of user's processes to prevent
  # accidental fork bombs.
  # See rhbz #432903 for reasoning.
 
  *          soft    nproc    1024
 
From '''man limits.conf''':
<blockquote>
''Also, please note that all limit settings are set per login. They are not global, nor are they permanent; existing only for the duration of the session.''
</blockquote>
 
The limits per login are applied via the ''pam'' stack. See '''man pam''' and '''man pam_limits''' for more details. As those limits are bound to sessions they don't affect most daemons started by our supported init systems or init utils. Most state that they are ignored by design, see [https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/938669/comments/4 upstart], [https://bugzilla.redhat.com/show_bug.cgi?id=754285#c1 systemd] and [https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=302079 start-stop-daemon]
 
==== Ulimit ====
From '''man bash'''
<blockquote>
ulimit [-HSTabcdefilmnpqrstuvx [limit]]
              Provides  control over the resources available to the shell and to processes started by it, on systems that allow such control.
</blockquote>
This is what we use in our System V compatible init scripts to increase resources for the open-xchange process across multiple distros. Currently only ''the maximum number of processes'' and ''the maximum number of open file descriptors available to a single user'' are increased via ulimit. The values are specified in '''/opt/open-xchange/ox-scriptconf.sh'''
 
==== Systemd ====
 
===== Control Groups=====
 
Control groups should only affect the OX middleware if you create/manage them yourself of if you are using a ''modern'' distribution that already uses systemd as init.
 
Citing from the kernel cgroup [https://www.kernel.org/doc/Documentation/cgroup-v2.txt documentation]:
 
<blockquote>
1-2. What is cgroup?
 
cgroup is a mechanism to organize processes hierarchically and
distribute system resources along the hierarchy in a controlled and
configurable manner.
 
cgroup is largely composed of two parts - the core and controllers.
cgroup core is primarily responsible for hierarchically organizing
processes.  A cgroup controller is usually responsible for
distributing a specific type of system resource along the hierarchy
although there are utility controllers which serve purposes other than
resource distribution.
 
cgroups form a tree structure and every process in the system belongs
to one and only one cgroup.  All threads of a process belong to the
same cgroup.  On creation, all processes are put in the cgroup that
the parent process belongs to at the time.  A process can be migrated
to another cgroup.  Migration of a process doesn't affect already
existing descendant processes.
 
Following certain structural constraints, controllers may be enabled or
disabled selectively on a cgroup.  All controller behaviors are
hierarchical - if a controller is enabled on a cgroup, it affects all
processes which belong to the cgroups consisting the inclusive
sub-hierarchy of the cgroup.  When a controller is enabled on a nested
cgroup, it always restricts the resource distribution further.  The
restrictions set closer to the root in the hierarchy can not be
overridden from further away.
</blockquote>
 
So processes are organized into a tree structure of control groups and controllers are responsible for the distribution of resources. So what kind of controllers exist?
 
<blockquote>
5. Controllers
 
5-1. CPU
 
The "cpu" controllers regulates distribution of CPU cycles.  This
controller implements weight and absolute bandwidth limit models for
normal scheduling policy and absolute bandwidth allocation model for
realtime scheduling policy.
 
5-2. Memory
 
The "memory" controller regulates distribution of memory.
...
While not completely water-tight, all major memory usages by a given
cgroup are tracked so that the total memory consumption can be
accounted and controlled to a reasonable extent.
 
5-3. IO
 
The "io" controller regulates the distribution of IO resources.  This
controller implements both weight based and absolute bandwidth or IOPS
limit distribution; however, weight based distribution is available
only if cfq-iosched is in use and neither scheme is available for
blk-mq devices.
</blockquote>
 
The open-xchange service is simply put into the default ''system.slice'' without applying further limits.
 
  singlenode$ systemd-cgls
  ├─1 /sbin/init
  ├─system.slice
  │ ├─avahi-daemon.service
  │ │ ├─501 avahi-daemon: running [singlenode]
  │ │ └─514 avahi-daemon: chroot helper
  │ ├─console-kit-daemon.service
  │ │ └─16164 /usr/sbin/console-kit-daemon --no-daemon
  │ ├─dbus.service
  │ │ └─508 /usr/bin/dbus-daemon --system --address=systemd: --nofork --nopidfile --systemd-activation
  │ ├─munin-node.service
  │ │ └─4290 /usr/bin/perl -wT /usr/sbin/munin-node
  │ ├─open-xchange.service
  │ │ └─6037 /usr/bin/java -Dsun.net.inetaddr.ttl=3600 -Dnetworkaddress.cache.ttl=3600 -Dnetworkaddress.cache.negative.ttl=10 ...
 
To check all the details use
 
singlenode:~ # systemctl show system.slice
 
===== Limits besides control groups =====
Besides control groups systemd allows you to apply other limits to the execution environment of your service. Here we can apply the limits that would normally be applied via ''limits.conf'' or ''ulimit''. Systemd uses [http://man7.org/linux/man-pages/man2/setrlimit.2.html setrlimit] for this. The options that we set by default are:
 
  * LimitNOFILE
  * LimitNPROC
 
You can check this by looking at the service file that is shipped by default ('''Note:''' Depending on your distro the service files may be located either in /usr/lib/systemd or /lib/systemd. We'll use /usr/lib in this oxpedia article)
 
  singlenode:~ # cat /usr/lib/systemd/system/open-xchange.service
  [Unit]
  After=remote-fs.target
  After=time-sync.target ypbind.service sendmail.service cyrus.service 
 
  [Service]
  User=open-xchange
  PermissionsStartOnly=true
  TimeoutStartSec=0
  ExecStartPre=/opt/open-xchange/sbin/triggerupdatethemes -u
  ExecStart=/opt/open-xchange/sbin/open-xchange
  ExecStop=/opt/open-xchange/sbin/shutdown -w
  ExecReload=/opt/open-xchange/sbin/triggerreloadconfiguration -d
  KillMode=process
  LimitNOFILE=65536
  LimitNPROC=65536
 
  [Install]
  WantedBy=multi-user.target
 
===== Drop-in configs =====
Drop in configs allow administrator to easily override the default service unit files. So if you want to change the default limits or add additional limits have a look at
 
  singlenode:~ # cat /etc/systemd/system/open-xchange.service.d/limits.conf
  # Override and add options in this file
  # See systemd.exec(5) for other limits
 
  [Service]
  #LimitNPROC=65536
  #LimitNOFILE=65536
 
== Open-Xchange middleware on specific distros ==
The support for the mentioned mechanism of resource control differ depending on the distribution and the init system in use.
 
=== Debian 7 ===
;Init
: System V style
;OX Configurable Limits/Defaults
: nofile, nproc
 
The mentioned limits can be configured via '''/opt/open-xchange/etc/ox-scriptconf.sh'''. The limits are applied via ulimit in the service's init script. The open-xchange service is finally started via start-stop-daemon which doesn't doesn't consider '''/etc/security/limits.*'''
 
=== RHEL 6 / CentOS 6 ===
;Init
: Upstart, System V compatible
;OX Configurable Limits/Defaults
: nofile, nproc
 
The mentioned limits can be configured via '''/opt/open-xchange/etc/ox-scriptconf.sh'''. The limits are applied via ulimit in the service's init script. Furthermore as the open-xchange service is finally started via '''su ... open-xchange''' on this distro a user session is opened via su/pam and the default CentOS pam config reads the '''/etc/security/limits.*''' configuration by loading the pam stack like:
 
: '''/etc/pam.d/su'''
:: '''-> /etc/pam.d/system-auth'''
::: '''-> pam_limits.so'''
 
If NPROC isn't configured for the open-xchange-server it's restricted to '''1024''' globally by default to prevent accidental fork bombs, see '''/etc/security/limits.d/90-nproc.conf''' which can result in severe problems modern multithreaded applications.
 
=== RHEL 7 / CentOS 7 / Debian 8 / SLE 12 ===
;Init
: Systemd
;OX Configurable Limits/Defaults
: nofile, nproc
 
For systemd the default limits are configured directly in the service's unit file that is shipped by OX and located at '''/usr/lib/systemd/system/open-xchange.service'''. The drop-in config to override or extend the default unit file is located at '''/etc/systemd/system/open-xchange.service.d/limits.conf'''. [https://www.freedesktop.org/software/systemd/man/systemd.exec.html Systemd.exec] shows a whole lot of options that can be used by admins to adapt the default service to their specific needs.
 
== Verify limits ==
 
=== System V ===
  singlenode:~ # read pid < /var/run/open-xchange.pid
  singlenode:~ # cat /proc/$pid/limits
  Limit                    Soft Limit          Hard Limit          Units
  Max cpu time              unlimited            unlimited            seconds
  Max file size            unlimited            unlimited            bytes
  Max data size            unlimited            unlimited            bytes
  Max stack size            8388608              unlimited            bytes
  Max core file size        0                    unlimited            bytes
  Max resident set          unlimited            unlimited            bytes
  Max processes            65536                65536                processes
  Max open files            65536                65536                files
  Max locked memory        65536                65536                bytes
  Max address space        unlimited            unlimited            bytes
  Max file locks            unlimited            unlimited            locks
  Max pending signals      24254                24254                signals
  Max msgqueue size        819200              819200              bytes
  Max nice priority        0                    0
  Max realtime priority    0                    0
  Max realtime timeout      unlimited            unlimited            us
 
=== Systemd ===
  singlenode:~ # systemctl show open-xchange | grep Limit
  StartLimitInterval=10000000
  StartLimitBurst=5
  StartLimitAction=none
  MemoryLimit=18446744073709551615
  LimitCPU=18446744073709551615
  LimitFSIZE=18446744073709551615
  LimitDATA=18446744073709551615
  LimitSTACK=18446744073709551615
  LimitCORE=18446744073709551615
  LimitRSS=18446744073709551615
  LimitNOFILE=65536
  LimitAS=18446744073709551615
  LimitNPROC=65536
  LimitMEMLOCK=65536
  LimitLOCKS=18446744073709551615
  LimitSIGPENDING=19827
  LimitMSGQUEUE=819200
  LimitNICE=0
  LimitRTPRIO=0
  LimitRTTIME=18446744073709551615

Latest revision as of 20:58, 24 April 2017