1 .. This work is licensed under a Creative Commons Attribution 4.0 International License.
2 .. http://creativecommons.org/licenses/by/4.0
3 .. (c) <optionally add copywriters name>
5 ===================================
6 OPNFV Barometer User Guide
7 ===================================
13 Barometer collectd plugins description
14 ---------------------------------------
15 .. Describe the specific features and how it is realised in the scenario in a brief manner
16 .. to ensure the user understand the context for the user guide instructions to follow.
18 collectd is a daemon which collects system performance statistics periodically
19 and provides a variety of mechanisms to publish the collected metrics. It
20 supports more than 90 different input and output plugins. Input plugins
21 retrieve metrics and publish them to the collectd deamon, while output plugins
22 publish the data they receive to an end point. collectd also has infrastructure
23 to support thresholding and notification.
25 Barometer has enabled the following collectd plugins:
27 * *dpdkstat plugin*: A read plugin that retrieve stats from the DPDK extended
30 * `ceilometer plugin`_: A write plugin that pushes the retrieved stats to
31 Ceilometer. It's capable of pushing any stats read through collectd to
32 Ceilometer, not just the DPDK stats.
34 * *hugepages plugin*: A read plugin that retrieves the number of available
35 and free hugepages on a platform as well as what is available in terms of
38 * *RDT plugin*: A read plugin that provides the last level cache utilitzation and
39 memory bandwidth utilization
41 * *Open vSwitch events Plugin*: A read plugin that retrieves events from OVS.
43 * *mcelog plugin*: A read plugin that uses mcelog client protocol to check for
44 memory Machine Check Exceptions and sends the stats for reported exceptions
46 All the plugins above are available on the collectd master, except for the
47 plugin as it's a python based plugin and only C plugins are accepted
48 by the collectd community. The ceilometer plugin lives in the OpenStack
51 Other plugins under development or existing as a pull request into collectd master:
53 * *dpdkevents*: A read plugin that retrieves DPDK link status and DPDK
54 forwarding cores liveliness status (DPDK Keep Alive).
56 * *Open vSwitch stats Plugin*: A read plugin that retrieve flow and interface
59 * *SNMP Agent*: A write plugin that will act as a AgentX subagent that receives
60 and handles queries from SNMP master agent and returns the data collected
61 by read plugins. The SNMP Agent plugin handles requests only for OIDs
62 specified in configuration file. To handle SNMP queries the plugin gets data
63 from collectd and translates requested values from collectd's internal format
64 to SNMP format. Supports SNMP: get, getnext and walk requests.
66 * *Legacy/IPMI*: A read plugin that reports platform thermals, voltages,
67 fanspeed, current, flow, power etc. Also, the plugin monitors Intelligent
68 Platform Management Interface (IPMI) System Event Log (SEL) and sends the
70 **Plugins included in the Danube release:**
77 collectd capabilities and usage
78 ------------------------------------
79 .. Describe the specific capabilities and usage for <XYZ> feature.
80 .. Provide enough information that a user will be able to operate the feature on a deployed scenario.
82 **NOTE** Plugins included in the OPNFV D release will be built-in to the fuel
83 plugin and available in the /opt/opnfv directory on the fuel master. You don't
84 need to clone the barometer/collectd repos to use these, but you can configure
85 them as shown in the examples below. Please note, the collectd plugins in OPNFV
86 are configured with reasonable defaults, but can be overriden.
88 Building all Barometer upstreamed plugins from scratch
89 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
90 The plugins that have been merged to the collectd master branch can all be
91 built and configured through the barometer repository.
93 **NOTE: sudo permissions are required to install collectd.**
95 **NOTE: These are instructions for Ubuntu 16.04.**
97 To build and install these dependencies, clone the barometer repo:
101 $ git clone https://gerrit.opnfv.org/gerrit/barometer
103 Install the build dependencies
107 $ ./src/install_build_deps.sh
109 To install collectd as a service and install all it's dependencies:
113 $ cd barometer/src && sudo make && sudo make install
115 This will install collectd as a service and the base install directory
118 Sample configuration files can be found in '/opt/collectd/etc/collectd.conf.d'
120 **Note**: Exec plugin requires non-root user to execute scripts. By default,
121 `collectd_exec` user is used. Barometer scripts do *not* create this user. It
122 needs to be manually added or exec plugin configuration has to be changed to use
123 other, existing user before starting collectd service.
125 Please note if you are using any Open vSwitch plugins you need to run:
130 $ sudo ovs-vsctl set-manager ptcp:6640
132 DPDK statistics plugin
133 ^^^^^^^^^^^^^^^^^^^^^^
134 Repo: https://github.com/collectd/collectd
138 Dependencies: DPDK (http://dpdk.org/)
140 To build and install DPDK to /usr please see:
141 https://github.com/collectd/collectd/blob/master/docs/BUILD.dpdkstat.md
143 Building and installing collectd:
147 $ git clone https://github.com/collectd/collectd.git
150 $ ./configure --enable-syslog --enable-logfile --enable-debug
155 This will install collectd to /opt/collectd
156 The collectd configuration file can be found at /opt/collectd/etc
157 To configure the hugepages plugin you need to modify the configuration file to
165 ProcessType "secondary"
167 EnabledPortMask 0xffff
170 For more information on the plugin parameters, please see:
171 https://github.com/collectd/collectd/blob/master/src/collectd.conf.pod
173 Please note if you are configuring collectd with the **static DPDK library**
174 you must compile the DPDK library with the -fPIC flag:
178 $ make EXTRA_CFLAGS=-fPIC
180 You must also modify the configuration step when building collectd:
184 $ ./configure CFLAGS=" -lpthread -Wl,--whole-archive -Wl,-ldpdk -Wl,-lm -Wl,-lrt -Wl,-lpcap -Wl,-ldl -Wl,--no-whole-archive"
186 Please also note that if you are not building and installing DPDK system-wide
187 you will need to specify the specific paths to the header files and libraries
188 using LIBDPDK_CPPFLAGS and LIBDPDK_LDFLAGS. You will also need to add the DPDK
189 library symbols to the shared library path using ldconfig. Note that this
190 update to the shared library path is not persistant (i.e. it will not survive a
191 reboot). Pending a merge of https://github.com/collectd/collectd/pull/2073.
195 $ ./configure LIBDPDK_CPPFLAGS="path to DPDK header files" LIBDPDK_LDFLAGS="path to DPDK libraries"
199 Repo: https://github.com/collectd/collectd
203 Dependencies: None, but assumes hugepages are configured.
205 To configure some hugepages:
209 sudo mkdir -p /mnt/huge
210 sudo mount -t hugetlbfs nodev /mnt/huge
211 sudo echo 14336 > /sys/devices/system/node/node0/hugepages/hugepages-2048kB/nr_hugepages
213 Building and installing collectd:
217 $ git clone https://github.com/collectd/collectd.git
220 $ ./configure --enable-syslog --enable-logfile --enable-hugepages --enable-debug
224 This will install collectd to /opt/collectd
225 The collectd configuration file can be found at /opt/collectd/etc
226 To configure the hugepages plugin you need to modify the configuration file to
237 ValuesPercentage false
240 For more information on the plugin parameters, please see:
241 https://github.com/collectd/collectd/blob/master/src/collectd.conf.pod
245 Repo: https://github.com/collectd/collectd
251 * PQoS/Intel RDT library https://github.com/01org/intel-cmt-cat.git
254 Building and installing PQoS/Intel RDT library:
258 $ git clone https://github.com/01org/intel-cmt-cat.git
261 $ make install PREFIX=/usr
263 You will need to insert the msr kernel module:
269 Building and installing collectd:
273 $ git clone https://github.com/collectd/collectd.git
276 $ ./configure --enable-syslog --enable-logfile --with-libpqos=/usr/ --enable-debug
280 This will install collectd to /opt/collectd
281 The collectd configuration file can be found at /opt/collectd/etc
282 To configure the RDT plugin you need to modify the configuration file to
287 <LoadPlugin intel_rdt>
294 For more information on the plugin parameters, please see:
295 https://github.com/collectd/collectd/blob/master/src/collectd.conf.pod
299 Repo: https://github.com/maryamtahhan/collectd
301 Branch: feat_ipmi_events, feat_ipmi_analog
303 Dependencies: OpenIPMI library
305 The IPMI plugin is already implemented in the latest collectd and sensors
306 like temperature, voltage, fanspeed, current are already supported there.
307 The list of supported IPMI sensors has been extended and sensors like flow,
308 power are supported now. Also, a System Event Log (SEL) notification feature
311 * The feat_ipmi_events branch includes new SEL feature support in collectd
312 IPMI plugin. If this feature is enabled, the collectd IPMI plugin will
313 dispatch notifications about new events in System Event Log.
315 * The feat_ipmi_analog branch includes the support of extended IPMI sensors in
316 collectd IPMI plugin.
318 On Ubuntu, install the dependencies:
322 $ sudo apt-get install libopenipmi-dev
324 Enable IPMI support in the kernel:
328 $ sudo modprobe ipmi_devintf
329 $ sudo modprobe ipmi_si
331 **Note**: If HW supports IPMI, the ``/dev/ipmi0`` character device will be
334 Clone and install the collectd IPMI plugin:
338 $ git clone https://github.com/maryamtahhan/collectd
340 $ git checkout $BRANCH
342 $ ./configure --enable-syslog --enable-logfile --enable-debug
346 Where $BRANCH is feat_ipmi_events or feat_ipmi_analog.
348 This will install collectd to default folder ``/opt/collectd``. The collectd
349 configuration file (``collectd.conf``) can be found at ``/opt/collectd/etc``. To
350 configure the IPMI plugin you need to modify the file to include:
356 SELEnabled true # only feat_ipmi_events branch supports this
359 **Note**: By default, IPMI plugin will read all available analog sensor values,
360 dispatch the values to collectd and send SEL notifications.
362 For more information on the IPMI plugin parameters and SEL feature configuration,
364 https://github.com/maryamtahhan/collectd/blob/feat_ipmi_events/src/collectd.conf.pod
366 Extended analog sensors support doesn't require additional configuration. The usual
367 collectd IPMI documentation can be used:
369 - https://collectd.org/wiki/index.php/Plugin:IPMI
370 - https://collectd.org/documentation/manpages/collectd.conf.5.shtml#plugin_ipmi
374 - https://www.kernel.org/doc/Documentation/IPMI.txt
375 - http://www.intel.com/content/www/us/en/servers/ipmi/ipmi-second-gen-interface-spec-v2-rev1-1.html
379 Repo: https://github.com/collectd/collectd
385 Start by installing mcelog. Note: The kernel has to have CONFIG_X86_MCE
386 enabled. For 32bit kernels you need at least a 2.6,30 kernel.
392 $ apt-get update && apt-get install mcelog
398 $ git clone git://git.kernel.org/pub/scm/utils/cpu/mce/mcelog.git
403 $ cp mcelog.service /etc/systemd/system/
404 $ systemctl enable mcelog.service
405 $ systemctl start mcelog.service
408 Verify you got a /dev/mcelog. You can verify the daemon is running completely
415 This should query the information in the running daemon. If it prints nothing
416 that is fine (no errors logged yet). More info @
417 http://www.mcelog.org/installation.html
419 Modify the mcelog configuration file "/etc/mcelog/mcelog.conf" to include or
424 socket-path = /var/run/mcelog-client
426 Clone and install the collectd mcelog plugin:
430 $ git clone https://github.com/maryamtahhan/collectd
432 $ git checkout feat_ras
434 $ ./configure --enable-syslog --enable-logfile --enable-debug
438 This will install collectd to /opt/collectd
439 The collectd configuration file can be found at /opt/collectd/etc
440 To configure the mcelog plugin you need to modify the configuration file to
449 McelogClientSocket "/var/run/mcelog-client"
452 For more information on the plugin parameters, please see:
453 https://github.com/maryamtahhan/collectd/blob/feat_ras/src/collectd.conf.pod
455 Simulating a Machine Check Exception can be done in one of 3 ways:
457 * Running $make test in the mcelog cloned directory - mcelog test suite
461 **mcelog test suite:**
463 It is always a good idea to test an error handling mechanism before it is
464 really needed. mcelog includes a test suite. The test suite relies on
465 mce-inject which needs to be installed and in $PATH.
467 You also need the mce-inject kernel module configured (with
468 CONFIG_X86_MCE_INJECT=y), compiled, installed and loaded:
472 $ modprobe mce-inject
474 Then you can run the mcelog test suite with
480 This will inject different classes of errors and check that the mcelog triggers
481 runs. There will be some kernel messages about page offlining attempts. The
482 test will also lose a few pages of memory in your system (not significant)
483 **Note this test will kill any running mcelog, which needs to be restarted
484 manually afterwards**.
487 A utility to inject corrected, uncorrected and fatal machine check exceptions
491 $ git clone https://git.kernel.org/pub/scm/utils/cpu/mce/mce-inject.git
494 $ modprobe mce-inject
496 Modify the test/corrected script to include the following:
501 STATUS 0xcc00008000010090
507 $ ./mce-inject < test/corrected
509 **Note: the uncorrected and fatal scripts under test will cause a platform reset.
510 Only the fatal script generates the memory errors**. In order to quickly
511 emulate uncorrected memory errors and avoid host reboot following test errors
512 from mce-test suite can be injected:
516 $ mce-inject mce-test/cases/coverage/soft-inj/recoverable_ucr/data/srao_mem_scrub
520 In addition an more in-depth test of the Linux kernel machine check facilities
521 can be done with the mce-test test suite. mce-test supports testing uncorrected
522 error handling, real error injection, handling of different soft offlining
523 cases, and other tests.
525 **Corrected memory error injection:**
527 To inject corrected memory errors:
529 * Remove sb_edac and edac_core kernel modules: rmmod sb_edac rmmod edac_core
530 * Insert einj module: modprobe einj param_extension=1
531 * Inject an error by specifying details (last command should be repeated at least two times):
535 $ APEI_IF=/sys/kernel/debug/apei/einj
536 $ echo 0x8 > $APEI_IF/error_type
537 $ echo 0x01f5591000 > $APEI_IF/param1
538 $ echo 0xfffffffffffff000 > $APEI_IF/param2
539 $ echo 1 > $APEI_IF/notrigger
540 $ echo 1 > $APEI_IF/error_inject
542 * Check the MCE statistic: mcelog --client. Check the mcelog log for injected error details: less /var/log/mcelog.
545 ^^^^^^^^^^^^^^^^^^^^^
546 OvS Events Repo: https://github.com/collectd/collectd
548 OvS Stats Repo: https://github.com/maryamtahhan/collectd
550 OvS Events Branch: master
552 OvS Stats Branch:feat_ovs_stats
554 OvS Events MIBs: The SNMP OVS interface link status is provided by standard
555 IF-MIB (http://www.net-snmp.org/docs/mibs/IF-MIB.txt)
557 Dependencies: Open vSwitch, Yet Another JSON Library (https://github.com/lloyd/yajl)
559 On Ubuntu, install the dependencies:
563 $ sudo apt-get install libyajl-dev openvswitch-switch
565 Start the Open vSwitch service:
569 $ sudo service openvswitch-switch start
571 configure the ovsdb-server manager:
575 $ sudo ovs-vsctl set-manager ptcp:6640
577 Clone and install the collectd ovs plugin:
583 $ git checkout $BRANCH
585 $ ./configure --enable-syslog --enable-logfile --enable-debug
589 where $REPO is one of the repos listed at the top of this section.
591 Where $BRANCH is master or feat_ovs_stats.
593 This will install collectd to /opt/collectd. The collectd configuration file
594 can be found at /opt/collectd/etc. To configure the OVS events plugin you
595 need to modify the configuration file to include:
599 <LoadPlugin ovs_events>
602 <Plugin "ovs_events">
604 Socket "/var/run/openvswitch/db.sock"
605 Interfaces "br0" "veth0"
606 SendNotification false
610 To configure the OVS stats plugin you need to modify the configuration file
615 <LoadPlugin ovs_stats>
621 Socket "/var/run/openvswitch/db.sock"
622 Bridges "br0" "br_ext"
625 For more information on the plugin parameters, please see:
626 https://github.com/collectd/collectd/blob/master/src/collectd.conf.pod
628 https://github.com/maryamtahhan/collectd/blob/feat_ovs_stats/src/collectd.conf.pod
632 Repo: https://github.com/maryamtahhan/collectd/
636 Dependencies: NET-SNMP library
638 Start by installing net-snmp and dependencies.
644 $ apt-get install snmp snmp-mibs-downloader snmpd libsnmp-dev
645 $ systemctl start snmpd.service
649 Become root to install net-snmp dependencies
653 $ apt-get install libperl-dev
655 Clone and build net-snmp
659 $ git clone https://github.com/haad/net-snmp.git
661 $ ./configure --with-persistent-directory="/var/net-snmp" --with-systemd --enable-shared --prefix=/usr
670 Copy default configuration to persistent folder
674 $ cp EXAMPLE.conf /usr/share/snmp/snmpd.conf
676 Set library path and default MIB configuration
681 $ echo export LD_LIBRARY_PATH=/usr/lib >> .bashrc
682 $ net-snmp-config --default-mibdirs
683 $ net-snmp-config --snmpconfpath
685 Configure snmpd as a service
690 $ cp ./dist/snmpd.service /etc/systemd/system/
691 $ systemctl enable snmpd.service
692 $ systemctl start snmpd.service
694 Add the following line to snmpd.conf configuration file
695 "/usr/share/snmp/snmpd.conf" to make all OID tree visible for SNMP clients:
699 view systemonly included .1
701 To verify that SNMP is working you can get IF-MIB table using SNMP client
702 to view the list of Linux interfaces:
706 $ snmpwalk -v 2c -c public localhost IF-MIB::interfaces
708 Clone and install the collectd snmp_agent plugin:
712 $ git clone https://github.com/maryamtahhan/collectd
714 $ git checkout feat_snmp
716 $ ./configure --enable-syslog --enable-logfile --enable-debug --enable-snmp --with-libnetsnmp
720 This will install collectd to /opt/collectd
721 The collectd configuration file can be found at /opt/collectd/etc
722 **SNMP Agent plugin is a generic plugin and cannot work without configuration**.
723 To configure the snmp_agent plugin you need to modify the configuration file to
724 include OIDs mapped to collectd types. The following example maps scalar
725 memAvailReal OID to value represented as free memory type of memory plugin:
729 LoadPlugin snmp_agent
730 <Plugin "snmp_agent">
731 <Data "memAvailReal">
735 OIDs "1.3.6.1.4.1.2021.4.6.0"
739 For more information on the plugin parameters, please see:
740 https://github.com/maryamtahhan/collectd/blob/feat_snmp/src/collectd.conf.pod
742 For more details on AgentX subagent, please see:
743 http://www.net-snmp.org/tutorial/tutorial-5/toolkit/demon/
745 Installing collectd as a service
746 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
747 **NOTE**: In an OPNFV installation, collectd is installed and configured as a
750 Collectd service scripts are available in the collectd/contrib directory.
751 To install collectd as a service:
755 $ sudo cp contrib/systemd.collectd.service /etc/systemd/system/
756 $ cd /etc/systemd/system/
757 $ sudo mv systemd.collectd.service collectd.service
758 $ sudo chmod +x collectd.service
760 Modify collectd.service
765 ExecStart=/opt/collectd/sbin/collectd
766 EnvironmentFile=-/opt/collectd/etc/
767 EnvironmentFile=-/opt/collectd/etc/
768 CapabilityBoundingSet=CAP_SETUID CAP_SETGID
774 $ sudo systemctl daemon-reload
775 $ sudo systemctl start collectd.service
776 $ sudo systemctl status collectd.service should show success
778 Additional useful plugins
779 ^^^^^^^^^^^^^^^^^^^^^^^^^^
781 * **Exec Plugin** : Can be used to show you when notifications are being
782 generated by calling a bash script that dumps notifications to file. (handy
783 for debug). Modify /opt/collectd/etc/collectd.conf:
789 # Exec "user:group" "/path/to/exec"
790 NotificationExec "user" "<path to barometer>/barometer/src/collectd/collectd_sample_configs/write_notification.sh"
793 write_notification.sh (just writes the notification passed from exec through
794 STDIN to a file (/tmp/notifications)):
799 rm -f /tmp/notifications
802 echo $x$y >> /tmp/notifications
805 output to /tmp/notifications should look like:
815 TypeInstance:link_status
816 uuid:f2aafeec-fa98-4e76-aec5-18ae9fc74589
818 linkstate of "br-ex" interface has been changed to "DOWN"
820 * **logfile plugin**: Can be used to log collectd activity. Modify
821 /opt/collectd/etc/collectd.conf to include:
828 File "/var/log/collectd.log"
834 Monitoring Interfaces and Openstack Support
835 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
836 .. Figure:: monitoring_interfaces.png
838 Monitoring Interfaces and Openstack Support
840 The figure above shows the DPDK L2 forwarding application running on a compute
841 node, sending and receiving traffic. collectd is also running on this compute
842 node retrieving the stats periodically from DPDK through the dpdkstat plugin
843 and publishing the retrieved stats to Ceilometer through the ceilometer plugin.
845 To see this demo in action please checkout: `Barometer OPNFV Summit demo`_
849 .. [1] https://collectd.org/wiki/index.php/Naming_schema
850 .. [2] https://github.com/collectd/collectd/blob/master/src/daemon/plugin.h
851 .. [3] https://collectd.org/wiki/index.php/Value_list_t
852 .. [4] https://collectd.org/wiki/index.php/Data_set
853 .. [5] https://collectd.org/documentation/manpages/types.db.5.shtml
854 .. [6] https://collectd.org/wiki/index.php/Data_source
855 .. [7] https://collectd.org/wiki/index.php/Meta_Data_Interface
857 .. _Barometer OPNFV Summit demo: https://prezi.com/kjv6o8ixs6se/software-fastpath-service-quality-metrics-demo/
858 .. _ceilometer plugin: https://github.com/openstack/collectd-ceilometer-plugin/tree/stable/mitaka