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 * *dpdkevents plugin*: A read plugin that retrieves DPDK link status and DPDK
31 forwarding cores liveliness status (DPDK Keep Alive).
33 * `gnocchi plugin`_: A write plugin that pushes the retrieved stats to
34 Gnocchi. It's capable of pushing any stats read through collectd to
35 Gnocchi, not just the DPDK stats.
37 * `aodh plugin`_: A notification plugin that pushes events to Aodh, and
38 creates/updates alarms appropriately.
40 * *hugepages plugin*: A read plugin that retrieves the number of available
41 and free hugepages on a platform as well as what is available in terms of
44 * *Open vSwitch events Plugin*: A read plugin that retrieves events from OVS.
46 * *Open vSwitch stats Plugin*: A read plugin that retrieves flow and interface
49 * *mcelog plugin*: A read plugin that uses mcelog client protocol to check for
50 memory Machine Check Exceptions and sends the stats for reported exceptions
52 * *RDT plugin*: A read plugin that provides the last level cache utilization and
53 memory bandwidth utilization
55 * *virt*: A read plugin that uses virtualization API *libvirt* to gather
56 statistics about virtualized guests on a system directly from the hypervisor,
57 without a need to install collectd instance on the guest.
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 All the plugins above are available on the collectd master, except for the
67 Gnocchi and Aodh plugins as they are Python-based plugins and only C plugins
68 are accepted by the collectd community. The Gnocchi and Aodh plugins live in
69 the OpenStack repositories.
71 Other plugins existing as a pull request into collectd master:
73 * *Legacy/IPMI*: A read plugin that reports platform thermals, voltages,
74 fanspeed, current, flow, power etc. Also, the plugin monitors Intelligent
75 Platform Management Interface (IPMI) System Event Log (SEL) and sends the
76 appropriate notifications based on monitored SEL events.
78 * *PCIe AER*: A read plugin that monitors PCIe standard and advanced errors and
79 sends notifications about those errors.
82 Third party application in Barometer repository:
84 * *Open vSwitch PMD stats*: An aplication that retrieves PMD stats from OVS. It is run
87 **Plugins included in the Danube release:**
94 collectd capabilities and usage
95 ------------------------------------
96 .. Describe the specific capabilities and usage for <XYZ> feature.
97 .. Provide enough information that a user will be able to operate the feature on a deployed scenario.
99 .. note:: Plugins included in the OPNFV D release will be built-in to the fuel
100 plugin and available in the /opt/opnfv directory on the fuel master. You don't
101 need to clone the barometer/collectd repos to use these, but you can configure
102 them as shown in the examples below.
104 The collectd plugins in OPNFV are configured with reasonable defaults, but can
107 Building all Barometer upstreamed plugins from scratch
108 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
109 The plugins that have been merged to the collectd master branch can all be
110 built and configured through the barometer repository.
113 * sudo permissions are required to install collectd.
114 * These are instructions for Ubuntu 16.04
116 To build all the upstream plugins, clone the barometer repo:
120 $ git clone https://gerrit.opnfv.org/gerrit/barometer
122 To install collectd as a service and install all it's dependencies:
126 $ cd barometer/systems && ./build_base_machine.sh
128 This will install collectd as a service and the base install directory
129 will be /opt/collectd.
131 Sample configuration files can be found in '/opt/collectd/etc/collectd.conf.d'
134 If you don't want to use one of the Barometer plugins, simply remove the
135 sample config file from '/opt/collectd/etc/collectd.conf.d'
137 If you plan on using the Exec plugin (for OVS_PMD_STATS or for executing scripts
138 on notification generation), the plugin requires a non-root
139 user to execute scripts. By default, `collectd_exec` user is used in the exec.conf
140 provided in the sample configurations directory under src/collectd in the Barometer
141 repo. The scripts *DO NOT* create this user. You need to create this user before you
142 run build_base_machine.sh. Or modify configuration in the sample configurations
143 directory under src/collectd to use another existing non root user before running
144 run build_base_machine.sh.
147 If you are using any Open vSwitch plugins you need to run:
151 $ sudo ovs-vsctl set-manager ptcp:6640
153 After this, you should be able to start collectd as a service
157 $ sudo systemctl status collectd
159 If you want to use granfana to display the metrics you collect, please see:
162 For more information on configuring and installing OpenStack plugins for
163 collectd, check out the `collectd-ceilometer-plugin GSG`_.
165 Below is the per plugin installation and configuration guide, if you only want
166 to install some/particular plugins.
170 Repo: https://github.com/collectd/collectd
174 Dependencies: DPDK (http://dpdk.org/)
176 .. note:: DPDK statistics plugin requires DPDK version 16.04 or later
178 To build and install DPDK to /usr please see:
179 https://github.com/collectd/collectd/blob/master/docs/BUILD.dpdkstat.md
181 Building and installing collectd:
185 $ git clone https://github.com/collectd/collectd.git
188 $ ./configure --enable-syslog --enable-logfile --enable-debug
192 .. note:: If DPDK was installed in a non standard location you will need to
193 specify paths to the header files and libraries using *LIBDPDK_CPPFLAGS* and
194 *LIBDPDK_LDFLAGS*. You will also need to add the DPDK library symbols to the
195 shared library path using *ldconfig*. Note that this update to the shared
196 library path is not persistant (i.e. it will not survive a reboot).
198 Example of specifying custom paths to DPDK headers and libraries:
202 $ ./configure LIBDPDK_CPPFLAGS="path to DPDK header files" LIBDPDK_LDFLAGS="path to DPDK libraries"
204 This will install collectd to /opt/collectd
205 The collectd configuration file can be found at /opt/collectd/etc
207 To configure the dpdkstats plugin you need to modify the configuration file to
217 ProcessType "secondary"
220 EnabledPortMask 0xffff
221 PortName "interface1"
222 PortName "interface2"
226 To configure the dpdkevents plugin you need to modify the configuration file to
231 LoadPlugin dpdkevents
232 <Plugin "dpdkevents">
237 ProcessType "secondary"
240 <Event "link_status">
241 SendEventsOnUpdate true
242 EnabledPortMask 0xffff
243 PortName "interface1"
244 PortName "interface2"
245 SendNotification false
248 SendEventsOnUpdate true
250 KeepAliveShmName "/dpdk_keepalive_shm_name"
251 SendNotification false
255 .. note:: Currently, the DPDK library doesn’t support API to de-initialize
256 the DPDK resources allocated on the initialization. It means, the collectd
257 plugin will not be able to release the allocated DPDK resources
258 (locks/memory/pci bindings etc.) correctly on collectd shutdown or reinitialize
259 the DPDK library if primary DPDK process is restarted. The only way to release
260 those resources is to terminate the process itself. For this reason, the plugin
261 forks off a separate collectd process. This child process becomes a secondary
262 DPDK process which can be run on specific CPU cores configured by user through
263 collectd configuration file (“Coremask” EAL configuration option, the
264 hexadecimal bitmask of the cores to run on).
266 For more information on the plugin parameters, please see:
267 https://github.com/collectd/collectd/blob/master/src/collectd.conf.pod
269 .. note:: dpdkstat plugin initialization time depends on read interval. It
270 requires 5 read cycles to set up internal buffers and states. During that time
271 no statistics are submitted. Also if plugin is running and the number of DPDK
272 ports is increased, internal buffers are resized. That requires 3 read cycles
273 and no port statistics are submitted in that time.
275 The Address-Space Layout Randomization (ASLR) security feature in Linux should be
276 disabled, in order for the same hugepage memory mappings to be present in all
277 DPDK multi-process applications.
283 $ sudo echo 0 > /proc/sys/kernel/randomize_va_space
285 To fully enable ASLR:
289 $ sudo echo 2 > /proc/sys/kernel/randomize_va_space
291 .. warning:: Disabling Address-Space Layout Randomization (ASLR) may have security
292 implications. It is recommended to be disabled only when absolutely necessary,
293 and only when all implications of this change have been understood.
295 For more information on multi-process support, please see:
296 http://dpdk.org/doc/guides/prog_guide/multi_proc_support.html
298 **DPDK stats plugin limitations:**
300 1. The DPDK primary process application should use the same version of DPDK
301 that collectd DPDK plugin is using;
303 2. L2 statistics are only supported;
305 3. The plugin has been tested on Intel NIC’s only.
307 **DPDK stats known issues:**
309 * DPDK port visibility
311 When network port controlled by Linux is bound to DPDK driver, the port
312 will not be available in the OS. It affects the SNMP write plugin as those
313 ports will not be present in standard IF-MIB. Thus addition work is
314 required to be done to support DPDK ports and statistics.
318 Repo: https://github.com/collectd/collectd
322 Dependencies: None, but assumes hugepages are configured.
324 To configure some hugepages:
328 sudo mkdir -p /mnt/huge
329 sudo mount -t hugetlbfs nodev /mnt/huge
330 sudo echo 14336 > /sys/devices/system/node/node0/hugepages/hugepages-2048kB/nr_hugepages
332 Building and installing collectd:
336 $ git clone https://github.com/collectd/collectd.git
339 $ ./configure --enable-syslog --enable-logfile --enable-hugepages --enable-debug
343 This will install collectd to /opt/collectd
344 The collectd configuration file can be found at /opt/collectd/etc
345 To configure the hugepages plugin you need to modify the configuration file to
356 ValuesPercentage false
359 For more information on the plugin parameters, please see:
360 https://github.com/collectd/collectd/blob/master/src/collectd.conf.pod
364 Repo: https://github.com/collectd/collectd
370 * PQoS/Intel RDT library https://github.com/01org/intel-cmt-cat.git
373 Building and installing PQoS/Intel RDT library:
377 $ git clone https://github.com/01org/intel-cmt-cat.git
380 $ make install PREFIX=/usr
382 You will need to insert the msr kernel module:
388 Building and installing collectd:
392 $ git clone https://github.com/collectd/collectd.git
395 $ ./configure --enable-syslog --enable-logfile --with-libpqos=/usr/ --enable-debug
399 This will install collectd to /opt/collectd
400 The collectd configuration file can be found at /opt/collectd/etc
401 To configure the RDT plugin you need to modify the configuration file to
406 <LoadPlugin intel_rdt>
413 For more information on the plugin parameters, please see:
414 https://github.com/collectd/collectd/blob/master/src/collectd.conf.pod
418 Repo: https://github.com/maryamtahhan/collectd
420 Branch: feat_ipmi_events, feat_ipmi_analog
422 Dependencies: OpenIPMI library (http://openipmi.sourceforge.net/)
424 The IPMI plugin is already implemented in the latest collectd and sensors
425 like temperature, voltage, fanspeed, current are already supported there.
426 The list of supported IPMI sensors has been extended and sensors like flow,
427 power are supported now. Also, a System Event Log (SEL) notification feature
430 * The feat_ipmi_events branch includes new SEL feature support in collectd
431 IPMI plugin. If this feature is enabled, the collectd IPMI plugin will
432 dispatch notifications about new events in System Event Log.
434 * The feat_ipmi_analog branch includes the support of extended IPMI sensors in
435 collectd IPMI plugin.
437 **Install dependencies**
439 On Ubuntu, the OpenIPMI library can be installed via apt package manager:
443 $ sudo apt-get install libopenipmi-dev
445 Anyway, it's recommended to use the latest version of the OpenIPMI library as
446 it includes fixes of known issues which aren't included in standard OpenIPMI
447 library package. The latest version of the library can be found at
448 https://sourceforge.net/p/openipmi/code/ci/master/tree/. Steps to install the
449 library from sources are described below.
451 Remove old version of OpenIPMI library:
455 $ sudo apt-get remove libopenipmi-dev
457 Download OpenIPMI library sources:
461 $ git clone https://git.code.sf.net/p/openipmi/code openipmi-code
464 Patch the OpenIPMI pkg-config file to provide correct compilation flags
465 for collectd IPMI plugin:
469 diff --git a/OpenIPMIpthread.pc.in b/OpenIPMIpthread.pc.in
470 index 59b52e5..fffa0d0 100644
471 --- a/OpenIPMIpthread.pc.in
472 +++ b/OpenIPMIpthread.pc.in
473 @@ -6,6 +6,6 @@ includedir=@includedir@
474 Name: OpenIPMIpthread
475 Description: Pthread OS handler for OpenIPMI
477 -Requires: OpenIPMI pthread
479 Libs: -L${libdir} -lOpenIPMIutils -lOpenIPMIpthread
480 -Cflags: -I${includedir}
481 +Cflags: -I${includedir} -pthread
483 Build and install OpenIPMI library:
487 $ autoreconf --install
488 $ ./configure --prefix=/usr
492 Enable IPMI support in the kernel:
496 $ sudo modprobe ipmi_devintf
497 $ sudo modprobe ipmi_si
500 If HW supports IPMI, the ``/dev/ipmi0`` character device will be
503 Clone and install the collectd IPMI plugin:
507 $ git clone https://github.com/maryamtahhan/collectd
509 $ git checkout $BRANCH
511 $ ./configure --enable-syslog --enable-logfile --enable-debug
515 Where $BRANCH is feat_ipmi_events or feat_ipmi_analog.
517 This will install collectd to default folder ``/opt/collectd``. The collectd
518 configuration file (``collectd.conf``) can be found at ``/opt/collectd/etc``. To
519 configure the IPMI plugin you need to modify the file to include:
525 SELEnabled true # only feat_ipmi_events branch supports this
529 By default, IPMI plugin will read all available analog sensor values,
530 dispatch the values to collectd and send SEL notifications.
532 For more information on the IPMI plugin parameters and SEL feature configuration,
534 https://github.com/maryamtahhan/collectd/blob/feat_ipmi_events/src/collectd.conf.pod
536 Extended analog sensors support doesn't require additional configuration. The usual
537 collectd IPMI documentation can be used:
539 - https://collectd.org/wiki/index.php/Plugin:IPMI
540 - https://collectd.org/documentation/manpages/collectd.conf.5.shtml#plugin_ipmi
544 - https://www.kernel.org/doc/Documentation/IPMI.txt
545 - http://www.intel.com/content/www/us/en/servers/ipmi/ipmi-second-gen-interface-spec-v2-rev1-1.html
549 Repo: https://github.com/collectd/collectd
555 Start by installing mcelog.
558 The kernel has to have CONFIG_X86_MCE enabled. For 32bit kernels you
559 need at least a 2.6,30 kernel.
565 $ apt-get update && apt-get install mcelog
571 $ git clone git://git.kernel.org/pub/scm/utils/cpu/mce/mcelog.git
576 $ cp mcelog.service /etc/systemd/system/
577 $ systemctl enable mcelog.service
578 $ systemctl start mcelog.service
581 Verify you got a /dev/mcelog. You can verify the daemon is running completely
588 This should query the information in the running daemon. If it prints nothing
589 that is fine (no errors logged yet). More info @
590 http://www.mcelog.org/installation.html
592 Modify the mcelog configuration file "/etc/mcelog/mcelog.conf" to include or
597 socket-path = /var/run/mcelog-client
599 Clone and install the collectd mcelog plugin:
603 $ git clone https://github.com/maryamtahhan/collectd
605 $ git checkout feat_ras
607 $ ./configure --enable-syslog --enable-logfile --enable-debug
611 This will install collectd to /opt/collectd
612 The collectd configuration file can be found at /opt/collectd/etc
613 To configure the mcelog plugin you need to modify the configuration file to
622 McelogClientSocket "/var/run/mcelog-client"
625 For more information on the plugin parameters, please see:
626 https://github.com/maryamtahhan/collectd/blob/feat_ras/src/collectd.conf.pod
628 Simulating a Machine Check Exception can be done in one of 3 ways:
630 * Running $make test in the mcelog cloned directory - mcelog test suite
634 **mcelog test suite:**
636 It is always a good idea to test an error handling mechanism before it is
637 really needed. mcelog includes a test suite. The test suite relies on
638 mce-inject which needs to be installed and in $PATH.
640 You also need the mce-inject kernel module configured (with
641 CONFIG_X86_MCE_INJECT=y), compiled, installed and loaded:
645 $ modprobe mce-inject
647 Then you can run the mcelog test suite with
653 This will inject different classes of errors and check that the mcelog triggers
654 runs. There will be some kernel messages about page offlining attempts. The
655 test will also lose a few pages of memory in your system (not significant)
657 This test will kill any running mcelog, which needs to be restarted
662 A utility to inject corrected, uncorrected and fatal machine check exceptions
666 $ git clone https://git.kernel.org/pub/scm/utils/cpu/mce/mce-inject.git
669 $ modprobe mce-inject
671 Modify the test/corrected script to include the following:
676 STATUS 0xcc00008000010090
682 $ ./mce-inject < test/corrected
685 The uncorrected and fatal scripts under test will cause a platform reset.
686 Only the fatal script generates the memory errors**. In order to quickly
687 emulate uncorrected memory errors and avoid host reboot following test errors
688 from mce-test suite can be injected:
692 $ mce-inject mce-test/cases/coverage/soft-inj/recoverable_ucr/data/srao_mem_scrub
696 In addition an more in-depth test of the Linux kernel machine check facilities
697 can be done with the mce-test test suite. mce-test supports testing uncorrected
698 error handling, real error injection, handling of different soft offlining
699 cases, and other tests.
701 **Corrected memory error injection:**
703 To inject corrected memory errors:
705 * Remove sb_edac and edac_core kernel modules: rmmod sb_edac rmmod edac_core
706 * Insert einj module: modprobe einj param_extension=1
707 * Inject an error by specifying details (last command should be repeated at least two times):
711 $ APEI_IF=/sys/kernel/debug/apei/einj
712 $ echo 0x8 > $APEI_IF/error_type
713 $ echo 0x01f5591000 > $APEI_IF/param1
714 $ echo 0xfffffffffffff000 > $APEI_IF/param2
715 $ echo 1 > $APEI_IF/notrigger
716 $ echo 1 > $APEI_IF/error_inject
718 * Check the MCE statistic: mcelog --client. Check the mcelog log for injected error details: less /var/log/mcelog.
721 ^^^^^^^^^^^^^^^^^^^^^
722 OvS Plugins Repo: https://github.com/collectd/collectd
724 OvS Plugins Branch: master
726 OvS Events MIBs: The SNMP OVS interface link status is provided by standard
727 IF-MIB (http://www.net-snmp.org/docs/mibs/IF-MIB.txt)
729 Dependencies: Open vSwitch, Yet Another JSON Library (https://github.com/lloyd/yajl)
731 On Ubuntu, install the dependencies:
735 $ sudo apt-get install libyajl-dev openvswitch-switch
737 Start the Open vSwitch service:
741 $ sudo service openvswitch-switch start
743 configure the ovsdb-server manager:
747 $ sudo ovs-vsctl set-manager ptcp:6640
749 Clone and install the collectd ovs plugin:
755 $ git checkout master
757 $ ./configure --enable-syslog --enable-logfile --enable-debug
761 This will install collectd to /opt/collectd. The collectd configuration file
762 can be found at /opt/collectd/etc. To configure the OVS events plugin you
763 need to modify the configuration file to include:
767 <LoadPlugin ovs_events>
770 <Plugin "ovs_events">
772 Socket "/var/run/openvswitch/db.sock"
773 Interfaces "br0" "veth0"
774 SendNotification false
778 To configure the OVS stats plugin you need to modify the configuration file
783 <LoadPlugin ovs_stats>
789 Socket "/var/run/openvswitch/db.sock"
790 Bridges "br0" "br_ext"
793 For more information on the plugin parameters, please see:
794 https://github.com/collectd/collectd/blob/master/src/collectd.conf.pod
798 Repo: https://gerrit.opnfv.org/gerrit/barometer
801 1. Open vSwitch dependencies are installed.
802 2. Open vSwitch service is running.
803 3. Ovsdb-server manager is configured.
804 You can refer `Open vSwitch Plugins`_ section above for each one of them.
806 OVS PMD stats application is run through the exec plugin.
808 To configure the OVS PMD stats application you need to modify the exec plugin configuration
817 Exec "user:group" "<path to ovs_pmd_stat.sh>"
818 #NotificationExec "nobody" "/usr/lib/collectd/notify.sh"
821 .. note:: Exec plugin configuration has to be changed to use appropriate user before starting collectd service.
823 ovs_pmd_stat.sh calls the script for OVS PMD stats application with its argument:
827 sudo python /usr/local/src/ovs_pmd_stats.py" "--socket-pid-file"
828 "/var/run/openvswitch/ovs-vswitchd.pid"
832 Repo: https://github.com/maryamtahhan/collectd/
836 Dependencies: NET-SNMP library
838 Start by installing net-snmp and dependencies.
844 $ apt-get install snmp snmp-mibs-downloader snmpd libsnmp-dev
845 $ systemctl start snmpd.service
849 Become root to install net-snmp dependencies
853 $ apt-get install libperl-dev
855 Clone and build net-snmp
859 $ git clone https://github.com/haad/net-snmp.git
861 $ ./configure --with-persistent-directory="/var/net-snmp" --with-systemd --enable-shared --prefix=/usr
870 Copy default configuration to persistent folder
874 $ cp EXAMPLE.conf /usr/share/snmp/snmpd.conf
876 Set library path and default MIB configuration
881 $ echo export LD_LIBRARY_PATH=/usr/lib >> .bashrc
882 $ net-snmp-config --default-mibdirs
883 $ net-snmp-config --snmpconfpath
885 Configure snmpd as a service
890 $ cp ./dist/snmpd.service /etc/systemd/system/
891 $ systemctl enable snmpd.service
892 $ systemctl start snmpd.service
894 Add the following line to snmpd.conf configuration file
895 "/usr/share/snmp/snmpd.conf" to make all OID tree visible for SNMP clients:
899 view systemonly included .1
901 To verify that SNMP is working you can get IF-MIB table using SNMP client
902 to view the list of Linux interfaces:
906 $ snmpwalk -v 2c -c public localhost IF-MIB::interfaces
908 Clone and install the collectd snmp_agent plugin:
912 $ git clone https://github.com/maryamtahhan/collectd
914 $ git checkout feat_snmp
916 $ ./configure --enable-syslog --enable-logfile --enable-debug --enable-snmp --with-libnetsnmp
920 This will install collectd to /opt/collectd
921 The collectd configuration file can be found at /opt/collectd/etc
922 **SNMP Agent plugin is a generic plugin and cannot work without configuration**.
923 To configure the snmp_agent plugin you need to modify the configuration file to
924 include OIDs mapped to collectd types. The following example maps scalar
925 memAvailReal OID to value represented as free memory type of memory plugin:
929 LoadPlugin snmp_agent
930 <Plugin "snmp_agent">
931 <Data "memAvailReal">
935 OIDs "1.3.6.1.4.1.2021.4.6.0"
941 * Object instance with Counter64 type is not supported in SNMPv1. When GetNext
942 request is received, Counter64 type objects will be skipped. When Get
943 request is received for Counter64 type object, the error will be returned.
944 * Interfaces that are not visible to Linux like DPDK interfaces cannot be
945 retreived using standard IF-MIB tables.
947 For more information on the plugin parameters, please see:
948 https://github.com/maryamtahhan/collectd/blob/feat_snmp/src/collectd.conf.pod
950 For more details on AgentX subagent, please see:
951 http://www.net-snmp.org/tutorial/tutorial-5/toolkit/demon/
955 Repo: https://github.com/maryamtahhan/collectd
957 Branch: feat_libvirt_upstream
959 Dependencies: libvirt (https://libvirt.org/), libxml2
961 On Ubuntu, install the dependencies:
965 $ sudo apt-get install libxml2-dev
969 libvirt version in package manager might be quite old and offer only limited
970 functionality. Hence, building and installing libvirt from sources is recommended.
971 Detailed instructions can bet found at:
972 https://libvirt.org/compiling.html
974 Certain metrics provided by the plugin have a requirement on a minimal version of
975 the libvirt API. *File system information* statistics require a *Guest Agent (GA)*
976 to be installed and configured in a VM. User must make sure that installed GA
977 version supports retrieving file system information. Number of *Performance monitoring events*
978 metrics depends on running libvirt daemon version.
980 .. note:: Please keep in mind that RDT metrics (part of *Performance monitoring
981 events*) have to be supported by hardware. For more details on hardware support,
983 https://github.com/01org/intel-cmt-cat
985 Additionally perf metrics **cannot** be collected if *Intel RDT* plugin is enabled.
987 libvirt version can be checked with following commands:
994 .. table:: Extended statistics requirements
996 +-------------------------------+--------------------------+-------------+
997 | Statistic | Min. libvirt API version | Requires GA |
998 +===============================+==========================+=============+
999 | Domain reason | 0.9.2 | No |
1000 +-------------------------------+--------------------------+-------------+
1001 | Disk errors | 0.9.10 | No |
1002 +-------------------------------+--------------------------+-------------+
1003 | Job statistics | 1.2.9 | No |
1004 +-------------------------------+--------------------------+-------------+
1005 | File system information | 1.2.11 | Yes |
1006 +-------------------------------+--------------------------+-------------+
1007 | Performance monitoring events | 1.3.3 | No |
1008 +-------------------------------+--------------------------+-------------+
1010 Start libvirt daemon:
1014 $ systemctl start libvirtd
1016 Create domain (VM) XML configuration file. For more information on domain XML
1017 format and examples, please see:
1018 https://libvirt.org/formatdomain.html
1020 .. note:: Installing additional hypervisor dependencies might be required before
1021 deploying virtual machine.
1023 Create domain, based on created XML file:
1027 $ virsh define DOMAIN_CFG_FILE.xml
1033 $ virsh start DOMAIN_NAME
1035 Check if domain is running:
1041 Check list of available *Performance monitoring events* and their settings:
1045 $ virsh perf DOMAIN_NAME
1047 Enable or disable *Performance monitoring events* for domain:
1051 $ virsh perf DOMAIN_NAME [--enable | --disable] EVENT_NAME --live
1053 Clone and install the collectd virt plugin:
1059 $ git checkout $BRANCH
1061 $ ./configure --enable-syslog --enable-logfile --enable-debug
1065 Where ``$REPO`` and ``$BRANCH`` are equal to information provided above.
1067 This will install collectd to ``/opt/collectd``. The collectd configuration file
1068 ``collectd.conf`` can be found at ``/opt/collectd/etc``. To load the virt plugin
1069 user needs to modify the configuration file to include:
1075 Additionally, user can specify plugin configuration parameters in this file,
1076 such as connection URI, domain name and much more. By default extended virt plugin
1077 statistics are disabled. They can be enabled with ``ExtraStats`` option.
1083 ExtraStats "cpu_util disk disk_err domain_state fs_info job_stats_background pcpu perf vcpupin"
1086 For more information on the plugin parameters, please see:
1087 https://github.com/maryamtahhan/collectd/blob/feat_libvirt_upstream/src/collectd.conf.pod
1089 Installing collectd as a service
1090 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1091 **NOTE**: In an OPNFV installation, collectd is installed and configured as a
1094 Collectd service scripts are available in the collectd/contrib directory.
1095 To install collectd as a service:
1099 $ sudo cp contrib/systemd.collectd.service /etc/systemd/system/
1100 $ cd /etc/systemd/system/
1101 $ sudo mv systemd.collectd.service collectd.service
1102 $ sudo chmod +x collectd.service
1104 Modify collectd.service
1109 ExecStart=/opt/collectd/sbin/collectd
1110 EnvironmentFile=-/opt/collectd/etc/
1111 EnvironmentFile=-/opt/collectd/etc/
1112 CapabilityBoundingSet=CAP_SETUID CAP_SETGID
1118 $ sudo systemctl daemon-reload
1119 $ sudo systemctl start collectd.service
1120 $ sudo systemctl status collectd.service should show success
1122 Additional useful plugins
1123 ^^^^^^^^^^^^^^^^^^^^^^^^^^
1125 * **Exec Plugin** : Can be used to show you when notifications are being
1126 generated by calling a bash script that dumps notifications to file. (handy
1127 for debug). Modify /opt/collectd/etc/collectd.conf:
1133 # Exec "user:group" "/path/to/exec"
1134 NotificationExec "user" "<path to barometer>/barometer/src/collectd/collectd_sample_configs/write_notification.sh"
1137 write_notification.sh (just writes the notification passed from exec through
1138 STDIN to a file (/tmp/notifications)):
1143 rm -f /tmp/notifications
1146 echo $x$y >> /tmp/notifications
1149 output to /tmp/notifications should look like:
1157 PluginInstance:br-ex
1159 TypeInstance:link_status
1160 uuid:f2aafeec-fa98-4e76-aec5-18ae9fc74589
1162 linkstate of "br-ex" interface has been changed to "DOWN"
1164 * **logfile plugin**: Can be used to log collectd activity. Modify
1165 /opt/collectd/etc/collectd.conf to include:
1172 File "/var/log/collectd.log"
1178 Monitoring Interfaces and Openstack Support
1179 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1180 .. Figure:: monitoring_interfaces.png
1182 Monitoring Interfaces and Openstack Support
1184 The figure above shows the DPDK L2 forwarding application running on a compute
1185 node, sending and receiving traffic. collectd is also running on this compute
1186 node retrieving the stats periodically from DPDK through the dpdkstat plugin
1187 and publishing the retrieved stats to OpenStack through the
1188 collectd-ceilometer-plugin.
1190 To see this demo in action please checkout: `Barometer OPNFV Summit demo`_
1192 For more information on configuring and installing OpenStack plugins for
1193 collectd, check out the `collectd-ceilometer-plugin GSG`_.
1197 .. [1] https://collectd.org/wiki/index.php/Naming_schema
1198 .. [2] https://github.com/collectd/collectd/blob/master/src/daemon/plugin.h
1199 .. [3] https://collectd.org/wiki/index.php/Value_list_t
1200 .. [4] https://collectd.org/wiki/index.php/Data_set
1201 .. [5] https://collectd.org/documentation/manpages/types.db.5.shtml
1202 .. [6] https://collectd.org/wiki/index.php/Data_source
1203 .. [7] https://collectd.org/wiki/index.php/Meta_Data_Interface
1205 .. _Barometer OPNFV Summit demo: https://prezi.com/kjv6o8ixs6se/software-fastpath-service-quality-metrics-demo/
1206 .. _gnocchi plugin: https://github.com/openstack/collectd-ceilometer-plugin/tree/stable/ocata/
1207 .. _aodh plugin: https://github.com/openstack/collectd-ceilometer-plugin/tree/stable/ocata/
1208 .. _collectd-ceilometer-plugin GSG: https://github.com/openstack/collectd-ceilometer-plugin/blob/master/doc/source/GSG.rst
1209 .. _grafana guide: https://wiki.opnfv.org/display/fastpath/Installing+and+configuring+InfluxDB+and+Grafana+to+display+metrics+with+collectd