Unit definition file (jpc_program-name.service)

Organization of this page

Format
Description
File
Storage directory
Character code
Line feed code
When the definitions are applied
Information that is specified

Format

[Unit]
Description = Unit description
After=local-fs.target remote-fs.target rsyslog.service network.target
[Service]
WorkingDirectory = Working directory
ExecStart = /bin/sh -c 'Program Command Line |& (trap "" 15 && exec install-directory/jp1ima/bin/rotatelogs -n 8^#1 log output directory^#2 10240K)'
Type = simple
Restart=no
User=root
TimeoutStopSec=Termination timeout period
LimitNOFILE=65536^#3
Environment=SIGDUMP_PATH=-^#3
[Install]
WantedBy=multi-user.target graphical.target

#1: Specify "100" only for unit definition file of Fluentd.
#2: The log-destination directory that is setup as the default value varies depending on the program. For the path of the default value, see List of files/directories that can be viewed/edited by the user in JP1/IM - Agent of integrated agent host (Linux) in Appendix A.4(4) Integrated agent host (Linux) in the JP1/Integrated Management 3 - Manager Overview and System Design Guide.
#3: Described only for unit definition file of Fluentd.

To Page Top

Description

This is a definition file for registering programs in systemctl in the Linux environment.

To Page Top

File

For a physical host

jpc_program-name.service

jpc_program-name.service.model (model file)
For a logical host

jpc_program-name_logical-host-name.service

To Page Top

Storage directory

■Integrated agent host

For a physical host (Model File storage destination)

/opt/jp1ima/conf/
For Physical host and Logical host (storage destination of definition file)

/usr/lib/systemd/system/

The target of the installation is as follows:

Target name	Description
multi-user.target	runlevel=3 Multi-user mode (text login)
graphical.target	runlevel=5 Multi-user mode (graphical login)

Target name

Description

multi-user.target

runlevel=3

Multi-user mode (text login)

graphical.target

runlevel=5

Multi-user mode (graphical login)

To Page Top

Character code

UTF-8 (without BOM)

To Page Top

Line feed code

To Page Top

When the definitions are applied

It is reflected when you run the systemctl daemon-reload command.

To Page Top

Information that is specified

The startup order of the services on which the Prometheus server, Alertmanager, and Exporter depend is shown below.

For Prometheus server

Startup order of dependent services	Description
local-fs.target	Local File System
remote-fs.target	Remote File System
rsyslog.service	rsyslog

For Exporter, and Alertmanager, Fluentd, imagent, imagentproxy, imagentaction

Startup order of dependent services	Description
local-fs.target	Local File System
remote-fs.target	Remote File System
rsyslog.service	rsyslog
network.target	Network

The ending timeout period is shown below.

Types of programs	End timeout period (in seconds)
Prometheus server	110
Alertmanager	60
Node exporter
Blackbox exporter
Yet another cloudwatch exporter
process_exporter
promitor_scraper
promitor_resource_discovery
script_exporter
Fluentd
OracleDB exporter
imagent	140
imagentproxy	140
imagentaction	60

The following are the areas that you can change:

If you want to change command line options

Edit ExecStart line.

■For Prometheus server

For the parameters of prometheus command, see the description of prometheus command options in Service definition file (jpc_program-name_service.xml).

■For Alertmanager

For the parameters of alertmanager command, see the description of prometheus command options in Service definition file (jpc_program-name_service.xml).

■For Node exporter for AIX

For the arguments of the Node exporter command, see the description of "node_exporter command options" below.

■For Node exporter

About arguments of the node_exporter command, see node_exporter command options.

■For Yet another cloudwatch exporter

About arguments of the yet-another-cloudwatch-exporter command, see yet-another-cloudwatch-exporter command options.

■For Process exporter

For the parameters of process exporter command, see the description of process_exporter command options in Process exporter configuration file (jpc_process_exporter.yml).

■For Primitor

No specifiable options for the promitor_scraper command and promitor_resource_discovery command.

■For Script exporter

For the parameters of script exporter command, see the description of script exporter command options in Service definition file (jpc_program-name_service.xml).

■For OracleDB exporter

For the arguments of oracledb_exporter command, see the description of oracledb_exporter command options in Service definition file (jpc_program-name_service.xml).
To change the number of log sectors

Edit the value after the -n option on the ExecStart line. The number of log sectors that can be specified is 8 to 512.
To change log size

Edit the value at the end of the ExecStart line.
If you want to change the log output destination

Edit the log output directory specified by the rotatelogs argument.
If you want the process to be restarted automatically when it stops abnormally, specify "always" for Restart. If the process is not restarted automatically, such as in a cluster configuration, specify "no" for Restart.

The options for the node_exporter command and the yet-another-cloudwatch-exporter command are shown below.

node_exporter command options

Items related to collectors can be enabled with the "--collector.collector name" flag. To invalidate entries for collectors, specify the "--no-collector.collector name" flag.

Item	Description	Changeability	What You Setup in Your JP1/IM - Agent	JP1/IM - Agent Defaults Value
--collector.cpu.info	Enable the collection of node_cpu_info metrics. Specify the ,--collector.cpu.info if you want to enable it. Specify the ,--no-collector.cpu.info if you want to disable it.	Y	--	--collector.cpu.info
--collector.filesystem.ignored-mount-points	Specifies the regular expression for mount points to ignore in the FileSystem collector.	Y	Regular expressions for mount points to ignore in the FileSystem collector	--collector.filesystem.ignored-mount-points="^/(dev\|proc\|sys\|var/lib/docker/.+)($\|/)"
--collector.filesystem.ignored-fs-types	Specifies a regular expression for file system types to be ignored by the FileSystem collector.	Y	Regular expressions for mount points to ignore in the FileSystem collector	--collector.filesystem.ignored-fs-types="^/(dev\|proc\|sys\|var/lib/docker/.+)($\|/)"
--collector.netstat.fields	Specifies the regular expression for the field to return to the netstat collector.	Y	Regular expression for fields to return to the netstat collector	--collector.netstat.fields="^(._(InErrors\|InErrs)\|Ip_Forwarding\|Ip(6\|Ext)_(InOctets\|OutOctets)\|Icmp6?_(InMsgs\|OutMsgs)\|TcpExt_(Listen.\|Syncookies.*\|TCPSynRetrans)\|Tcp_(ActiveOpens\|InSegs\|OutSegs\|OutRsts\|PassiveOpens\|RetransSegs\|CurrEstab)\|Udp6?_(InDatagrams\|OutDatagrams\|NoPorts\|RcvbufErrors\|SndbufErrors))$"
--path.procfs	Specifies the procfs mount point.	Y	The procfs mount point	--path.procfs="/proc"
--path.sysfs	Specifies the sysfs mount point.	Y	The sysfs mount point	--path.sysfs="/sys"
--path.rootfs	Specifies the rootfs mount point.	Y	The rootfs mount point	--path.rootfs="/"
--collector.systemd.unit-include	Specify the unit filename of Systemd to be monitored by the service monitoring function in regular expressions. The default is an empty character and no service is monitored. If the regular expressions are incorrect, Node exporter fails to start servicing. Depending on how the regular expression is specified, it may take time to collect performance information. For more information, see Appendix G.4 Tips on using regular expressions in the JP1/Integrated Management 3 - Manager Overview and System Design Guide.	Y	Regular expression that matches the unit file name	--collector.systemd.unit-include=""
--collector.cpu	Enable the cpu collector. If enabled,-- specify collector.cpu. If you want to disable it,-- specify no-collector.cpu.	Y	--	--collector.cpu
--collector.diskstats	Enable the diskstats collector. If enabled,-- specify collector.diskstats. To disable it,-- specify no-collector.diskstats.	Y	--	--collector.diskstats
--collector.filesystem	Enable the FileSystem collector. If enabled,-- specify collector.filesystem. If disabled,-- specify no-collector.filesystem.	Y	--	--collector.filesystem
--collector.loadavg	Enable the loadavg collector. If enabled,-- specify collector.loadavg. If you want to disable it,-- specify no-collector.loadavg.	Y	--	--collector.loadavg
--collector.meminfo	Enable the meminfo collector. If enabled,-- specify collector.meminfo. To disable it,-- specify no-collector.meminfo.	Y	--	--collector.meminfo
--collector.netclass	Enable the netclass collector. If enabled,-- specify collector.netclass. If disabled,-- specify no-collector.netclass.	Y	--	--collector.netclass
--collector.netdev	Enable the netdev collector. If enabled,-- specify collector.netdev. If disabled,-- specify no-collector.netdev.	Y	--	--collector.netdev
--collector.netstat	Enable the netstat collector. If enabled,-- specify collector.netstat. To disable it,-- specify no-collector.netstat.	Y	--	--collector.netstat
--collector.nfs	Enable the nfs collector. If enabled,-- specify collector.nfs. To disable it,-- specify no-collector.nfs.	Y	--	--collector.nfs
--collector.nfsd	Enable the nfsd collector. If enabled,-- specify collector.nfsd. To disable it,-- specify no-collector.nfsd.	Y	--	--collector.nfsd
--collector.stat	Enable the stat collector. If enabled,-- specify collector.stat. To disable it,-- specify no-collector.stat.	Y	--	--collector.stat
--collector.systemd	Enable systemd collector. When enabled, specifies --collector.systemd. When disabled, specifies --nocollector.systemd.	Y	--	--collector.systemd
--collector.time	Enable the time collector. If enabled,-- specify collector.time. To disable it,-- specify no-collector.time.	Y	--	--collector.time
--collector.uname	Enable the uname collector. If enabled,-- specify collector.uname. To disable it,-- specify no-collector.uname.	Y	--	--collector.uname
--collector.vmstat	Enable the vmstat collector. If enabled,-- specify collector.vmstat. To disable it,-- specify no-collector.vmstat.	Y	--	--collector.vmstat
--web.listen-address	Specify the listening port. If you want to limit the listening IP address, also include the host name or internet address. If you specify a host name, you can specify up to 255 bytes. If a host name is specified for this option, the targets of the Node exporter's discovery configuration file (jpc_file_sd_config_node.yml) on the same host must be set to the same host name as this option, and the Prometheus server must be restarted or reloaded. If an internet address is specified for this option, the targets of the Node exporter's discovery configuration file (jpc_file_sd_config_node.yml) on the same host must be set to a host name that resolves to the same IP address as this option, and the Prometheus server must be restarted or reloaded. <Configuration Example> To accept requests for all IP addresses, set the following: --web.listen-address=":20716" If you want to limit the IP addresses that are accepted, you can specify a host name or an internet address as follows: --web.listen-address="`hostname or IP address`: 20716"	Y	Specify this when you want to change the port or limit the listening IP address.	--web.listen-address=":20716"
--web.max-requests	Specifies the maximum number of scrape requests to accept at the same time.	N	--	--web.max-requests=40
--log.level=info	Only messages above the specified level are logged. You can specify one of the following levels: debug, info, warn, or error.	N	Change Setup when you are prompted to change Value from the support desk. Normally, no change is required.	--log.level=debug

Legend:: Y: Changeable, N: Not changeable, --: Not applicable

yet-another-cloudwatch-exporter command options

Item	Description	Changeability	What You Setup in Your JP1/IM - Agent	JP1/IM - Agent Defaults Value
-cloudwatch-concurrency	Specify the maximum number of concurrent requests to the Amazon CloudWatch API.	Y	Maximum concurrent requests	-cloudwatch-concurrency=5
-config.file	Specify the path to the Yet another cloudwatch exporter configuration file.	Y	--	-config.file="`Agent-path`/jp1ima/conf/jpc_ya_cloudwatch_exporter.yml"
-fips	Specify this in when using API of FIPS compliant Amazon Web Services (AWS)^#. #: This should be specified, for example, if your security requirements require compliance with Federal Information Processing Standard (FIPS) 140-2).	Y	Specify if you want to enable it.	Do not specify (do not enable)
-listen-address	Specify the listening port. If you want to limit the listening IP address, also include the host name or internet address. If you specify a host name, you can specify up to 255 bytes. If you specify Host name for this option, you must Setup the same Host name as this option to targets of Yet another cloudwatch exporter discovery configuration file (jpc_file_sd_config_cloudwatch.yml) Yet another cloudwatch exporter discovery configuration file on the same host, and Restart or Reload Prometheus server. If you specify IP address for this option, you must Yet another cloudwatch exporter discovery configuration file (jpc_file_sd_config_cloudwatch.yml) targets on the same host, Setup Host name that resolves to the same IP address as this option and Restart or Reload Prometheus server. <Configuration Example> To accept requests for all IP addresses, set the following: -listen-address=":20718" If you want to limit the IP addresses that are accepted, you can specify a host name or an internet address as follows: -listen-address="`hostname or IP address`:20718"	Y	Specify this when you want to change the port or limit the listening IP address.	-listen-address=":20718"
-metrics-per-query	Specifies the number of metrics created in a single GetMetricsData request.	Y	Normally, you don't change it.	-metrics-per-query=500
-tag-concurrency	Specifies the maximum number of concurrent requests to the Resource Tagging API.	Y	Normally, you don't change it.	-tag-concurrency=5

Legend:: Y: Changeable, --: Not applicable

To Page Top