Hardware or systems monitoring for servers/hosts (SNMP-based versus NON-SNMP)
Article ID: 191920
Updated On:
Products
Issue/Introduction
Hardware or system monitoring inquiry.
Environment
- Any UIM release
Cause
- Guidance on host/server monitoring and SNMP-based probes
Resolution
DX UIM SNMP-based probes include:
- snmpcollector (SNMP GETs, NIC, CPU cores, RAM, FAN, Power Supply, temperature, battery, array, reachability, availability, memory, etc.)
- snmptd (SNMP traps, whatever is included in the Linux OS/Standard/Specific available device MIBs)
- snmpget (simple probe for SNMP GETs, requires loading of MIBs/Dependent MIBs for the metrics you need to monitor)
Also, there are some storage probes that may still include the ability to monitor via SNMP if its part of the probe design and the SNMP agent is allowed to be enabled on the device.
SNMP-based monitoring can be used in all environments that require network device monitoring which include:
- routers
- switches
- load balancers
- switch-routers
- gateways
- bridges
- etc.
...essentially any device that can have an SNMP agent enabled on it so you can connect to and query the device.
- cdm (disks and memory utilization) may be used for hardware monitoring, but this is not common for server monitoring and it ONLY applies to server/host hardware monitoring but more indirectly, and it doesn't 'require' SNMP. cdm is a locally installed probe. Also, rsp can monitor systems remotely and can use telnet/SSH for connectivity when its enabled and allowed.
- ntperf (installed locally, uses perfmon counters on the host, and can be used to monitor Network Interfaces)
- SNMP-based hardware monitoring is most commonly implemented using the snmp-based probes, snmpcollector, snmptd, and snmpget in that order.
- Note on snmpget: Host/server monitoring with snmpget normally requires the relevant MIB(s) to be loaded but most of the common/standard MIBs are already loaded by default (out-of-the-box).
OID Description
HOST-RESOURCES-MIB::hrSystem Contains general system information such as uptime, number of users, and number of running processes.
HOST-RESOURCES-MIB::hrStorage Contains data on memory and file system usage.
HOST-RESOURCES-MIB::hrDevices Contains a listing of all processors, network devices, and file systems.
HOST-RESOURCES-MIB::hrSWRun Contains a listing of all running processes.
HOST-RESOURCES-MIB::hrSWRunPerf Contains memory and CPU statistics on the process table from HOST-RESOURCES-MIB::hrSWRun.
HOST-RESOURCES-MIB::hrSWInstalled Contains a listing of the RPM database.
IF-MIB
etc.
- cdm installs locally on a server/host (not a device) can help with an alarm if the memory is above or below the memory/disk/cpu usage, and can help to check for new disks being removed or added.
- dirscan will help to monitor changes to directories/files.
In terms of detecting/monitoring changes on a server/host system, please refer to the techdocs:
dirscan
cdm
Additional Information
Non-SNMP
- cdm (server monitoring, disks and memory utilization)
- processes/services
- dirscan (file/directory scan)
- net_connect (application service:port monitoring)
- logmon (run ANY Linux utility/command/script and parse the output, collect QOS data/generate alarms). For example, see:
16 commands to check hardware information on Linux
https://www.binarytides.com/linux-commands-hardware-info/
Feedback
