Monitoring ECDN servers via SNMP

Monitoring of ECDN servers via SNMP

You can monitor the health of ECDN servers in the ECDN Management Portal on the 'Servers' list page and 'Servers details' page. The health is indicated by a set of dot indicators. See the Status of ECDN servers article for details.

In addition, you may optionally self monitor the health and see resource utilization metrics of the ECDN servers running in your environment. You will need a SNMP (Simple Network Management Protocol) based monitoring tool.

ECDN supports SNMP protocol version 2c (v2c). On a per server basis you may enable publishing of utilization metrics via SNMP. By default SNMP is disabled.

Enabling SNMP on ECDN servers

To enable SNMP, in the ECDN Management Portal, on the servers details page, click on 'Change configuration'. Click 'Next' twice to navigate to the third 'Configure server' step. Under 'SNMP Polling' click the checkbox to enable it. You will then have the opportunity to specify a 'Community string' of your choice.

Click next and then the "Complete setup" button to save the configuration. This new configuration will be auto-propagated to the ECDN server instance, and the SNMP daemon process will be started. It will now allow a SNMP client to retrieve metrics.

You can monitor the status of the SNMP daemon on the ECDN server details page in the 'Services' widget in the row titled 'SNMP Polling'. A green dot indicates the daemon is running. A grey dot indicates SNMP has not been enabled.

When enabled on the same page in the left hand column under 'SNMP Polling' you can also see the community string assigned to this server.

SNMP OIDs available

The following SNMP OIDs (object identifiers) are enabled.

1. .1.3.6.1.2.1.1 - .iso.org.dod.internet.mgmt.mib-2.system
   These represent the SNMPv2-MIB definitions that is a collection of objects common to all managed systems. See the above link for details.
2. .1.3.6.1.2 .1.25 - .iso.org.dod.internet.mgmt.mib-2.host
   These represent a collection of objects that provide server metrics such as cpu, memory and disk utilization. See the above link for details.

The following SNMP OIDs are initialized to:

1. sysLocation = <your-ECDN-server-fqdn>
2. sysContact = https://support.ustream.tv/hc

Retrieving metrics using a SNMP client

To retrieve utilization metrics, you will need a SNMP client on your laptop or a client machine. In this article we will use the snmp command line utilities snmpget, snmpwalk, and snmptable to retrieve the metrics. These programs comes natively on the Macs. On Ubuntu, you may install it with sudo apt-get install snmp. On Windows, you may have to search and install one of GUI tools.

To retrieve the metrics you must specify the same community string.

snmpbulkwalk -v 2c -c public 10.100.11.212 .iso.org.dod.internet.mgmt.mib-2.host

snmpbulkwalk -v 2c -c public 10.100.11.212 .1.3.6.1.2.1.25

The output will be look similar to:

    HOST-RESOURCES-MIB::hrSystemUptime.0 = Timeticks: (315737241) 36 days,
    13:02:52.41 HOST-RESOURCES-MIB::hrSystemDate.0 = STRING:
    2018-6-8,1:58:40.0,+0:0 HOST-RESOURCES-MIB::hrSystemInitialLoadDevice.0 =
    INTEGER: 393216 HOST-RESOURCES-MIB::hrSystemInitialLoadParameters.0 =
    STRING: "BOOT_IMAGE=/vmlinuz-4.4.0-122-generic
    root=/dev/mapper/ecdn--edge--vg-root ro quiet"
    HOST-RESOURCES-MIB::hrSystemNumUsers.0 = Gauge32: 1
    HOST-RESOURCES-MIB::hrSystemProcesses.0 = Gauge32: 139
    HOST-RESOURCES-MIB::hrSystemMaxProcesses.0 = INTEGER: 0
    HOST-RESOURCES-MIB::hrMemorySize.0 = INTEGER: 4046140 KBytes
    HOST-RESOURCES-MIB::hrStorageIndex.1 = INTEGER: 1
    HOST-RESOURCES-MIB::hrStorageIndex.3 = INTEGER: 3
    ...
  

SNMP OIDs of interest

Here's a list of OIDs that you may find useful. You may additionally include other metrics as per your best practices:

• snmptable -v 2c -c public 10.200.11.115 .iso.org.dod.internet.mgmt.mib-2.host.hrStorage.hrStorageTable

  
  The highlighted rows are of interest:
  • Total physical memory assigned to this ECDN server VM.
  • / file system size and amount used.
  • /opt/cache file system size and amount used.
  • /opt/cache/vod-disk file system size and amount used to store the VOD video content.

• snmpget -v 2c -c public 10.200.11.115 .iso.org.dod.internet.mgmt.mib-2.host.hrSystem.hrSystemUptime.0

  HOST-RESOURCES-MIB::hrSystemUptime.0 = Timeticks: (316068432) 36 days, 13:58:04.32

  This is the server uptime since the last boot.

• snmpwalk -v 2c -c dhh 10.200.11.115 .iso.org.dod.internet.mgmt.mib-2.system.sysDescr.0

  SNMPv2-MIB::sysDescr.0 = STRING: Linux xxx.ecdn-demo.ecdn-ustream.com 4.4.0-122-generic #146-Ubuntu SMP Mon Apr 23 15:34:04 UTC 2018 x86_64

  This has the ECDN server hostname and linux version it is running.

• snmptable -v 2c -c public 10.200.11.115 .iso.org.dod.internet.mgmt.mib-2.host.hrDevice.hrProcessorTable

  
  The second column represents the cpu load.

• snmptable -v 2c -c dhh 10.200.11.115 .iso.org.dod.internet.mgmt.mib-2.ip.ipAddrTable

  
  
  The first highlighted entries represent the IP address of the ECDN server VM. The second highlighted entry represents the IP address assigned to this ECDN server VM if it successfully established a VPN tunnel connection with the ECDN backend. Note, ECDN server will only establish a VPN connection if you have explicitly enabled it in the ECDN server configuration in the ECDN Management Portal.