Monitoring Shibboleth Idp v3

Alerts and Statuses

There are two components to monitoring the Shib v3 production system, alerts and status indicators. Alerts are forwarded by the Pager Duty Application and status is managed by NEGSUE.

Pager Duty

Negsue Monitoring Page

Once notified that a situation is critical, examination of the NEGSUE page provides more insight into the issues. The URL ​https://negsue.alaska.edu opens to a list of clients monitored by NEGSUE.

(1) NEGSUE U Alaska Client List by Server Name

Shibboleth v3 Idp is served by george.alaska.edu. Clicking on george takes you to the status page for the server.

(2) Shibboleth Status List

Check Statuses

Checks convey status of a system. The status can be one of four colors: green, yellow, red and gray.

• Green: Everything is good
• Yellow: Warning things may continue to deteriorate
• Red: Critical non-functional state
• Gray: Something happened, but no information as to what it is

check-app-shib-idp-status : Disabled

This custom status is not yet ready for monitoring.

• Meaning:
• Contact:

check-os-sshd

• Meaning:
• Contact:

check-os-crond

• Meaning:
• Contact:

check-os-rsyslogd

• Meaning:
• Contact:

check-os-memory

• Meaning:
• Contact:

keepalive

• Meaning:
• Contact:

check-app-https-cert

• Meaning:
• Contact:

check-os-disk-usage

• Meaning:
• Contact:

check-os-load

• Meaning:
• Contact:

Metrics

Metrics are gathering data for grahic display. No alerts are sent from Metrics to Pager Duty. Below is an example of the CPU graph.

Metric status can be one of three colors: gren, yellow, and red.

• Green: Data is being gathered as expected
• Yellow: Issues are presenting themselves in the collection of data
• Red: No data is being collected

metric-os-disk

• Meaning:

metric-os-disk-usage

• Meaning:

metric-os-vmstat

• Meaning:

metric-os-load

• Meaning:

metric-os-interface

• Meaning:

metric-os-memory

• Meaning: