See https://lists.gt.net/nsp/juniper/66756
The two main options are syslog and SNMP traps.
I added the following alert rule in LibreNMS, to email only me for now.
Description
Description
Related Objects
Related Objects
- Mentioned In
- T333210: Investigate Junos Prometheus exporter
Event Timeline
Comment Actions
Mentioned in SAL (#wikimedia-operations) [2019-12-19T06:51:59Z] <XioNoX> cr2-eqdfw:delete chassis alarm management-ethernet - T241105
Comment Actions
As a test, I pushed the following:
[edit chassis]
- alarm {
- management-ethernet {
- link-down ignore;
- }
- }As cr2-eqdfw doesn't have a mgmt link, it did detect the issue:
ayounsi@cr2-eqdfw# run show system alarms 1 alarms currently active Alarm time Class Description 2019-12-19 06:52:10 UTC Major Management Ethernet Links Down
But... nothing in syslog...
Comment Actions
I reached out to our Juniper account rep, after a few emails they opened ER-080949 (Enhancement Request).
Since Junos 15, it's possible to generate custom OIDs using python scripts: https://www.juniper.net/documentation/en_US/junos/topics/example/junos-script-automation-snmp-script-example.html. Might be worth investigating.