This article describes the behavior of jnxOverTemperature and jnxTemperatureOK SNMP traps.

When an SRX device is in a high temperature state, there are times when it does not send ​a jnxOverTemperature trap.

For more information on temperature threshold, see KB22153 - Junos temperature thresholds in SRX devices and the actions taken when it exceeds the threshold [juniper.net] .

The interval of temperature checks for SNMP trap is 1 minute. If the temperature exceeds the threshold and goes below within 1 minute, an SNMP trap will not be sent depending on the timing.
Note: Interval on syslog is 4 seconds.

Example

Below is a message log when the temperature of SRX240 RE exceeded the yellow alarm threshold.
SRX detected a high temperature at 16:07:26, but an SNMP trap was generated at 16:08:10 due to the 1 minute interval.
If the temperature goes down before 16:08:10, it would not have the high temperature information by the SNMP trap.

 Jul 21 16:07:26  SRX240H2POEA alarmd[1507]: Alarm set: RE color=YELLOW, class=CHASSIS, reason=Host 0 Temperature Warm
 Jul 21 16:07:26  SRX240H2POEA chassisd[1506]: CHASSISD_RE_WARM_TEMP_CONDITION: Routing Engine 0 temperature (51 C) is above warm temperature limit (50 C)
 Jul 21 16:08:10  SRX240H2POEA chassisd[1506]: CHASSISD_SNMP_TRAP6: SNMP trap generated: Over Temperature!  (jnxContentsContainerIndex 9, jnxContentsL1Index 1, jnxContentsL2Index 0, jnxContentsL3Index 0, jnxContentsDescr Routing Engine, jnxOperatingState/Temp 51)
 Jul 21 16:09:10  SRX240H2POEA chassisd[1506]: CHASSISD_SNMP_TRAP6: SNMP trap generated: Over Temperature!  (jnxContentsContainerIndex 9, jnxContentsL1Index 1, jnxContentsL2Index 0, jnxContentsL3Index 0, jnxContentsDescr Routing Engine, jnxOperatingState/Temp 51)​
 Jul 21 16:10:10  SRX240H2POEA chassisd[1506]: CHASSISD_SNMP_TRAP6: SNMP trap generated: Over Temperature! (jnxContentsContainerIndex 9, jnxContentsL1Index 1, jnxContentsL2Index 0,  jnxContentsL3Index 0, jnxContentsDescr Routing Engine, jnxOperatingState/Temp 51)​

This behavior is by design and the interval is not configurable. Check with the syslog message for real time status.

Or use event-option to generate SNMP trap for the following syslog messages to know whether or not SRX temperature exceeds temperature threshold:

Configuration Example:

 set event-options policy temp-alarm-set1 events SYSTEM
 set event-options policy temp-alarm-set1 attributes-match system.message matches "Alarm set.*Temperature"
 set event-options policy temp-alarm-set1 then raise-trap

 set event-options policy temp-alarm-clear events SYSTEM
 set event-options policy temp-alarm-clear attributes-match system.message matches "Alarm cleared.*Temperature"
 set event-options policy temp-alarm-clear then raise-trap