Juniper QSFP Temperature low warning set

One day, after connecting several 40G transceivers to the Juniper QFX5100 and MX204 devices, messages from these devices began to arrive on the syslog server:

fpc0 qsfp-0/0/0: Temperature low warning cleared
fpc0 qsfp-0/0/0: Temperature low alarm cleared
fpc0 qsfp-0/0/0: Temperature low warning set

At the same time, judging by the graph in Zabbix, the temperature of the optical module was stable at 24 degrees, and the threshold for triggering a warning/alarm was 0/-5 degrees, which was not possible, since the devices were located in a data center with a stable temperature of 20-21 degrees Celsius.

show interfaces diagnostics optics et-0/0/0 | match "Module temperature"
Module temperature                        :  24 degrees C / 74 degrees F
Module temperature high alarm             :  Off
Module temperature low alarm              :  Off
Module temperature high warning           :  Off
Module temperature low warning            :  Off
Module temperature high alarm threshold   :  78 degrees C / 172 degrees F
Module temperature low alarm threshold    :  -5 degrees C / 23 degrees F
Module temperature high warning threshold :  75 degrees C / 167 degrees F
Module temperature low warning threshold  :  0 degrees C / 32 degrees F

More logs from another transceiver, this one has more of them:

fpc0 qsfp-0/0/0 Chan# 1: Rx power low warning cleared "
fpc0 qsfp-0/0/0 Chan# 1: Rx power low alarm cleared "
fpc0 qsfp-0/0/0 Chan# 0: Rx power low warning cleared "
fpc0 qsfp-0/0/0 Chan# 0: Rx power low alarm cleared "
fpc0 qsfp-0/0/0 Chan# 3: Rx loss cleared "
fpc0 qsfp-0/0/0 Chan# 3: Rx power low warning cleared "
fpc0 qsfp-0/0/0 Chan# 3: Rx power low alarm cleared "
fpc0 qsfp-0/0/0 Chan# 2: Rx loss cleared "
fpc0 qsfp-0/0/0 Chan# 2: Rx power low warning cleared "
fpc0 qsfp-0/0/0 Chan# 2: Rx power low alarm cleared "
fpc0 qsfp-0/0/0: Temperature low warning cleared "
fpc0 qsfp-0/0/0: Temperature low alarm cleared "
fpc0 qsfp-0/0/0 Chan# 3: Rx power low warning set "
fpc0 qsfp-0/0/0 Chan# 3: Rx power low alarm set "
fpc0 qsfp-0/0/0 Chan# 2: Rx power low warning set "
fpc0 qsfp-0/0/0 Chan# 2: Rx power low alarm set "
fpc0 qsfp-0/0/0 Chan# 1: Rx power low warning set "
fpc0 qsfp-0/0/0 Chan# 1: Rx power low alarm set "
fpc0 qsfp-0/0/0 Chan# 0: Rx power low warning set "
fpc0 qsfp-0/0/0 Chan# 0: Rx power low alarm set "
fpc0 qsfp-0/0/0: Temperature low warning set "
fpc0 qsfp-0/0/0: Temperature low alarm set "
mib2d[2071]: SNMP_TRAP_LINK_DOWN: ifIndex 517, ifAdminStatus up(1), ifOperStatus down(2), ifName et-0/0/0"

Such events were observed in the logs a couple of times a day, there was even a case when the link disappeared and came back up, after which the transceiver was replaced with a transceiver from another manufacturer (ordered from OmniLink), after which such messages no longer appeared in the logs.

I also saw information that the manufacturer FINISAR CORP. had a similar problem with false positives, which fixed it in new firmware versions, in my case, on the problematic transceivers, OEM is indicated in the manufacturer field:

show chassis hardware |find FPC
FPC 0
  PIC 0
    Xcvr 0                NON-JNPR     QSFP+-40G-LR4

 show chassis pic fpc-slot 0 pic-slot 0
PIC port information:
                         Fiber                    Xcvr vendor       Wave-    Xcvr         JNPR
  Port Cable type        type  Xcvr vendor        part number       length   Firmware     Rev
  0    40GBASE LR4       SM    OEM                LQ-LR4-10         1310 nm  0.0          "

See also my article:
Juniper SNMP OIDs DDM for optical signal transceivers

Leave a comment

Leave a Reply