Details
-
Bug
-
Resolution: Fixed
-
Major
-
None
-
None
-
None
-
Version info:
Version Information
Observium 26.3.14711 (rolling) (27th March 2026)
OS Linux 6.12.74+deb13+1-amd64 [amd64] (Debian 13)
VM VMware Virtual Machine
Apache 2.4.66 (Debian)
PHP 8.4.16 (OPcache: Yes; FastCache: Yes, Apcu, 842kB) (Memory: 128MB)
Python 3.13.5
MariaDB 11.8.6-MariaDB-0+deb13u1 from Debian (client: MySQLnd 8.4.16)
SNMP NET-SNMP 5.9.4.pre2
RRDtool 1.7.2
Fping 5.1 (IPv4 and IPv6)
Fetch cURL 8.14.1 (OpenSSL/3.5.5, LibZ 1.3.1, LibIDN 2.3.8)Version info: Version Information Observium 26.3.14711 (rolling) (27th March 2026) OS Linux 6.12.74+deb13+1-amd64 [amd64] (Debian 13) VM VMware Virtual Machine Apache 2.4.66 (Debian) PHP 8.4.16 (OPcache: Yes; FastCache: Yes, Apcu, 842kB) (Memory: 128MB) Python 3.13.5 MariaDB 11.8.6-MariaDB-0+deb13u1 from Debian (client: MySQLnd 8.4.16) SNMP NET-SNMP 5.9.4.pre2 RRDtool 1.7.2 Fping 5.1 (IPv4 and IPv6) Fetch cURL 8.14.1 (OpenSSL/3.5.5, LibZ 1.3.1, LibIDN 2.3.8)
Description
Hi we currently have some 400G optics in a few Juniper devices, name the PTX10001 device.
Discovery has ran on the device for the JUNIPER-IFOPTICS-MIB.
It has discovered the module temperature which according to the mib is in .1 celcius. OID:
".1.3.6.1.4.1.2636.3.71.1.2.1.1.39"
https://mibs.observium.org/mib/JUNIPER-IFOPTICS-MIB/#jnxPMCurTemperature
As i understand Observium should be dividing this value by 100, otherwise the temperature given from the device get's intrepted as 556c rather than 55.6c.
I've attached a problem of the issue.
On the same device, the correct temperature also seems to be discovered at.
https://mibs.observium.org/mib/JUNIPER-DOM-MIB/#jnxDomCurrentModuleTemperature
So there does seem to be a duplication in data too.
However i don't think we can just disable the JUNIPER-IFOPTICS-MIB and use only the JUNIPER-DOM-MIB as things like "https://mibs.observium.org/mib/JUNIPER-DOM-MIB/#jnxDomCurrentTxLaserBiasCurrent" are being discovered and then reporting back 0A.
Where as back on the IFOPTICS mib, "https://mibs.observium.org/mib/JUNIPER-IFOPTICS-MIB/#jnxPMCurTxLaserBiasCurrent", is correctly reporting back the 0.22A.
Overall i'm not sure if this is a vendor issue with having data across two mibs, but the the very least the problem in Observium seems to be with the handling of the temperature and values in the IFOPTICS-MIB that probably needs to be divided if the values in that mib are a .X value, DB, dbm and Celcius.
