SystemUpTime in NodesStatistics

TL;DR: Hundredths of a second

sysUpTime OBJECT-TYPE
    SYNTAX      TimeTicks
    MAX-ACCESS  read-only
    STATUS      current
    DESCRIPTION
            "The time (in hundredths of a second) since the
            network management portion of the system was last
            re-initialized."
    ::= { system 3 }

note the key here is TimeTicks which defines the format of the object -- which sysUptime comments is in Hundredths of a second

the SYNTAX for MIB objects should be flly described in the SMI MIB:

(unless a vendor decides to extend the syntax, which is OK, but challenging for generic management applications to deal with)

-- hundredths of seconds since an epoch
TimeTicks ::=
    [APPLICATION 3]
        IMPLICIT INTEGER (0..4294967295)

note the TimeTicks says 'an epoch' rather than 'The Epoch' [which is grok as 1970-01-01 00:00:00.000]

Some systems don't allow the full range of TimeTicks values, with the results that some of our switches report a reboot unexpectedly which causes a mild panic until we realize they didn't actually reboot in the middle of the day.

Mr Letts, thank you very much for the write up.  This particular data is being transferred into ServiceDesk (our new helpdesk ticketing system).  They want the systemuptime to be presented in hours.  Is there a better way of determining a devices uptime via database query other than this value ?