RFC 3877, "Alarm Management Information Base (MIB)", September 2004Source of RFC: disman (ops)
Errata ID: 1652
Reported By: Brian Bidulock
Date Reported: 2009-01-13
Rejected by: Dan Romascanu
Date Rejected: 2010-05-11
Section 5.4 says:
alarmModelState -> ituAlarmPerceivedSeverity 1 -> clear (1) 2 -> indeterminate (2) 3 -> warning (6) 4 -> minor (5) 5 -> major (4) 6 -> critical (3)
It should say:
alarmModelState -> ituAlarmPerceivedSeverity 1 -> clear (1) 2 -> warning (6) 3 -> indeterminate (2) 4 -> minor (5) 5 -> major (4) 6 -> critical (3)
alarmModelState requires that the states be defined from less severe to more severe; however, under ITU-T PerceivedSeverity from ITU-T Rec. X.721 | ISO/IEC 10165-2 "indeterminate" is more severe than "warning". This change corrects the order to match the requirement for order of severity for alarmModelState.
While the discrepancy between the documents is unfortunate, there is not a technical requirement for the enumeration values to be identical, nor is there a technical requirement for the labels to be identical, even though there is obviously considerable documentation value in avoiding gratuitous differences.
What *is* technically important is that the MIB be able to uniquely represent all the cases from M.3100, and it accomplishes that goal.
In a future version of the document we can add an informative note alerting implementors to the discrepancies in numbering and spelling, so their implementations can include appropriate mapping functions to avoid losing information.