>>>>> On Thu, 29 Jul 2004 14:37:12 +0200, Thomas Anders <[EMAIL PROTECTED]> said:
>> Net-SNMP uses a combination of time and random number values to >> seed the engineID. Doing it any other way (like the common IP >> address way) is generally bad practice. Thomas> Are you willing to elaborate on the latter statement? The Thomas> SnmpEngineID TC from RFC3411 defines and allows for lots of Thomas> those "other ways", provided that they result in a value that Thomas> will be unique in the agent's administrative domain. AFAICS Thomas> this may well be the case with appropriate use of those "other Thomas> ways". The last statement is intended to say that basing an engineID on an IP address is a stupid thing to do, which is the standard way that most agents do things. IP addresses change. Thus when they do, either the agent reinitializes its user database and thus no previous users are valid (not common) or they keep the old engineID (A) defined by the old IP address which may conflict with another machine that generated a new engineID (B) based its current IP address which might happen to be the same as the old engineID (A == B), which is more common. In either case you've just hosed the management station by reassigning a box to a new address. MAC addresses are better, as they're typically unique. However, even those can typically be assigned by the user in common interfaces these days. -- Wes Hardaker Sparta ------------------------------------------------------- This SF.Net email is sponsored by OSTG. Have you noticed the changes on Linux.com, ITManagersJournal and NewsForge in the past few weeks? Now, one more big change to announce. We are now OSTG- Open Source Technology Group. Come see the changes on the new OSTG site. www.ostg.com _______________________________________________ Net-snmp-users mailing list [EMAIL PROTECTED] Please see the following page to unsubscribe or change other options: https://lists.sourceforge.net/lists/listinfo/net-snmp-users
