Hi, [speaking as co-chair]
I think these comments from Tom have not been addressed in mib-12. Does the WG agree that Figure 1 and discussion of facilities and hosts and so on should be removed in mib-xx? Does the WG agree that Diagram 1 from protocol-19 should be duplicated in mib-xx, as a way to bring the language used in the mib module closer to the "standard terminology" proposed by this WG? Does the WG agree that Diagram 1 represents all likely deployment scenarios (most notably those encountered on dumb-device, *NIX and Windows environments)? Would it be adequate to simply **reference** Diagram 1 in [RFCPROT], and the Terminology of Section 3 of [RFCPROT], and then explain how the terminology in the mib module relates to the [RFCPROT] terminology and diagram? David Harrington [EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED] co-chair, Syslog WG > -----Original Message----- > From: tom.petch [mailto:[EMAIL PROTECTED] > Sent: Friday, December 15, 2006 9:35 AM > To: David Harrington; 'Rainer Gerhards' > Cc: [EMAIL PROTECTED] > Subject: Re: [Syslog] Syslog-mib-11 > > mmmmmm > > Rainer's explanation shows me why I have not encountered > this; it is only > present in Windows servers and then only, if I read it > aright, because Windows > has not done a satisfactory implementation yet. The question > then is, how much > do we adapt to this behaviour of Windows? > > I find your remedy insufficient. Look at the diagram at the > start of -mib and > compare it with the diagram at the start of -protocol. > Seeing those two > diagrams convinced me that the two I-Ds came from different > planets and that, > until that was resolved - and only now, months later, do I > begin to see a > resolution - it was not worth expending much effort on -mib. > > I think this will be the reaction of others and that we MUST > include the diagram > in -mib that appears in all the others and then -mib must > explain how the terms > it chooses relates to those in the diagram and why. > Simplying saying this I-D > is different because we are on planet Microsoft is not > sufficient for me. > > I agree that there must also be a brief explanation in the > DESCRIPTION clause. > > Another problem I have with -mib is the statement that > "The syslog entities may be on the same host or on different hosts." > If on different hosts, what is the protocol used to > communicate between the two > hosts? If this is syslog, then how is the MIB information > communicated from > syslog sender to the entity with the SNMP agent? This is not a new > question:-(. > > > Tom Petch > > > ----- Original Message ----- > From: "David Harrington" <[EMAIL PROTECTED]> > To: "'Rainer Gerhards'" <[EMAIL PROTECTED]> > Cc: <[EMAIL PROTECTED]> > Sent: Thursday, December 14, 2006 11:23 PM > Subject: RE: [Syslog] Syslog-mib-11 > > > > Hi, > > > > [speaking as a contributor] > > > > Thank you Rainer for such a clear response. > > > > I recommend that text similar to Rainer's response be > included in the > > DESCRIPTION clause for the syslogEntityControlTable, to explain why > > multiple syslog entities are modeled in the MIB module. > > > > I recommend capturing the discussion within the MIB module > definition > > rather than in the document introductory sections, because > MIB modules > > are often distributed already-extracted from the > surrounding document, > > and NMS help screens are often fashioned from the > DESCRIPTION clauses. > > So putting this info in the table description clause will get the > > explanation to the users. I would not object to **also** having it > > discussed in the introductory text sections. > > > > David Harrington > > [EMAIL PROTECTED] > > [EMAIL PROTECTED] > > [EMAIL PROTECTED] > > > > https://www1.ietf.org/mailman/listinfo/syslog > > _______________________________________________ Syslog mailing list [email protected] https://www1.ietf.org/mailman/listinfo/syslog
