Re: Model: FE Attributes

There is a clear need to store stats about TML somewhere. 
I am not sure if the FEO is the right place. I dont think the idea of a
TML LFB is the right one either. 
Our implementation at the moment leaves the details of the TML to the
xEM plane - which is not standardized. I think thats where such details
belong.

cheers,
jamal

On Mon, 2006-16-10 at 14:09 +0800, Weiming Wang wrote:
> If we are religious about this, yes, I agree the attributes related to TML 
> are more relavent to FE Protocol LFB, but we may not be too religious about
> this. 
> 
> I don't think it is a good way to set a specific TML LFB for this purspose. 
> I'v already thought of this idea, but I was just drawn back by the odd appearance 
> of the TML LFB if we deifine it. Thinking that it have to have an odd relationship 
> with FE Protocol LFB, between them a receive and send channel have to exist, while  
> it is out of scope of FE topology.  
> 
> Note that events on congestion state  is triggered as an alert notification. I don't 
> think transmission of such one notification will add to the congestion state. It can 
> be sent with highest priority to make it not too late to report to CE. 
> 
> I object to report individual TML specific statistics to PL layer, which are too 
> implementation oriented and seem not very useful. I think we'v discussed this before 
> and had a rough consensus. 
> 
> thanks,
> Weiming
> 
> ----- Original Message ----- 
> From: "Joel M. Halpern" <[email protected]>
> 
> > The obvious question is whether that belongs in the FE Object LFB, 
> > the FE Protocol Object LFB, or an additional TML LFB?  I am inclined 
> > to think that at least some of it goes in one of the first two.
> > 
> > The other question is whether some of that information will actually work.
> > A ForCES Notification based on TML congestion is likely to increase 
> > congestion, and if it arrives at all it will arrive to late to be 
> > helpful.  We probably do need the FE side TML statistics, and there 
> > may be some parameters for some TMLs.  But those would presumably be 
> > in a TML LFB, since they may be specific to the TML.
> > 
> > Yours,
> > Joel
> > 
> > At 10:54 PM 10/14/2006, Wang,Weiming wrote:
> >>Hi Joel and Allen,
> >>
> >>The following text is about the xml description of the FE Object 
> >>attributes, capability, events that are related to TML management, 
> >>and are expected to incorparate in current FE Object definition. An 
> >>xml file is also included in the attachment. To better understand 
> >>the attributes, the latest tml service primitive draft revision 
> >>version is also attached, though it is not yet submitted due to some 
> >>last modification.
> >>
> >>thanks,
> >>Weiming
> >>----------
> >><?xml version="1.0" encoding="UTF-8"?>
> >><!-- FE Object Attributes, events, and capabilities that relates to TML -->
> >>
> >><dataTypeDefs>
> >>   <dataTypeDef>
> >>  <name>TMLErrorType</name>
> >>  <struct>
> >>    <element elementID="1">
> >>   <name>ErrorType</name>
> >>   <atomic>
> >>   <baseType>uint16</baseType>
> >>   <specialValues>
> >>    <specialValue value="1">
> >>     <name>AllLocalTMLlinkFailed</name>
> >>                 <synopsis>
> >>                 All local TML links failed.
> >>                 </synopsis>
> >>    </specialValue>
> >>    <specialValue value="2">
> >>     <name>PartLocalTMLlinkFailed</name>
> >>      <synopsis>
> >>      Part of local TML links failed.
> >>      </synopsis>
> >>    </specialValue>
> >>    <specialValue value="3">
> >>     <name>PeerTMLlinkUnavailable</name>
> >>     <synopsis>
> >>     Peer TML link unavailable
> >>     </synopsis>
> >>    </specialValue>
> >>    <specialValue value="4">
> >>     <name>PeerTMLlinkLeft</name>
> >>     <synopsis>
> >>     Peer TML link abnormally left
> >>     </synopsis>
> >>    </specialValue>
> >>   </specialValues>
> >>   </atomic>
> >>    </element>
> >>    <element elementID="2">
> >>   <name>ErrorStatus</name>
> >>   <atomic>
> >>   <baseType>uchar</baseType>
> >>   <specialValues>
> >>    <specialValue value="1">
> >>     <name>ErrorHappened</name>
> >>     </specialValue>
> >>    <specialValue value="0">
> >>     <name>ErrorReleased</name>
> >>     </specialValue>
> >>   </specialValues>
> >>   </atomic>
> >>    </element>
> >>  </struct>
> >>   </dataTypeDef>
> >>   <dataTypeDef>
> >>  <name>TMLCongestionAlertType</name>
> >>  <struct>
> >>    <element elementID="1">
> >>   <name>AlertType</name>
> >>   <atomic>
> >>     <baseType>uint16</baseType>
> >>     <specialValues>
> >>    <specialValue value="1">
> >>             <name>ControlCongestionAlert</name>
> >>                 <synopsis>
> >>                 Congestion alert from control channel
> >>                 </synopsis>
> >>    </specialValue>
> >>    <specialValue value="2">
> >>     <name>RedirectCongestionAlert</name>
> >>                 <synopsis>
> >>                 Congestion alert from redirect channel
> >>                 </synopsis>
> >>    </specialValue>
> >>    <specialValue value="3">
> >>     <name>DoSAlert</name>
> >>     <synopsis>
> >>     Alert of DoS attack from redirect data
> >>     </synopsis>
> >>    </specialValue>
> >>     </specialValues>
> >>   </atomic>
> >>    </element>
> >>    <element elementID="2">
> >>   <name>AlertStatus</name>
> >>   <atomic>
> >>   <baseType>uchar</baseType>
> >>   <specialValues>
> >>    <specialValue value="1">
> >>     <name>AlertHappened</name>
> >>     </specialValue>
> >>    <specialValue value="0">
> >>     <name>AlertReleased</name>
> >>     </specialValue>
> >>   </specialValues>
> >>   </atomic>
> >>    </element>
> >>   </struct>
> >>   </dataTypeDef>
> >>   <dataTypeDef>
> >>  <name>TMLType</name>
> >>  <synopsis>
> >>   Transportation Mapping Layer(TML) media type
> >>  </synopsis>
> >>  <atomic>
> >>   <baseType>uint16</baseType>
> >>   <specialValues>
> >>     <specialValue/>
> >>   </specialValues>
> >>  </atomic>
> >>   </dataTypeDef>
> >></dataTypeDefs>
> >>
> >><attributes>
> >>   <attribute access="read-only" elementID="10">
> >>  <name>TMLError</name>
> >>  <synopsis>
> >>  The attribute records latest error report from TML
> >>  </synopsis>
> >>  <typeRef>TMLErrorType</typeRef>
> >>   </attribute>
> >>   <attribute access="read-only" elementID="11">
> >>  <name>TMLCongestionAlert</name>
> >>  <synopsis>
> >>  The attribute records latest congestion alert from TML
> >>  </synopsis>
> >>  <typeRef>TMLCongestionAlertType</typeRef>
> >>   </attribute>
> >>   <attribute access="read-write" elementID="12">
> >>  <name>WorkingTMLType</name>
> >>  <synopsis>
> >>
> >>  </synopsis>
> >>  <typeRef>TMLType</typeRef>
> >>  <optional/>
> >>   </attribute>
> >></attributes>
> >>
> >><capabilities>
> >>   <capability elementID="15">
> >>  <name>SupportedTMLType</name>
> >>  <synopsis>
> >>  Supported TML media types for this TML
> >>  </synopsis>
> >>  <array type="variable-size">
> >>   <typeRef>TMLType</typeRef>
> >>  </array>
> >>  <optional/>
> >>   </capability>
> >></capabilities>
> >>
> >><events baseID="10">
> >>   <event eventID="1">
> >>   <name>TMLErrorReport</name>
> >>   <synopsis>
> >>   The FE TML error report to CE
> >>   </synopsis>
> >>   <eventTarget>
> >>   <eventField>TMLError</eventField>
> >>   </eventTarget>
> >>   <eventChanged/>
> >>   <eventReports>
> >>  <eventReport>
> >>   <eventField>TMLError</eventField>
> >>  </eventReport>
> >>   </eventReports>
> >>   </event>
> >>   <event eventID="2">
> >>   <name>TMLCongestionAlert</name>
> >>   <synopsis>
> >>   Alert of TML Congestion in the FE
> >>   </synopsis>
> >>   <eventTarget>
> >>   <eventField>TMLCongestionAlert</eventField>
> >>   </eventTarget>
> >>   <eventChanged/>
> >>   <eventReports>
> >>  <eventReport>
> >>   <eventField>TMLCongestionAlert</eventField>
> >>  </eventReport>
> >>   </eventReports>
> >>   </event>
> >></events>
> >>------------------------
> >>
> >>----- Original Message -----
> >>From: "Wang,Weiming" <[email protected]>
> >>  > Joel,
> >> >
> >> > The submission deadline is October 23, Monday, I 
> >> suppose?  Because the model draft is inactive right now, so, if it 
> >> can not comply with this deadline and must be with Oct 16 as an 
> >> initial draft deadline, I suppose to go with the first way,  or, we 
> >> may directly go with one submission. I can promise to finish my input by 15th.
> >> >
> >> > thanks,
> >> > weiming
> >> >
> >> >
> >> > ----- Original Message -----
> >> > From: "Joel M. Halpern" <[email protected]>
> >> >
> >> > > Any input will be appreciated.
> >> > > I was hoping to get this in to the I-D editor before the flood for
> >> > > the deadline, but that may not happen.
> >> > >
> >> > > There are a couple of ways forward.
> >> > > If we get the draft revised without the FE Object information, we can
> >> > > submit that, and then submit a revision with the information as soon
> >> > > as you make it available to us.
> >> > > If there are a number of other items where we find that folks are
> >> > > promising updates, we can hold off submitting the draft until 
> >> we get those.
> >> > >
> >> > > Yours,
> >> > > Joel  M. Halpern
> >> > >
> >> > > At 11:10 AM 9/28/2006, Wang,Weiming wrote:
> >> > > >Joel,
> >> > > >
> >> > > >I'm very sorry that I defered inputing something for the FE object
> >> > > >as I propsed months ago, owing to the ugly busy business for the
> >> > > >university which may have to be ttill Oct. 1st.
> >> > > >I have to  make the TML SP update first ( which I'm doing) before
> >> > > >making things for the FE object, for things I'm to input  is tied to
> >> > > >the SP in some way, as we discussed .  Could give me another one or
> >> > > >two weeks for this?
> >> > > >
> >> > > >While at the same time, I also urge other guys to have more inputs
> >> > > >for the FE object attributes.
> >> > > >
> >> > > >thanks
> >> > > >weiming
> >> > > >
> >> > > >----- Original Message -----
> >> > > >From: "Joel M. Halpern" <[email protected]>
> >> > > >
> >> > > > > I am resending this note to the Forces list.
> >> > > > > Folks, the FE Object is insufficient.
> >> > > > > I am quite sure that it needs more information.
> >> > > > > But without input I can not fix it.
> >> > > > > I can pretend it is correct.
> >> > > > > Likely, no reviewer will stop us publishing as is.
> >> > > > >
> >> > > > > Please!
> >> > > > >
> >> > > > > Section 5.3 notes that the FE Attributes in the FE Object is woefully
> >> > > > > incomplete.
> >> > > > > I believe this to be true.
> >> > > > > However, without input I can not repair this.
> >> > > > >
> >> > > > > In the absence of text I will remove the note and we will simply be
> >> > > > > insufficient.
> >> > > > >
> >> > > > > Yours,
> >> > > > > Joel M. Halpern
> >> > >
> >>