Re: [netmod] Question on how to design a Yang model to reflect auto-asignment of a give leaf
Agree that ideally we avoid (a) in running. Unfortunately I think we used (a)
in ietf-interfaces though:
leaf type {
type identityref {
base interface-type;
}
mandatory true;
description
"The type of the interface.
When an interface entry is created, a server MAY
initialize the type leaf with a valid value, e.g., if it
is possible to derive the type from the name of the
interface.
I do like (c) but note that it means the value calculated by the server may not
be persistent (e.g. across reboot). It depends how the server calculates the
value. If it is "allocated" from some sort of free list, then at reboot it may
select a different value than the previously selected value.
Jason
> -Original Message-
> From: netmod On Behalf Of Martin Bjorklund
> Sent: Wednesday, February 12, 2020 2:48 AM
> To: Oscar González de Dios
> Cc: ops...@ietf.org; netmod@ietf.org
> Subject: Re: [netmod] Question on how to design a Yang model to reflect auto-
> asignment of a give leaf
>
> Oscar González de Dios wrote:
> >
> >
> > -Mensaje original-
> > De: Martin Bjorklund
> > Enviado el: martes, 11 de febrero de 2020 11:00
> > Para: Oscar González de Dios
> > CC: ops...@ietf.org; netmod@ietf.org
> > Asunto: Re: [netmod] Question on how to design a Yang model to reflect
> > auto-asignment of a give leaf
> >
> > Hi,
> >
> > Oscar González de Dios wrote:
> > > Dear OPSAWG and Netmod colleagues,
> > >
> > > During last IETF Opsawg meeting we raised a question (and
> > > there was some discussion during the meeting) that we have
> > > found yet no good answer and we would like to discuss it
> > > with
> > > operations and Yang experts.
> > >
> > > The use case is the following: We have a yang module which
> > > holds certain optional leafs. The behaviors that we would
> > > like
> > > to have (and distinguish between them) are:
> > >
> > >
> > > a) The user does not provide the value and such value is auto-assigned
> > > by the system (a device (if it is a device module) or a controller (if
> > > it is a network/service module)).
> >
> > I assume that this value not a static default value?
> >
> > [Oscar] True. Should the leaf have a default value, it implies that
> > "if the value is not set, the default value is taken".
> >
> > > b) The user does not provide a value and wants that such value IS NOT
> > > set by the system (as assigning a value has implications). That is,
> > > intentionally it is aimed at being left "empty" and should not be
> > > expanded. So, either the value is set or should remain empty
> >
> > Do you mean that you want (a) and (b) at the same time for the same
> > leaf?
> >
> > [Oscar] No. Depending on the leaf, we would like to specify behavior a
> > or behavior b. Behavior a is ok for most of the cases. The problem is
> > that in some cases, assigning a value has way more implications and
> > the service will not work properly. Those case are the ones we wanted
> > to specifically tackle.
>
> Ok. See below.
>
> > > What is the best way to model this behavior? I see that some yang
> > > modules have added an "auto-assignment" leaf to express if
> > > auto-assignment is desired or not. (hence, auto-assignment false, and
> > > leaf not set, would do not assign).
> > >
> > > Which is the "default" rule for a leaf that is not set? It is that the
> > > system is free to create it (via template or any means of
> > > auto-assignment) or should leave it as is, that is, empty?
> > >
> > > In NMDA, the system is allowed to expand a given configuration. This
> > > fact, in my personal view, implies that by "default" any system could
> > > implement the "auto-assignment" behavior being compliant with
> > > Neconf/Restconf/NMDA rules (but I am not sure if the interpretation is
> > > correct).
> >
> > There are (at least) three ways to interpret "auto-assign". The
> > client writes to running, and then the server auto-assigns X:
> >
> > (a) in running
> > (b) in intended
> > (c) in the operational state
> >
> > (c) is uncontroversial and simple to implement in all servers, and
> > simple to understand.
> > [Oscar] agree
> >
> > (b) is allowed by NMDA but requires more of the server implementation;
> > specifically it requires the server to support that intended is
> > different from running.
> > [Oscar] Agree . "Theoretically speaking" this is the behavior I would
> > consider strictly follows NMDA guidelines. Reality is implementations
> > are yet far from this...
> >
> > (a) is not recommended in general; running should be fully owned by
> > the client(s) and not modified by the server.
> > [Oscar] Agree.
> >
> > [Oscar] So... what would be the best way to specify the behavior?
> > Explicitly adding an
Re: [netmod] Question on how to design a Yang model to reflect auto-asignment of a give leaf
Oscar González de Dios wrote: > > > -Mensaje original- > De: Martin Bjorklund > Enviado el: martes, 11 de febrero de 2020 11:00 > Para: Oscar González de Dios > CC: ops...@ietf.org; netmod@ietf.org > Asunto: Re: [netmod] Question on how to design a Yang model to reflect > auto-asignment of a give leaf > > Hi, > > Oscar González de Dios wrote: > > Dear OPSAWG and Netmod colleagues, > > > > During last IETF Opsawg meeting we raised a question (and > > there was some discussion during the meeting) that we have > > found yet no good answer and we would like to discuss it > > with > > operations and Yang experts. > > > > The use case is the following: We have a yang module which > > holds certain optional leafs. The behaviors that we would > > like > > to have (and distinguish between them) are: > > > > > > a) The user does not provide the value and such value is auto-assigned > > by the system (a device (if it is a device module) or a controller (if > > it is a network/service module)). > > I assume that this value not a static default value? > > [Oscar] True. Should the leaf have a default value, it implies that > "if the value is not set, the default value is taken". > > > b) The user does not provide a value and wants that such value IS NOT > > set by the system (as assigning a value has implications). That is, > > intentionally it is aimed at being left "empty" and should not be > > expanded. So, either the value is set or should remain empty > > Do you mean that you want (a) and (b) at the same time for the same > leaf? > > [Oscar] No. Depending on the leaf, we would like to specify behavior a > or behavior b. Behavior a is ok for most of the cases. The problem is > that in some cases, assigning a value has way more implications and > the service will not work properly. Those case are the ones we wanted > to specifically tackle. Ok. See below. > > What is the best way to model this behavior? I see that some yang > > modules have added an "auto-assignment" leaf to express if > > auto-assignment is desired or not. (hence, auto-assignment false, and > > leaf not set, would do not assign). > > > > Which is the "default" rule for a leaf that is not set? It is that the > > system is free to create it (via template or any means of > > auto-assignment) or should leave it as is, that is, empty? > > > > In NMDA, the system is allowed to expand a given configuration. This > > fact, in my personal view, implies that by "default" any system could > > implement the "auto-assignment" behavior being compliant with > > Neconf/Restconf/NMDA rules (but I am not sure if the interpretation is > > correct). > > There are (at least) three ways to interpret "auto-assign". The > client writes to running, and then the server auto-assigns X: > > (a) in running > (b) in intended > (c) in the operational state > > (c) is uncontroversial and simple to implement in all servers, and > simple to understand. > [Oscar] agree > > (b) is allowed by NMDA but requires more of the server implementation; > specifically it requires the server to support that intended is > different from running. > [Oscar] Agree . "Theoretically speaking" this is the behavior I would > consider strictly follows NMDA guidelines. Reality is implementations > are yet far from this... > > (a) is not recommended in general; running should be fully owned by > the client(s) and not modified by the server. > [Oscar] Agree. > > [Oscar] So... what would be the best way to specify the behavior? > Explicitly adding an auto-assign leaf to identify the behavior? Just > "obey" NMDA rules? For your "auto-assignment" case, I would describe the behaviour in the description statement. Something like: "If this leaf has not been configured, the server will calculate a value [... specify how ... ] and use that value operationally. This calculated value is available in the operational state." For your other case (the user really doesn't want a value) I also would document this in the description (unless it's obvious). /martin > > > /martin > > > > > > > Best Regards, > > > > Óscar > > > > > > > > Este mensaje y sus adjuntos se dirigen exclusivamente a su > > destinatario, puede contener información privilegiada o confidencial y > > es para uso exclusivo de la persona o entidad de destino. Si no es > > usted. el destinatario indicado, queda notificado de que la lectura, > > utilización, divulgación y/o copia sin autorización puede estar > > prohibida en virtud de la legislación vigente. Si ha recibido este > > mensaje por error, le rogamos que nos lo comunique inmediatamente por > > esta misma vía y proceda a su destrucción. > > > > The information contained in this transmission is privileged and > > confidential information intended only
Re: [netmod] Question on how to design a Yang model to reflect auto-asignment of a give leaf
-Mensaje original- De: Martin Bjorklund Enviado el: martes, 11 de febrero de 2020 11:00 Para: Oscar González de Dios CC: ops...@ietf.org; netmod@ietf.org Asunto: Re: [netmod] Question on how to design a Yang model to reflect auto-asignment of a give leaf Hi, Oscar González de Dios wrote: > Dear OPSAWG and Netmod colleagues, > > During last IETF Opsawg meeting we raised a question (and > there was some discussion during the meeting) that we have > found yet no good answer and we would like to discuss it with > operations and Yang experts. > > The use case is the following: We have a yang module which > holds certain optional leafs. The behaviors that we would like > to have (and distinguish between them) are: > > > a) The user does not provide the value and such value is auto-assigned > by the system (a device (if it is a device module) or a controller (if > it is a network/service module)). I assume that this value not a static default value? [Oscar] True. Should the leaf have a default value, it implies that "if the value is not set, the default value is taken". > b) The user does not provide a value and wants that such value IS NOT > set by the system (as assigning a value has implications). That is, > intentionally it is aimed at being left "empty" and should not be > expanded. So, either the value is set or should remain empty Do you mean that you want (a) and (b) at the same time for the same leaf? [Oscar] No. Depending on the leaf, we would like to specify behavior a or behavior b. Behavior a is ok for most of the cases. The problem is that in some cases, assigning a value has way more implications and the service will not work properly. Those case are the ones we wanted to specifically tackle. > What is the best way to model this behavior? I see that some yang > modules have added an "auto-assignment" leaf to express if > auto-assignment is desired or not. (hence, auto-assignment false, and > leaf not set, would do not assign). > > Which is the "default" rule for a leaf that is not set? It is that the > system is free to create it (via template or any means of > auto-assignment) or should leave it as is, that is, empty? > > In NMDA, the system is allowed to expand a given configuration. This > fact, in my personal view, implies that by "default" any system could > implement the "auto-assignment" behavior being compliant with > Neconf/Restconf/NMDA rules (but I am not sure if the interpretation is > correct). There are (at least) three ways to interpret "auto-assign". The client writes to running, and then the server auto-assigns X: (a) in running (b) in intended (c) in the operational state (c) is uncontroversial and simple to implement in all servers, and simple to understand. [Oscar] agree (b) is allowed by NMDA but requires more of the server implementation; specifically it requires the server to support that intended is different from running. [Oscar] Agree . "Theoretically speaking" this is the behavior I would consider strictly follows NMDA guidelines. Reality is implementations are yet far from this... (a) is not recommended in general; running should be fully owned by the client(s) and not modified by the server. [Oscar] Agree. [Oscar] So... what would be the best way to specify the behavior? Explicitly adding an auto-assign leaf to identify the behavior? Just "obey" NMDA rules? /martin > > Best Regards, > > Óscar > > > > Este mensaje y sus adjuntos se dirigen exclusivamente a su > destinatario, puede contener información privilegiada o confidencial y > es para uso exclusivo de la persona o entidad de destino. Si no es > usted. el destinatario indicado, queda notificado de que la lectura, > utilización, divulgación y/o copia sin autorización puede estar > prohibida en virtud de la legislación vigente. Si ha recibido este > mensaje por error, le rogamos que nos lo comunique inmediatamente por > esta misma vía y proceda a su destrucción. > > The information contained in this transmission is privileged and > confidential information intended only for the use of the individual > or entity named above. If the reader of this message is not the > intended recipient, you are hereby notified that any dissemination, > distribution or copying of this communication is strictly prohibited. > If you have received this transmission in error, do not read it. > Please immediately reply to the sender that you have received this > communication in error and then delete it. > > Esta mensagem e seus anexos se dirigem exclusivamente ao seu > destinatário, pode conter informação privilegiada ou confidencial e é > para uso exclusivo da pessoa ou entidade de destino. Se não é vossa > senhoria o destinatário indicado, fica notificado de que a leitura, > utilização, divulgação e/ou cópia sem autorização
Re: [netmod] Question on how to design a Yang model to reflect auto-asignment of a give leaf
Hi, Oscar González de Dios wrote: > Dear OPSAWG and Netmod colleagues, > > During last IETF Opsawg meeting we raised a question (and > there was some discussion during the meeting) that we have > found yet no good answer and we would like to discuss it with > operations and Yang experts. > > The use case is the following: We have a yang module which > holds certain optional leafs. The behaviors that we would like > to have (and distinguish between them) are: > > > a) The user does not provide the value and such value is auto-assigned > by the system (a device (if it is a device module) or a controller (if > it is a network/service module)). I assume that this value not a static default value? > b) The user does not provide a value and wants that such value IS NOT > set by the system (as assigning a value has implications). That is, > intentionally it is aimed at being left "empty" and should not be > expanded. So, either the value is set or should remain empty Do you mean that you want (a) and (b) at the same time for the same leaf? > What is the best way to model this behavior? I see that some yang > modules have added an "auto-assignment" leaf to express if > auto-assignment is desired or not. (hence, auto-assignment false, and > leaf not set, would do not assign). > > Which is the "default" rule for a leaf that is not set? It is that the > system is free to create it (via template or any means of > auto-assignment) or should leave it as is, that is, empty? > > In NMDA, the system is allowed to expand a given configuration. This > fact, in my personal view, implies that by "default" any system could > implement the "auto-assignment" behavior being compliant with > Neconf/Restconf/NMDA rules (but I am not sure if the interpretation is > correct). There are (at least) three ways to interpret "auto-assign". The client writes to running, and then the server auto-assigns X: (a) in running (b) in intended (c) in the operational state (c) is uncontroversial and simple to implement in all servers, and simple to understand. (b) is allowed by NMDA but requires more of the server implementation; specifically it requires the server to support that intended is different from running. (a) is not recommended in general; running should be fully owned by the client(s) and not modified by the server. /martin > > Best Regards, > > Óscar > > > > Este mensaje y sus adjuntos se dirigen exclusivamente a su > destinatario, puede contener información privilegiada o confidencial y > es para uso exclusivo de la persona o entidad de destino. Si no es > usted. el destinatario indicado, queda notificado de que la lectura, > utilización, divulgación y/o copia sin autorización puede estar > prohibida en virtud de la legislación vigente. Si ha recibido este > mensaje por error, le rogamos que nos lo comunique inmediatamente por > esta misma vía y proceda a su destrucción. > > The information contained in this transmission is privileged and > confidential information intended only for the use of the individual > or entity named above. If the reader of this message is not the > intended recipient, you are hereby notified that any dissemination, > distribution or copying of this communication is strictly > prohibited. If you have received this transmission in error, do not > read it. Please immediately reply to the sender that you have received > this communication in error and then delete it. > > Esta mensagem e seus anexos se dirigem exclusivamente ao seu > destinatário, pode conter informação privilegiada ou confidencial e é > para uso exclusivo da pessoa ou entidade de destino. Se não é vossa > senhoria o destinatário indicado, fica notificado de que a leitura, > utilização, divulgação e/ou cópia sem autorização pode estar proibida > em virtude da legislação vigente. Se recebeu esta mensagem por erro, > rogamos-lhe que nos o comunique imediatamente por esta mesma via e > proceda a sua destruição ___ netmod mailing list netmod@ietf.org https://www.ietf.org/mailman/listinfo/netmod
Re: [netmod] Question on how to design a Yang model to reflect auto-asignment of a give leaf
Isnt it possible to handle case b) by defining a value to have the meaning no value has been assigned and then the user an explicitly set that value? Adrian From: netmod On Behalf Of Oscar González de Dios Sent: 11 February 2020 02:40 To: ops...@ietf.org; netmod@ietf.org Subject: [netmod] Question on how to design a Yang model to reflect auto-asignment of a give leaf Dear OPSAWG and Netmod colleagues, During last IETF Opsawg meeting we raised a question (and there was some discussion during the meeting) that we have found yet no good answer and we would like to discuss it with operations and Yang experts. The use case is the following: We have a yang module which holds certain optional leafs. The behaviors that we would like to have (and distinguish between them) are: a. The user does not provide the value and such value is auto-assigned by the system (a device (if it is a device module) or a controller (if it is a network/service module)). b. The user does not provide a value and wants that such value IS NOT set by the system (as assigning a value has implications). That is, intentionally it is aimed at being left empty and should not be expanded. So, either the value is set or should remain empty What is the best way to model this behavior? I see that some yang modules have added an auto-assignment leaf to express if auto-assignment is desired or not. (hence, auto-assignment false, and leaf not set, would do not assign). Which is the default rule for a leaf that is not set? It is that the system is free to create it (via template or any means of auto-assignment) or should leave it as is, that is, empty? In NMDA, the system is allowed to expand a given configuration. This fact, in my personal view, implies that by default any system could implement the auto-assignment behavior being compliant with Neconf/Restconf/NMDA rules (but I am not sure if the interpretation is correct). Best Regards, Óscar _ Este mensaje y sus adjuntos se dirigen exclusivamente a su destinatario, puede contener información privilegiada o confidencial y es para uso exclusivo de la persona o entidad de destino. Si no es usted. el destinatario indicado, queda notificado de que la lectura, utilización, divulgación y/o copia sin autorización puede estar prohibida en virtud de la legislación vigente. Si ha recibido este mensaje por error, le rogamos que nos lo comunique inmediatamente por esta misma vía y proceda a su destrucción. The information contained in this transmission is privileged and confidential information intended only for the use of the individual or entity named above. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this transmission in error, do not read it. Please immediately reply to the sender that you have received this communication in error and then delete it. Esta mensagem e seus anexos se dirigem exclusivamente ao seu destinatário, pode conter informação privilegiada ou confidencial e é para uso exclusivo da pessoa ou entidade de destino. Se não é vossa senhoria o destinatário indicado, fica notificado de que a leitura, utilização, divulgação e/ou cópia sem autorização pode estar proibida em virtude da legislação vigente. Se recebeu esta mensagem por erro, rogamos-lhe que nos o comunique imediatamente por esta mesma via e proceda a sua destruição ___ netmod mailing list netmod@ietf.org https://www.ietf.org/mailman/listinfo/netmod
