RFC 6121 section 8.5 "active resource" and "Connected Resource"
Hi, Section 8.5.4 provides a summary of earlier sections. It uses the term "active resource" twice. This term is not used in any of the earlier sections. Conversely, the earlier sections use the term "Connected Resource" a couple of times. A "Connected Resource" is defined by RFC6121/6120 as "[has] bound a resource to the stream" Although the term "Connected Resource" is used in section 8.5, none of the specifications therein seem to apply to Connected Resources. The usage of the term "active resource" in section 8.5.4 could be an artifact of RFC3921/3920, which defines an "active resource" as: "Upon establishing a session, a connected resource (in the terminology of [XMPP-CORE]) is said to be an "active resource." Although section 8.5 advertises that it will define how stanzas related to connected resources are to be processed ("If there is at least one available resource or connected resource, how the stanza is processed depends on the stanza type.") it does not seem to do that. All definitions seem to depend on having at least one available resource (which is defined as a connected resource that has sent initial presence), leaving scenarios in which only connected resources are present, undefined. Am I misinterpreting this? If not, is this something that should be cleared up (or explicitly be left undefined)? As a practical example of a question that I believe is left unanswered: How should a server process a message stanza of type 'normal', addressed to a bare JID that represents a local user (the scenario of section 8.5.2.1.1 of RFC 6121) if the corresponding user only has one or more Connected Resources (but not any Available Resources)? I believe that there are many possible variations on this question, I'm limiting it to one to illustrate the larger issue that I described in this email. For additional context: this was discussed in the XSF's Discussion MUC earlier. Logs of that conversation can be found at https://logs.xmpp.org/xsf/2025-05-28#2025-05-28-a5d887bd6428222c Kind regards, Guus
How should a server process a message stanza of type 'normal', addressed to a bare JID that represents a local user (the scenario of section 8.5.2.1.1 of RFC 6121) if the corresponding user only has one or more Connected Resources (but not any Available Resources)?
I think that RFC 6120 §10.5.3.2 covers this:
For a message stanza, if there exists at least one connected resource for the account then the server SHOULD deliver it to at least one of the connected resources. If there exists no connected resource then the server MUST either (a) store the message offline for delivery when the account next has a connected resource or (b) return a <service-unavailable/> stanza error (Section 8.3.3.19).
Ah, yes, that seems very relevant, thank you! As RFC 6120 does not describe "presence sessions", it doesn't describe "Available Resources" either (as those are closely related to presence). That's all in RFC 6121. In that reasoning, it makes sense to look back at RFC 6120 to find specifications related to "Connected Resources" (rather than "Available Resources". Two things do not sit well with me: - The usage of the term "active resource" in RFC 6121, which appears to be an artifact of RFC 3920/3921. I believe that term should not be used at all in RFC 6121. - The section titles in RFC 6121 suggest that specifications are provided for "Connected Resources", which they do not appear to do at all. Would it be appropriate to suggest editorial changes to the RFC, in an attempt to remove some ambiguity? Does such a mechanism even exist for RFCs? On Sat, Jun 7, 2025 at 8:46 PM Tedd Sterr <teddsterr@outlook.com> wrote:
How should a server process a message stanza of type 'normal', addressed to a bare JID that represents a local user (the scenario of section 8.5.2.1.1 of RFC 6121) if the corresponding user only has one or more Connected Resources (but not any Available Resources)?
I think that RFC 6120 §10.5.3.2 covers this:
For a message stanza, if there exists *at least one connected resource* for the account then the server SHOULD *deliver it to at least one of the connected resources*. If there exists * no connected resource* then the server MUST either (a) *store the message offline for delivery* when the account next has a connected resource *or* (b) *return a <service-unavailable/> stanza error* (Section 8.3.3.19).
_______________________________________________ Standards mailing list -- standards@xmpp.org To unsubscribe send an email to standards-leave@xmpp.org
On 6/13/25 5:08 AM, Guus der Kinderen wrote:
Ah, yes, that seems very relevant, thank you!
As RFC 6120 does not describe "presence sessions", it doesn't describe "Available Resources" either (as those are closely related to presence). That's all in RFC 6121. In that reasoning, it makes sense to look back at RFC 6120 to find specifications related to "Connected Resources" (rather than "Available Resources".
Two things do not sit well with me:
* The usage of the term "active resource" in RFC 6121, which appears to be an artifact of RFC 3920/3921. I believe that term should not be used at all in RFC 6121. * The section titles in RFC 6121 suggest that specifications are provided for "Connected Resources", which they do not appear to do at all.
Ah, I see what you mean - I think the one instance of "active resource" in the table within Section 8.5.4 of RFC 6121 is a typo (a holdover from RFC 3921, as you say) - it should be "available resource".
Would it be appropriate to suggest editorial changes to the RFC, in an attempt to remove some ambiguity? Does such a mechanism even exist for RFCs?
One can file errata against an RFC, but they are not incorporated until and unless the RFC is replaced by a new one (at least if they are "held for document update", as they usually are). In the case of RFCs 6120 and 6121, I haven't heard that anyone has volunteered to take on that task! It might make sense to do that eventually given that we have defined new ways to handle authentication and the like, but the process might require a new working group and that involves a lot of effort and overhead... Peter
On Sat, Jun 7, 2025 at 8:46 PM Tedd Sterr <teddsterr@outlook.com <mailto:teddsterr@outlook.com>> wrote:
> How should a server process a message stanza of type 'normal', addressed to a bare JID that represents a local user (the scenario of section 8.5.2.1.1 of RFC 6121) if the corresponding user only has one or more Connected Resources (but not any Available Resources)?
I think that RFC 6120 §10.5.3.2 covers this:
> For a message stanza, if there exists *at least one connected resource* for the account then the server SHOULD *deliver it to at least one of the connected resources*. If there exists *no connected resource* then the server MUST either (a) *store the message offline for delivery* when the account next has a connected resource *or* (b) *return a <service-unavailable/> stanza error* (Section 8.3.3.19).
_______________________________________________ Standards mailing list -- standards@xmpp.org <mailto:standards@xmpp.org> To unsubscribe send an email to standards-leave@xmpp.org <mailto:standards-leave@xmpp.org>
_______________________________________________ Standards mailing list -- standards@xmpp.org To unsubscribe send an email to standards-leave@xmpp.org
Meu amigo. Desculpa e aí. E q meu dados e de raíz. Não tive direito de ter um estudo para mim. Não. Sofrie o sofro muito mesmo... Em sex., 13 de jun. de 2025 20:17, Peter Saint-Andre <stpeter@stpeter.im> escreveu:
On 6/13/25 5:08 AM, Guus der Kinderen wrote:
Ah, yes, that seems very relevant, thank you!
As RFC 6120 does not describe "presence sessions", it doesn't describe "Available Resources" either (as those are closely related to presence). That's all in RFC 6121. In that reasoning, it makes sense to look back at RFC 6120 to find specifications related to "Connected Resources" (rather than "Available Resources".
Two things do not sit well with me:
* The usage of the term "active resource" in RFC 6121, which appears to be an artifact of RFC 3920/3921. I believe that term should not be used at all in RFC 6121. * The section titles in RFC 6121 suggest that specifications are provided for "Connected Resources", which they do not appear to do at all.
Ah, I see what you mean - I think the one instance of "active resource" in the table within Section 8.5.4 of RFC 6121 is a typo (a holdover from RFC 3921, as you say) - it should be "available resource".
Would it be appropriate to suggest editorial changes to the RFC, in an attempt to remove some ambiguity? Does such a mechanism even exist for RFCs?
One can file errata against an RFC, but they are not incorporated until and unless the RFC is replaced by a new one (at least if they are "held for document update", as they usually are). In the case of RFCs 6120 and 6121, I haven't heard that anyone has volunteered to take on that task! It might make sense to do that eventually given that we have defined new ways to handle authentication and the like, but the process might require a new working group and that involves a lot of effort and overhead...
Peter
On Sat, Jun 7, 2025 at 8:46 PM Tedd Sterr <teddsterr@outlook.com <mailto:teddsterr@outlook.com>> wrote:
> How should a server process a message stanza of type 'normal', addressed to a bare JID that represents a local user (the scenario of section 8.5.2.1.1 of RFC 6121) if the corresponding user only has one or more Connected Resources (but not any Available Resources)?
I think that RFC 6120 §10.5.3.2 covers this:
> For a message stanza, if there exists *at least one connected resource* for the account then the server SHOULD *deliver it to at least one of the connected resources*. If there exists *no connected resource* then the server MUST either (a) *store the message offline for delivery* when the account next has a connected resource *or* (b) *return a <service-unavailable/> stanza error* (Section 8.3.3.19).
_______________________________________________ Standards mailing list -- standards@xmpp.org <mailto:
standards@xmpp.org>
To unsubscribe send an email to standards-leave@xmpp.org <mailto:standards-leave@xmpp.org>
_______________________________________________ Standards mailing list -- standards@xmpp.org To unsubscribe send an email to standards-leave@xmpp.org
_______________________________________________ Standards mailing list -- standards@xmpp.org To unsubscribe send an email to standards-leave@xmpp.org
On 13/06/2025 13.08, Guus der Kinderen wrote:
Ah, yes, that seems very relevant, thank you!
As RFC 6120 does not describe "presence sessions", it doesn't describe "Available Resources" either (as those are closely related to presence). That's all in RFC 6121. In that reasoning, it makes sense to look back at RFC 6120 to find specifications related to "Connected Resources" (rather than "Available Resources".
Two things do not sit well with me:
* The usage of the term "active resource" in RFC 6121, which appears to be an artifact of RFC 3920/3921. I believe that term should not be used at all in RFC 6121. * The section titles in RFC 6121 suggest that specifications are provided for "Connected Resources", which they do not appear to do at all.
Would it be appropriate to suggest editorial changes to the RFC, in an attempt to remove some ambiguity?
Yes, please do so. An RFC errata record will be very helpful once a new revision of the RFCs is on its way. - Flow
participants (5)
-
Florian Schmaus -
George. Soares -
Guus der Kinderen -
Peter Saint-Andre -
Tedd Sterr