George, Wes
2014-02-18 17:55:20 UTC
I had a fairly quick pass through this draft before the meeting and have some comments, as an individual (no WG chair hat) unless otherwise specified:
Section 2, 2nd paragraphâ itâs unclear what this testing that youâre referring to is. If itâs external (e.g. A whitepaper or other published study), a citation would be useful. If itâs internal, some information on who, what, where, when and methodology used would help to establish the credibility of the testing as something that adds value as a data point in this discussion.
3.1 â probably should cite the relevant RFCs/sections for NAT vs NAPT. While many readers may be familiar with the difference, the reference is useful for those who are not.
3.3 âfeature of double-translationâ â citation/more explanation needed
4.1/4.4 â <WG Chair hat> I thought that the direction of the WG was to integrate the referenced drafts so that we had one draft that covers the solution space, not continue to reference them externally from an overview draft. Since there is no longer a Behave WG, these referenced drafts are orphans, and unless consensus changes, Sunset4 will not be progressing multiple overlapping NAT/CGN logging/mapping drafts. There is significant overlap between the solutions proposed by Donley and Tsou such that they can be merged into one solution with a couple of different variants of port assignment, I.e. Tsou is the simple case, and Donley is a potential optimization. This along with some discussion as to the pros/cons of each will be useful guidance.
</WG Chair hat>
5.1 âa testing was madeâ â see comment for section 2. Also, details about how the 40T value was reached (what amount of data per record, number of records per unit time. See donleyâs intro section for an example. Also, I think that this section is really something that belongs earlier in the document, since it is the reason for there to be multiple methods of port allocation. This is the problem statement to the solutions being provided in donley/tsou.
There also needs to be discussion of state sync and failover/HA for the different solutions â how are things like existing mappings managed if a CGN fails and traffic switches to a backup, how do logging messages get updated, etc. See Donley section 3.1
Thanks,
Wes
Anything below this line has been added by my companyâs mail server, I have no control over it.
-----------
________________________________
This E-mail and any of its attachments may contain Time Warner Cable proprietary information, which is privileged, confidential, or subject to copyright belonging to Time Warner Cable. This E-mail is intended solely for the use of the individual or entity to which it is addressed. If you are not the intended recipient of this E-mail, you are hereby notified that any dissemination, distribution, copying, or action taken in relation to the contents of and attachments to this E-mail is strictly prohibited and may be unlawful. If you have received this E-mail in error, please notify the sender immediately and permanently delete the original and any copy of this E-mail and any printout.
Section 2, 2nd paragraphâ itâs unclear what this testing that youâre referring to is. If itâs external (e.g. A whitepaper or other published study), a citation would be useful. If itâs internal, some information on who, what, where, when and methodology used would help to establish the credibility of the testing as something that adds value as a data point in this discussion.
3.1 â probably should cite the relevant RFCs/sections for NAT vs NAPT. While many readers may be familiar with the difference, the reference is useful for those who are not.
3.3 âfeature of double-translationâ â citation/more explanation needed
4.1/4.4 â <WG Chair hat> I thought that the direction of the WG was to integrate the referenced drafts so that we had one draft that covers the solution space, not continue to reference them externally from an overview draft. Since there is no longer a Behave WG, these referenced drafts are orphans, and unless consensus changes, Sunset4 will not be progressing multiple overlapping NAT/CGN logging/mapping drafts. There is significant overlap between the solutions proposed by Donley and Tsou such that they can be merged into one solution with a couple of different variants of port assignment, I.e. Tsou is the simple case, and Donley is a potential optimization. This along with some discussion as to the pros/cons of each will be useful guidance.
</WG Chair hat>
5.1 âa testing was madeâ â see comment for section 2. Also, details about how the 40T value was reached (what amount of data per record, number of records per unit time. See donleyâs intro section for an example. Also, I think that this section is really something that belongs earlier in the document, since it is the reason for there to be multiple methods of port allocation. This is the problem statement to the solutions being provided in donley/tsou.
There also needs to be discussion of state sync and failover/HA for the different solutions â how are things like existing mappings managed if a CGN fails and traffic switches to a backup, how do logging messages get updated, etc. See Donley section 3.1
Thanks,
Wes
Anything below this line has been added by my companyâs mail server, I have no control over it.
-----------
________________________________
This E-mail and any of its attachments may contain Time Warner Cable proprietary information, which is privileged, confidential, or subject to copyright belonging to Time Warner Cable. This E-mail is intended solely for the use of the individual or entity to which it is addressed. If you are not the intended recipient of this E-mail, you are hereby notified that any dissemination, distribution, copying, or action taken in relation to the contents of and attachments to this E-mail is strictly prohibited and may be unlawful. If you have received this E-mail in error, please notify the sender immediately and permanently delete the original and any copy of this E-mail and any printout.