Re: [ppsp] Reivew of the base tracker protocol

"Huangyihong (Rachel)" <rachel.huang@huawei.com> Fri, 24 October 2014 03:06 UTC

From: "Huangyihong (Rachel)" <rachel.huang@huawei.com>
To: "hishigh@sina.com" <hishigh@sina.com>, draft-ietf-ppsp-base-tracker-pro <draft-ietf-ppsp-base-tracker-protocol@tools.ietf.org>, ppsp <ppsp@ietf.org>
Thread-Topic: [ppsp] Reivew of the base tracker protocol
Thread-Index: AQHP7zB8wkizZtXH2UKe//j4ZQGajJw+iiQg
Date: Fri, 24 Oct 2014 03:06:18 +0000
Message-ID: <51E6A56BD6A85142B9D172C87FC3ABBB8624B861@nkgeml501-mbs.china.huawei.com>
References: <20141024021415.84CC9405C0@webmail.sinamail.sina.com.cn>
In-Reply-To: <20141024021415.84CC9405C0@webmail.sinamail.sina.com.cn>
Accept-Language: zh-CN, en-US
Content-Language: zh-CN
Content-Type: multipart/alternative; boundary="_000_51E6A56BD6A85142B9D172C87FC3ABBB8624B861nkgeml501mbschi_"
MIME-Version: 1.0
Archived-At: http://mailarchive.ietf.org/arch/msg/ppsp/Fb0FsjyuyD-frgUVy27h1kPuxq4
Subject: Re: [ppsp] Reivew of the base tracker protocol
Precedence: list

Hi Yunfei,

Please see inline.

BR,
Rachel

From: ppsp [mailto:ppsp-bounces@ietf.org] On Behalf Of yunfei
Sent: Friday, October 24, 2014 10:14 AM
To: draft-ietf-ppsp-base-tracker-pro; ppsp
Subject: [ppsp] Reivew of the base tracker protocol

Hi,

   (Speaking individually) I have reviewed the latest three versions of the base tracker protocol and have the following comments. Wish it helpful for today's online discussion.

General: I do like the idea in the latest two versions that the new section (section 3 and the corresponding update in section 4) has been added to address to accomodate different encodings in unified data type.

Comments:

1) Section 4.1:

typedef struct {

                     CONCRETE_PROTO_SPEC_DEPENDENT    address;

                     // string, addr_in4,

                     // addr_in6, addr_storage

                     CONCRETE_PROTO_SPEC_DEPENDENT    port;

                     // string (IANA service port),

                     // or uint16 (service port)

Is the definition of the type of address and port missing? I didn't find them before.

 [Rachel]: Yes. Comparing to previous xml format, the address type is missing.

2) In the end of section 5.1 and 5.2, there is an IMPLEMENTATION NOTE saying: If no PeerNum attributes are present in the request the tracker MAY return a random sample from the peer population.

It's suggested to set a DEFAULT value for the return number if no explict request of number.

 [Rachel]: Agree. The tracker should set a DEFAULT value. And I think this should also apply in the case of peer_count= 0.

BR

Yunfei
typedef struct {

CONCRETE_PROTO_SPEC_DEPENDENT address;

// string, addr_in4,

// addr_in6, addr_storage

CONCRETE_PROTO_SPEC_DEPENDENT port;

// string (IANA service port),

// or uint16 (service port)

CONCRETE_PROTO_SPEC_DEPENDENT address;

// string, addr_in4,

// addr_in6, addr_storage

CONCRETE_PROTO_SPEC_DEPENDENT port;

// string (IANA service port),

// or uint16 (service port)

CONCRETE_PROTO_SPEC_DEPENDENT address;

// string, addr_in4,

// addr_in6, addr_storage

CONCRETE_PROTO_SPEC_DEPENDENT port;

// string (IANA service port),

// or uint16 (service port)

[ppsp] Reivew of the base tracker protocol yunfei
Re: [ppsp] Reivew of the base tracker protocol Huangyihong (Rachel)