Hi Michael,

 

I agree with you that there is a hole when subsequent store requests have the 
same storage_time value.

But should every equality case be responded with a TOO-OLD error?

 

If there is  only one writer to the data object, it seems like a repetition, 
and it makes sense to discard it like you said.

Whereas if there are more than one writers to the data object, I am not sure 
what is expected. 

 

Do we have to consider the multi-writer case?

 

BR

Lingli

 

发件人: [email protected] [mailto:[email protected]] 代表 Michael Chen
发送时间: 2013年7月5日 1:52
收件人: [email protected]
主题: [P2PSIP] storage_time check wording inconsistency

 

Hi,

 

In the current p2psip-base-26, all descriptions of storage_time check (e.g 
7.4.1.1 and 13.5.3) states that a Store request must satisfy condition:

    new_item->storage_time > existing_item->storage_time

 

However, these two places have wordings to the contrary in terms of equality in 
the above comparison:

 

Section 7. Data Storage Protocol

   storage_time
      ... Any attempt to store a data value with a storage
      time before that of a value already stored at this location MUST
      generate a Error_Data_Too_Old error.

Section 6.3.3.1

   Error_Data_Too_Old:  A store cannot be completed because the
      storage_time precedes the existing value.

The word "before" must be changed to "older than or equal to", and

the word "precedes" must be changed to "is older than or equal to".

 

Thanks

 

--Michael

_______________________________________________
P2PSIP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/p2psip

Reply via email to