cnfs cycbuff problem

[Joe St Sauver]

>Date: Tue, 11 Apr 2000 11:59:37 -0700
>From: Kevin McKinnon <kevin at sunshinecable.com>
>Subject: Re: cnfs cycbuff problem
>To: inn-workers at isc.org, ab at icsmedia.de

[snip]

>> 	 Buffer REST02, len: 50.00 Mbytes, used: 12.10 Mbytes (24.2%)   1 cycles
>> 	  Newest: 2000-04-11 11:53:12,    0 days,  0:00:01 ago
>> 
>>  
>> After that lines, I think I misunderstand the cycbuff. 
>> Is that correct, that the innd erase all exist messages in a store ,
>> if he reached the next cycle ? I think that were really bad.
>
>Hi Axel,
>
>The information shown as used is a bit misleading.
>
>After the buffer has rolled over, it is always 100% full.  The 
>utilization shown is the point in the buffer that is being written 
>to.  So anything in the first 24.2 % is new, anything in the rest of 
>the buffer is still there (and readable) from the previous cycle.

Related to that, wouldn't it be more useful to report the OLDEST rather
than the Newest content in the buffer? I've often wondered about that...

I understand that it is nice to know if the buffer is being used to store
new content, but it would also be nice to know the relative "expiration"
being obtained on a given buffer (e.g., nothing older than 3 days on a 
binaries buffer, nothing older than 26.2 days on a big 7 buffer, whatever).

Regards,

Joe