DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://nagoya.apache.org/bugzilla/show_bug.cgi?id=22870>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=22870

Apache 2.0 logging bogus transfer info





------- Additional Comments From [EMAIL PROTECTED]  2003-11-25 12:24 -------
In a comment on 2002-10-16 (over a year ago) in bug 10459 (which has been 
closed), Joshua identified the root of the problem:

>But there is a bug in apache related to this, which is that it always logs the
>full amount that it intends to send, rather than the amount it actually put
>on the wire.  There is active work to correct this, but it is not an easy
>problem.

My recent trial of Apache 2.0 revealed exactly this issue.

(BTW, this ALSO applies to requests with status 200 as well, not just 206.)

Using %I and %O (mod_log_io) does not help.  I do not want the bytecount to 
include header bytes - only content bytes.

Apache 2.0 is useless to us until this gets fixed.  Our business is downloads, 
and we need accurate byte counts, *especially* to indicate aborted downloads.  
We had to throw away several days worth of download logs because the 
bytecounts were so extremely inflated.  We *never suspected* that this basic 
logging behavior would have changed from Apache 1.3!

In bug 8996 there is a comment on 2003-02-16 about this 'feature' being 
intentional because "Some rely on having the %b and %B be what was intended to 
be written rather than what was really written - don't ask - it was a major 
point of contention on the development lists.".

What a crock!  Let *those* users be forced to use some snotty compile-time 
option that the Apache 2.0 developers seem to be so inordinately fond of 
(e.g. -DI_INSIST_ON_EXTRA_CYCLES_FOR_CLF_COMPLIANCE - see bug 21500).

Perhaps something like -DI_INSIST_ON_NON_CLF_COMPLIANT_BOGUS_BYTECOUNTS.  ;-P

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to