Stas Bekman <[EMAIL PROTECTED]> writes: > On Mon, 11 Mar 2002, Stas Bekman wrote: > > > this looks like a copy-n-paste porting bug. in httpd-2.0's scoreboard > > stop/start timestamps are in apr_time_t (==usecs only). > > > > the problem wasn't noticed before since these structures are never set in > > workers. Any ideas why? See mod_status's output for yourself.
because unlike apache 1.3, nobody calls ap_time_process_request() :) (no idea why) > here is one more porting leftover fix, including the previous one: > > Index: modules/generators/mod_status.c > =================================================================== > RCS file: /home/cvspublic/httpd-2.0/modules/generators/mod_status.c,v > retrieving revision 1.54 > diff -u -r1.54 mod_status.c > --- modules/generators/mod_status.c 4 Mar 2002 10:46:22 -0000 1.54 > +++ modules/generators/mod_status.c 11 Mar 2002 09:43:25 -0000 > @@ -570,13 +570,11 @@ > req_time = 0L; > #endif /* HAVE_TIMES */ > #else > - if (ws_record.start_time == 0L > - && ws_record.start_time == 0L) > + if (ws_record.start_time == 0L) > req_time = 0L; > else > req_time = (long) > - (((ws_record.stop_time - ws_record.start_time) * 1000) > - + ((ws_record.stop_time - ws_record.start_time) / 1000)); > + ((ws_record.stop_time - ws_record.start_time) / 1000); > #endif > if (req_time < 0L) > req_time = 0L; I'll commit this shortly. -- Jeff Trawick | [EMAIL PROTECTED] Born in Roswell... married an alien...
