httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rob Hartill <>
Subject [BUG]: "Apache httpd's consuming lots of time" on Solaris 2.x (fwd)
Date Tue, 18 Feb 1997 14:55:46 GMT

not acked.

---------- Forwarded message ----------
Date: Tue Feb 18  5:30:05 1997
Subject: [BUG]: "Apache httpd's consuming lots of time" on Solaris 2.x

Operating system: Solaris 2.x, version: 
Version of Apache Used: 1.1.3
Extra Modules used: mod_proxy (compiled in, not used here)
URL exhibiting problem: 

   We are seeing httpd's consuming lots of CPU time, in what
appears to be a tight loop as noted in this stack trace.  It is
difficult for us to tell if this is a bug, but it looks like it
is in an I/O loop but not managing to do any I/O despite the fact
that it is spinning away.

   Based on the source, this appears to be in the code that does
POST requests, but I'll defer to your more experienced judgement
there.  Let me know if you need any additional data to
assist you in debugging

-- Sincerely
Rand S. Huntzinger

Stack trace and I/O statistics for a httpd process follows:

Function Name             + Offset     Current PC   File Name
====================      ===========  ==========   =========
bread                     20           0x00033C24   httpd
read_client_block         56           0x0002B928   httpd
cgi_handler               1992         0x00048860   httpd
invoke_handler            316          0x000231C4   httpd
process_request_internal  1076         0x000276DC   httpd
process_request           16           0x00027748   httpd
child_main                2052         0x0001D994   httpd
make_child                276          0x0001DC7C   httpd
standalone_main           1716         0x0001E67C   httpd
main                      740          0x0001E994   httpd

I/O Sample:
                                                         Rate per Second
                                  Totals      Delta     (Total)   (Delta)
Input Blocks:                          0            0
Output Blocks:                         0            0
Messages send:                        50            0      0.0      0.0
Messages received:                    66            0      0.1      0.0
Bytes read+writen:                199585            0    177.9      0.0

NOTE: The I/O Total's aren't changing but the I/O rate per second
is declining.  This suggests that the code read 199,585 bytes and
then got stuck.




View raw message