www-apache-bugdb mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mathieu GUILLAUME <...@isdnet.net>
Subject os-bsdi/223: Child processes don't die.
Date Sun, 09 Mar 1997 11:00:01 GMT

	The contract type is `' with a response time of 3 business hours.
	A first analysis should be sent before: Mon Mar 10 11:00:01 PST 1997

>Number:         223
>Category:       os-bsdi
>Synopsis:       Child processes don't die.
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    apache (Apache HTTP Project)
>State:          open
>Class:          sw-bug
>Submitter-Id:   apache
>Arrival-Date:   Sun Mar  9 03:00:01 1997
>Originator:     mat@isdnet.net
>Release:        1.2b7
BSDI 2.1, patchlevel 27 (i.e. including timeouts for FIN_WAIT_2).
Compiler is gcc.
This is very similar to the bug number 23 in the database :
the server runs for some time, then freezes, usually when the load
increases, because child processes are not closed (in the morning, I
find the server frozen with the max number of child processes allowed.
Those processes are several hours old).
I tried to disable keepalives, reduce the timeouts, ... to no avail.
Bug 23 in the database was reported fixed in 1.2b7 : I believe it is
the same and it was not really fixed...
I've got no idea how you could simulate this on a regular server.
Nope :( I've got no idea where it comes from. Perhaps mod_cgi, if it
IS the same bug as number 23 in the database

View raw message