You are viewing a plain text version of this content. The canonical link for it is here.
Posted to modperl@perl.apache.org by Oscar Serrano <os...@fondos.net> on 2002/01/16 14:09:54 UTC
weird problem. Lost of the POST data
Hello:
Perhaps some of you have had this problem before, but searching throug the
archive I've haven't seen nothing similar.
First, I need to be sure of one thing:
If I have 30 mod_perl (Apache::Registry) cgis in my system, once the all
are run, I must have 30 httpd process in memory; is that right?
I don't really know what I've done, but now, when I am in a web page and I
submit the content of a form by the POST method, the mod_perl cgi that
should receive that data doesn't receive it. Then I look at the access log
and I see this:
62.37.124.2 - - [16/Jan/2002:12:31:35 +0100]
"pa=fondos&ticket=0&language=es&group=ANY&name=bbva+2007&gestora=ANYGET
/graficos/fondos/arrowup.gif HTTP/1.1" 501 522
"http://www.fondos.net/buscafondos/buscafondos.mpcgi" "Mozilla/4.0
(compatible; MSIE 5.5; Windows 98; Win 9x 4.90)"
What we have here is that the mod_perl cgi
(http://www.fondos.net/buscafondos/buscafondos.mpcgi) is called, then this
cgi generate a page in which we have some .gif files. But the first .gif
file of that page is not shown because is called in a bad way. Before the
GET we see all the parameters the form sent via POST.
Why is happening this? This only happens with mod_perl cgis. I also run
some CGI's but they do not throw this error.
There is something that may give a clue. When I restart apache, it takes
some minutes to throw the first error. But the error get more and more
frecuently. But not always I get the error. It seems as if some httpd child
get corrupted and then, when one of those corrupted childs processes a
request, it throws this error.
Also tell you, that when I start Apache I get 6 httpd process (that's what
I have in my startservers directive in httpd.conf) and then the number of
httpd child process changes constantly. I have more that 50 mod_perl cgis
but normally I only have 9 or 10 httpd process and execpt for the six
first, the others die and born constatly, as if the where attending normal
CGI and not mod_perl CGI. The size in memory of this http process starts at
8184 bytes but grows until 22180. This maybe normal because of caching
modules I use in some CGIs. But, shouldn't I have one httpd process for
each mod_perl CGI executed? At least I think that's how I had it. But
perphaps I've done something in some module of my own that makes Apache
work so weirdly.
Thank you very much.
oscar