[Zope] zope crawls - the day after

Fri, 14 Jul 2000 21:38:57 +0200

"Bak @ kedai" wrote:
> 
> hi all
> 
> zope is faster now.  i think i know why, just wanted some confirmation from the
> zopistas.
> 
> setup:
> apache in front of zope on a rh 6.1 linux
> 
> what happened:
> -got a lot of simultaneous connect in one time.  whe the number of connect
> exceeds 200, zope started slowing down.   and i mean really slow.
> -tried putting squid in front, still not much improvement.
> -set up cache headers again.  i did this earlier on, but was not successful.
> 
> now:
> -got less then 100 simultaneous connect.  zope runs beautifully.
> 
> what i think that solved this:
> -setting the cache headers.  at first, i put the headers in
> standard_html_header, before any html documents.  later, i put it after
> <head><title> tags.  and i think i can see the cache headers on.
> 
> HTTP/1.1 200 OK
> Date: Tue, 11 Jul 2000 09:15:33 GMT
> Server: Apache/1.3.12 (Unix)  (Red Hat/Linux) mod_fastcgi/2.2.4 (mod_pcgi2/1.0.1; PCGI/2.0a5) PHP/3.0.12
> Cache-Control: must-revalidate,max-age=120
> Expires: Tue, 11 Jul 2000 09:17:06 GMT
> X-Powered-By: Zope (www.zope.org), Python (www.python.org)
> Content-Length: 14508
> Last-Modified: Mon, 10 Jul 2000 16:24:22 GMT
> Content-Type: text/html
> 
> so, i guess, it does matter where you put the cache headers.  is there a limit
> on how many simultaneous connect zope can handle?  apache ( i know it's
> different) can handle much higher traffic with not much complaints.

Uhm. Headers are headers, not content. As example, image files have
headers too but no
HTML inside. So you should set headers with RESPONSE.setHeader()

hth
Tino Wildenhain