under heavy load, apache begins to start other processes of my fcgi-script.
since I am caching some basic code in this main process, I want every
request to be handled by the same fcgi-script. So how can I force this
behaviour?
under heavy load, apache begins to start other processes of my
fcgi-script. since I am caching some basic code in this main process, I
want every request to be handled by the same fcgi-script.
errr! meant to say "I want every request to be handled by the same
*process*"
google fastcgi and folows links to apache config directive for how to do this.
a warning : caching state is __exactly__ analogous to memory leak under nearly
all circumstances. consider which, if any, guarantee the cached data is
scrubbed and what limits the cache size.
under heavy load, apache begins to start other processes of my fcgi-script.
since I am caching some basic code in this main process, I want every
request to be handled by the same fcgi-script. So how can I force this
behaviour?
Any ideas?
--
email :: ara [dot] t [dot] howard [at] noaa [dot] gov
phone :: 303.497.6469
although gold dust is precious, when it gets in your eyes, it obstructs
your vision. --hsi-tang
under heavy load, apache begins to start other processes of my
fcgi-script. since I am caching some basic code in this main process, I
want every request to be handled by the same fcgi-script.
errr! meant to say "I want every request to be handled by the same
*process*"
under heavy load, apache begins to start other processes of my
fcgi-script. since I am caching some basic code in this main process, I
want every request to be handled by the same fcgi-script.
errr! meant to say "I want every request to be handled by the same
*process*"
I assume you have some state you are trying to maintain?
In general, I would have to waggle my finger in warning
against such practice
why? its simply that I have build a framework around the fcgi-script and
want to avoid this framework to be reloaded. In fact this was propagaded as
advantage of fastcgi: that a script doesn't have to be loaded on each
request.
my framework got so large that I don't want it so be several times in RAM.
but I am very interested to hear what disadvantages this configuration might
have to me.
is it simply an issue of threads-performance vs. RAM overhead?
benny
···
Le 24/4/2005, "benny" <listen@marcrenearns.de> a écrit:
but I am very interested to hear what disadvantages this configuration might
have to me.
is it simply an issue of threads-performance vs. RAM overhead?
benny
I use 2 * number of CPUs after doing some benchmarking.
The disadvantage to your configuration is that if you have more than 1 CPU. You might find 1 * number of CPUs more to your liking...
why? its simply that I have build a framework around the fcgi-script and
want to avoid this framework to be reloaded. In fact this was propagaded as
advantage of fastcgi: that a script doesn't have to be loaded on each
request.
Hmm... I've been wondering for some time if Instiki can be run under FCGI this way (it has to be only one process for each storage, due to Madeleine persistence).
my framework got so large that I don't want it so be several times in RAM.
but I am very interested to hear what disadvantages this configuration might
have to me.
is it simply an issue of threads-performance vs. RAM overhead?
You'll only be able to serve one client at a time, unless I misunderstand fcgi's
process loop model.
···
--
'This must be Thursday. I never could get the hang of Thursdays.'
-- Arthur Dent
Rasputin :: Jack of All Trades - Master of Nuns