On Unix systems most daemon processes are written to trap a
SIGHUP sent to their proccess ID (PID)
and reinitialize.
This is common behaviour and handy if you either need to
propagate a change in the daemon's configuration,
or want it to release locks or file handles.
The latter is mostly required
Hello Joshua,
that was what I was suspecting (looking over the scoreboard
stats),
that the No. of MinSpareServers was way too high.
So the error message fooled me a little.
I have set it, as you suggested down to 10.
So what about the StartServers?
Don't you think 40 isn't a bit too high as well
Hello Apache Users/Admins,
I think I have a performance issue in that the master httpd
cannot fork enough children
to satisfy all concurrent connect requests,
because I get these in error_log:
# grep server\ seems\ busy $SERVERROOT/logs/error_log|tail
[Wed Aug 09 21:21:27 2006] [info] server seem
What's the content of /etc/logrotate.d/httpd,
could it be you are signalling the wrong process?
You also might want to check if these to relate to the same proc
$ ps -p $(cat /var/run/httpd.pid) -o args
$ ps -p $(pgrep -P 1 -u 0 -x httpd) -o args
> -Original Message-
> From: Farid Hamj
For instance assuming your Apache is 2.0,
have you looked here yet?
http://httpd.apache.org/docs/2.0/ssl/
http://httpd.apache.org/docs/2.0/mod/mod_ssl.html
http://httpd.apache.org/docs/2.0/vhosts/
> -Original Message-
> From: Michael W Cocke [mailto:[EMAIL PROTECTED]
> Sent: Wednesday,
You merely need to change the path to your apachectl of your 2nd
httpd in the init script.
e.g. here I have a Fedora 3 where httpd was installed from an FC3
RPM
# rpm -qf /etc/init.d/httpd
httpd-2.0.52-3
In this init script a variable apachectl is set as
# grep ^apachectl /etc/init.d/httpd
ap
I'm not that much into Solaris package management
but you could try if you could relocate the installation base
to a directory on which your user account has write permissions
(e.g. beneath your $HOME)
So you could try the -R switch (see man pkgadd)
Prior to installation you could check where the p
There's absolutely no need to use .htaccess (or other named files
that serve the same purpose),
and in fact the general recommendation is to avoid them unless
you are forced to give away
configuration privileges to third parties (e.g. web developers)
Read here
http://httpd.apache.org/docs/2.0/howt
I think that's what I meant by "embedded Linux HTTP server"
I simply forgot its name.
But of course should it been glaring to me that it only could be
named "tux" ;-)
> -Original Message-
> From: Eugene [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, March 01, 2006 8:13 PM
> To: users@httpd.a
Ok guys, got that.
I will have to watch the load on the server
and then possibly decide on compiling a lean static httpd with
only rudimentary functionality
(e.g. this one wouldn't need mod_ssl, mod_vhost, mod_include,
mod_rewrite, ...)
which exclusively serves my private LAN.
Maybe even CPAN's H
Hi Nick,
of course does it make sense to me.
Why would I want to have 10 preforked httpd children listening on
192.168.*.*:80
when I know that there will be only occassional single requests
to this private address.
As I said this URL should only server internal occassional
maintenance and stats r
Hi Joshua,
ouch, I knew I missed something vital.
You see, I errantly first tried to put the additional Listen
in the vhost container where it definitely didn't belong.
# apachectl configtest
Syntax error on line 5 of /var/www/apache/conf.d/vhosts.conf:
Listen cannot occur within section
That
Hello,
I think I have a flaw in my httpd.conf somewhere,
but the httpd syntax check isn't complaining.
I use this httpd release on Fedora 3
# /usr/sbin/httpd -v
Server version: Apache/2.0.52
Server built: Oct 15 2004 11:39:29
and it's a DSO build as supplied from Fedora 3
# /usr/sbin/httpd -
I think you simply mixed up the order of method calls from CGI.pm
Your script needs to write the HTTP header first.
That's why the "print header" must come before
"print start_html".
If you are using the methods from CGI.pm
(which is always preferable)
then you can safely ignore worrying about fo
In fact, Solaris has a utility command to administer who and to
what extend is allowed
to have core files generated.
See "man coreadm", or ask your sysadmin for help.
Its config file is /etc/coreadm.conf
Here is for instance a default setup of it
# cat /etc/coreadm.conf
#
# coreadm.conf
#
# Para
Hi AndreĀ“,
I removed the IfModule tags as you suggested.
I also removed the RedirectMatch directive from mod_alias
an instead load the bulk of mod_rewrite.
In the directory container I included these directives
Alias /loo/ /var/www/turd/
Options SymlinksIfOwnerMatch
RewriteEngin
Hi Nick,
no unusual umask
# umask
0022
# touch /tmp/touched
# ls -l /tmp/touched
-rw-r--r-- 1 root root 0 Feb 21 13:35 /tmp/touched
/tmp as with all my Unix boxes only has the sticky bit set
to prevent users' mutual file deletion.
That's why I was alarmed when I saw htdigest doing this.
Hi,
I needed to "protect" a URI of my webserver.
Since the Base64 encoding of mod_auth seemed too revealing to me,
and on the other hand SSL/TLS too involved
I looked for something in between and came accross
mod_auth_digest.
With mod_auth_digest comes the utility program htdigest
to create and m
Hello Apache httpd users,
I have just subscribed to this list to clarify a few things for
me that so far
I couldn't find in the abundance of Apache and its modules'
refeernce and howto
material.
Probably something like this has been asked a million of times
here and thus already
found sufficient
19 matches
Mail list logo