I have apache installed on a box with 2 nics. I've seen how to force the listener to an IP, but the issue is that this box uses DHCP. Is it possible to lock the listener to a network/subnet or to a specific NIC?
View 2 Replies
My company currently have some spam filtering problem with mailscanner and the Windows team was given a project to come up with a better solution to fight spams.
I work in Unix dept. I suggested to Windows admin I use ASSP personally and works great. I gave them the specs on my setups. Since the current front-end proxy is on RHEL, we all settled to try out ASSP on Ubuntu based server.
We scrap-find an older Dell PE 2850 we can use. I finally convinced company to deploy OpenVZ, this will be our first OpenVZ server public facing.
The Dell have two Intel 82541GI Gigabit NICs. We are VLAN-ing on Cisco switch level; eth0 will be on internal 10.0 network and eth1 on public port.
I already installed CentOS 5.2 plus HyperVM. I configured and brought up eth1 without TCP/IP, just on layer 2. Looks like OpenVZ is using eth0 right now.
For this new proxy, will be routing traffics through host eth1. What's the best way of going with this? The new proxy will be using veth so it will have its own MAC (for security reasons; and network team said this is mandatory).
Should I be using bridging? Or simple routing guests through VZ configured eth1 would work?
Can anyone give me some ideas? I'm asking in WHT is because a lot of hosting companies probably have this setup already. I'm just absorbing ideas...
i ask of cpanel.net that why : my apache any 1h going to down and i force run restart httpd
cpanel :
Howdy,
This is occurring because your server is consistently hitting the maximum
allowed concurrent connection limit, defined by MaxClients in httpd.conf:
[root@part7 ~]# grep MaxClient /usr/local/apache/logs/error_log | tail -5
[Mon Oct 22 13:07:46 2007] [error] server reached MaxClients setting, consider
raising the MaxClients setting
[Mon Oct 22 13:32:03 2007] [error] server reached MaxClients setting, consider
raising the MaxClients setting
[Mon Oct 22 14:03:29 2007] [error] server reached MaxClients setting, consider
raising the MaxClients setting
[Mon Oct 22 14:16:07 2007] [error] server reached MaxClients setting, consider
raising the MaxClients setting
[Mon Oct 22 16:48:36 2007] [error] server reached MaxClients setting, consider
raising the MaxClients setting
[root@part7 ~]#
You will need to adjust this setting to a value which better accommodates your
server to prevent any further downtime.
We are struggling to configure our Apache reverse proxy (on WIN 2008) server to force https.
We have the cert installed on the proxy server, and it seems to be working but we are unable to force connections to https: and the site is still available via http:
How do you enforce https on the site?
reading read about the .htaccess file, virtual hosts but still having a hell of a time putting it all together
Internet -> SSL -> Apache 2.2 -> HTTP -> web app
My issue is with my code I have it rewwrites my addon domains to the main domain. I have a addon domain exclusion line. But each time I add a domain I have to go in and edit the htaccess. I foresee lots of addon domains in my future.
View 2 Replies View RelatedDepending on where u are at on my site (documents pages, training, main root, etc.) will depend on which type of background, footer, header and the like you'll get. Now I was thinking. Is there a way to have multiple error messages for more then one page depending on where you are at on a site? Right now it's intranet site and a modded snitz forum. What is the code and were does it go and in which apache conf file(s) does it go in?
View 1 Replies View RelatedI have some instructions written in the .htaccess of root directory for my website. So how to force those rules to apply to all sub directories even if those sub directories have .htaccess that overwrite the rules of the mother .htaccess on the root directory ? For example the root .htaccess deny access to all *.log files, so how to force that rule even if in some subdirectories there is some .htaccess files that allow access of *.log files.
View 1 Replies View Relatedfor a project i'm setting up a WDS server for rapid deployment of Windows 2008 servers.
The information regarding WDS and automated deployments is clear for me.
The only, and probably biggest problem is how to configure the linux dhcp/pxe server to able to boot an image from the WDS to start the installation.
Unfortunately I haven't found much information about it.
I hope some one here can push me in the right direction.
I'm currently working on a little project to get a rescue image loaded through PXE. It shouldn't be that difficult, but personally I don't really work with DHCP (obviously we usually just use static IPs), and DHCP is required for PXE.
I've setup the PXE host server and it's also running DHCPd without errors. This is the configuration of dhcpd.conf:
Code:
ddns-update-style ad-hoc;
option subnet-mask 255.255.255.0;
option domain-name-servers x.x.x.37;
subnet x.x.x.0 netmask 255.255.255.0 {
range x.x.x.155 x.x.x.159;
option broadcast-address x.x.x.255;
option routers x.x.x.1;
next-server x.x.x.153;
filename "pxelinux.0";
max-lease-time 100;
default-lease-time 100;
}
(I have replaced the first three numbers of the IPs with x.x.x for privacy reasons)
That configuration seems fine.. Now the problem is that when I try to get a server to boot with PXE, after a minute or so, it reports that it hasn't received any DHCP offers.
Some side information: both the host and client server are on the same network, but not on the same switch though. I guess that shouldn't matter, or does it?
I follow instruction from redhat to setup my Centos acts like a router. When i setup the dhcp server, at the end when i start the dhcp, it keeps getting me the failed error. So i believe that i need to use 'touch' command for first time start dhcp server but seem that no such file or directory for dhcpd.leases.
Quote:
[root@localhost dhcp-3.0.5]# touch /var/lib/dhcp/dhcpd.leases
touch: cannot touch `/var/lib/dhcp/dhcpd.leases': No such file or directory
I just had a question about servers that have 2 Network Cards...
Does each card end up having a different IP address?
Like are they there for redundancy reasons or is the reason behind having 2 just to aid in processing power?
If card #1 fails does card #2 automatically "pick up the slack?"
I'm using XEN to setup my VPS. The installation have no problem for the first two server. On the 3rd server, I install using DHCP IP and it's ok.The problem is when I change the DHCP IP to static IP .
Using Centos 5.2, I'm using vi to edit /etc/sysconfig/network-scripts/ifcfg-eth0
and restart the network service network restart, the IP change, but outside or other static IP VPS network can't ping to this server.
I've tried to change the /etc/xen/the-host-config to include the IP . the results is still the same. Is there some script or way I can use to bridge this new 3rd server to the Xen host ant other network?
* I've restart the Xen server.
some gigabit NICs for server upgrading. Will use with CentOS (mixed of 32, 64 bits, centOS 4-5) on PCI slot. Which brand do you suggest? I don't want to face to compatibility issue
View 7 Replies View RelatedI read WHT quite a bit and I'd rather not weigh in on any provider issues with a 1 post count, so I'll ask a generic question of my own.
Is there any particular reason that Windows Updates seem to periodically cause network inaccessibility? Particularly with our PD-SMI+ boards I notice patch days bring at least 10 boxes down. No known solution other than to leave settings identical, disable, then enable the network adapters.
Does anyone have experience comparing the two? Any comments on why one would be better over the other?
View 7 Replies View RelatedCan i run multiple sites on apache? and if so how do i do it.
View 3 Replies View RelatedApache keeps stopping. MULTIPLE times per day! There is no logic to when it dies. But about every 2 hours.
Load stays below .30 and there is free memory available.
This is on a VPS machine. None of the other VPS's are having an issue. Just this one.
Centos release 5.3 (Final)
Apache/2.2.3
Here is what is in the httpd.conf file. I realize the numbers are way too high, but just trying to get this issue to go away.
Code:
<IfModule prefork.c>
StartServers 100
MinSpareServers 100
MaxSpareServers 100
ServerLimit 512
MaxClients 512
MaxRequestsPerChild 4000
</IfModule>
<IfModule worker.c>
StartServers 100
MaxClients 500
MinSpareThreads 25
MaxSpareThreads 75
ThreadsPerChild 25
MaxRequestsPerChild 100
</IfModule>
Here is what is in the /var/log/httpd/error_log file before it dies:
Code:
[Wed Jul 01 18:06:32 2009] [info] server seems busy, (you may need to increase StartServers, or Min/MaxSpareServers), spawning 8 children, there are 98 idle, and 108 total children
[Wed Jul 01 18:08:17 2009] [info] server seems busy, (you may need to increase StartServers, or Min/MaxSpareServers), spawning 8 children, there are 74 idle, and 76 total children
[Wed Jul 01 18:08:18 2009] [info] server seems busy, (you may need to increase StartServers, or Min/MaxSpareServers), spawning 16 children, there are 63 idle, and 63 total children
[Wed Jul 01 18:08:19 2009] [info] server seems busy, (you may need to increase StartServers, or Min/MaxSpareServers), spawning 32 children, there are 79 idle, and 79 total children
[Wed Jul 01 18:11:36 2009] [info] server seems busy, (you may need to increase StartServers, or Min/MaxSpareServers), spawning 8 children, there are 93 idle, and 108 total children
Our servers has been updated recently and my site is not working properly. It was working before the server change. When I click on the links, it doesn't update correctly, instead it gets the last cached data. I have to refresh each link for it to update which shouldn't be the case. I'm a front-end and have no access to servers. Could this be server-related?
View 1 Replies View RelatedI wanted to configure 2 clusters on Apache 2.4. Here are my sample configurations
Cluster 1 config:
----------------
<VirtualHost *:80>
ServerName mywebsite1.com
<IfModule mod_proxy_balancer.c>
<Proxy balancer://sgcluster>
BalancerMember http://IPaddress1:80/ loadfactor=1
BalancerMember http://IPaddress2:80/ loadfactor=2
ProxySet lbmethod=byrequests
[Code] ....
I've included both configuration file in httpd.conf. I'm getting the errors, when I tried to start up the apache.
[Thu May 02 11:01:34.243583 2013] [slotmem_shm:debug] [pid 9240606:tid 1] mod_slotmem_shm.c(448): AH02301: attach looking for /opt/apache2/logs/slo
tmem-shm-mod_heartmonitor.shm
[Thu May 02 11:01:34.243667 2013] [lbmethod_heartbeat:notice] [pid 9240606:tid 1] AH02282: No slotmem from mod_heartmonitor
[Code] .....
How do I add multiple web pages to my server?
View 6 Replies View RelatedI very much like the "readme.txt" that Apache appends to the directory listings. That is, when a browser GETs a directory, the text in that file is put underneath the list of files therein.
View 1 Replies View RelatedI am completely new to apache and I had the load balancer set up for a single domain that was working perfectly fine, but when it came to multiple domains i couldn't, for the life of me, figure out what to do.
There is example.com sub.example.com and otherexample.com
I had a single node set up for this but I want to experiment with the balancer. I have scratched Google and I only found broken info which couldn't explain much to me. Here are my domains
<VirtualHost *:80>
DocumentRoot "/var/www/domain.com"
ServerName domain.com
</VirtualHost>
[Code] ....
Any example configuration of the default file.
I inherited an Apache HTTP instance from a colleague. That's why I'm not deep into that stuff. My current scenario is the following: We are running several backend instances which are published to the internet via Apache HTTP Server which is configured as Reverse Proxy (RP). In the past there was only one port and one protocol to handle by the Apache HTTP (RP).
Now I have the following need. There has to be published a Web Service from one instance using two Ports with two protocols (HTTP and HTTPs). I'm not sure how to solve that need.
A solution I tried already was to change and add the variable DEF_BACKEND_PORT to DEF_BACKEND_PORT_1 and DEF_BACKEND_PORT_2 but it didn't worked.
I must say that I tried a lot of possibilities but my knowledgment in Apache WS is not too much advanced.
Apache Web Server version: 2.2.22
OS: Centos 6 64 bits
Tomcat version: 7.0.57
What I need to achieve is the following:
I have some plain websites defined this way:
<VirtualHost *:80>
ServerAdmin example1@example1.com
DocumentRoot "/opt/sites/example1/"
ServerName example1.com
ErrorLog logs/example1-error_log
CustomLog logs/example1-access_log common
RedirectMatch permanent ^/(.*) http://www.example1.com/$1
[Code] ....
I have my Apache 2.4.9 for Windows with multiple Virtual Hosts, the main one is configure like this:
Code:
NameVirtualHost vhost1.domain.org:80
<VirtualHost vhost1.domain.org:80>
ServerName vhost1.domain.org
DocumentRoot "C:/Apache24/htdocs"
ServerAdmin techsupport@domain.org
# ErrorLog "logs/vhost1.domain.org-error.log"
# CustomLog "logs/vhost1.domain.org-access.log" common
</VirtualHost>
The rest like this:
Code:
NameVirtualHost vhost2.domain.org:80
<VirtualHost vhost2.domain.org:80>
ServerName vhost2.domain.org
DocumentRoot "C:/Apache24/htdocs/vhost2"
ServerAdmin techsupport@domain.org
# ErrorLog "logs/vhost2.domain.org-error.log"
# CustomLog "logs/vhost2.domain.org-access.log" common
</VirtualHost>
There is no *:80 anywhere and still when I go to any vhost it always load the main one.
What else do I need to check?
I'm kinda new to apache and I have the following situation.
I have a vps, on which I have set the following:
WordPress Site A, has it's own VirtualHost and domain
WordPress Site B, has it's own VirtualHost and domain
Canvas LMS (not connected to a domain).
Canvas LMS can add multiple accounts, which may (or may not) be accessed individually using a URL that looks as follows:
{canvas_root}/accounts/{account_id}/
For example : http://....../accounts/4/
I need to add a link to each of WP sites, which points to each site's account on Canvas, such that it appears as .../learning/ folder
For example:
Code:
http://WPSiteA/learning/
http://WPSiteB/learning/
which actually represents
Code:
http://WPSiteA/accounts/4/
http://WPSiteB/accounts/5/
Which in turn represents
Code:
http://{my_vps_ip}/accounts/4/
http://{my_vps_ip}/accounts/5/
respectively
Noticing the following:
I do not want the users to see the /account/4/ , just /learning/
I also want to prevent accessing
Canvas LMS is located beyond the website's DocumentRoot
I want this to work with both HTTP and HTTPS
when I had multiple (at least 5) vhosts set to Apache when it just hanged, this was fixed adding:
Code:
Acceptfilter http none
Acceptfilter https none
To my httpd.conf, but now it's not hanging, it just times out, I get no error, nothing in the error.log file, it just times out.
After doing a refresh, it comes back like nothing happened.
Are there particular version of mysql or apache that are best suited when you want to utilize a multi-core/multi-cpu system? For example, we are currently using apache 1.3 and am not sure if it has the inherent ability to use multiple cpus to its benefit.
Or would the fact that it spawns child processes take advantage as the processes will be spawned on the different CPUs?
I need to configure multiple Apache Web Servers on redhat server. I have copied and extracted Apache 2.2 into redhat server and extracted but not able to install because I don't understand setting prefix. Please let me explain about prefix configuration and how to set it. At the same time I would like to know is it possible to setup 4 Apache Web Servers on same machine if possible then how to. Can we set up different versions of Apache HTTP Servers?
View 17 Replies View RelatedI'm trying to get multiple GUN groups to have access to certain directories.
Access to transport, delete
When you go to access transport or delete you get prompted for user name and password when entered you get in. However when I add GUNtest10 in the required ldap group. It doesnt accept your user name and password.
######ACL Directives######
LoadModule ldap_module modules/mod_ldap.so
LoadModule authnz_ldap_module modules/mod_authnz_ldap.so
<Directory /bills/>
AuthType Basic
AuthBasicProvider ldap
AuthName "1Login with your Windows ID and password"
[Code]....
In my old server (VPS) I had my dir structure as:
sitea.com was pointing to
/home/me/public_html/sitea
siteb.com was pointing to
/home/me/public_html/siteb
On my new server, I have root permissions and used Plesk to create two domains sitea.com and siteb.com . Now plesk asks for a user to be created for each domain, so created usera and userb for sitea and siteb respectively.
Now as root on my server I created the dir /home/me/ and untarred the whole backup from old server to new server and I have dirs :
/home/me/public_html/sitea and /home/me/public_html/siteb
In apache configuration in file:
httpd.include_sitea under
dir /var/www/vhosts/sitea.com/conf/httpd.include
I changed the document root to /home/me/public_html/sitea
So I thought I am all set. But it does not work. When I try sitea.com in the browser it works but for all subdirs, for eg sitea.com/images it says Access denied.
This is because the dirs I created are owned by root.
So the owner of /home/me/public_html/sitea needs to be usera for this to work ?
Again for siteb to work I need to change the owner of /home/me/public_html/siteb to userb ?
This will be painful ? At least I should be able to change the owner of /home/me/public_html/ to one owner and ensure all sites under that work fine. How do I do that ?
I think the problem is clear by now. Its that I want all my sites to work off from .../public_html/ sub dirs.
