夏期劇場's questions

In my Apache, I have a long list of IPs to be blocked (for some reasons). And then I'm currently putting those rules inside the .htaccess files of the each and every single docroot (domains) I have.

The rules look like this:

SetEnvIF X-Forwarded-For "1.2.3.*" DenyIP
SetEnvIF X-Forwarded-For "100.200.*.*" DenyIP
..
..
..
..
..
..
Order Deny,Allow
Deny from env=DenyIP

These same .htaccess files are now located like:

/var/www/html/www.site-a.com/.htaccess
/var/www/html/www.site-b.com/.htaccess
/var/www/html/www.site-c.com/.htaccess
/var/www/html/www.site-d.com/.htaccess
/var/www/html/www.site-e.com/.htaccess
/var/www/html/www.site-f.com/.htaccess
/var/www/html/www.site-g.com/.htaccess
..

It works that way. But then whenever I have to update the IPs, I have to edit in all of these files.

Question:

How do I apply these common rules Server-wide (Apache-wide), so that they are applied on every single websites docroots (domains) I have in my Apache?

I have a collectively very long list of black listed IPs (for different reasons). Something like this (just an example):

SetEnvIF X-Forwarded-For "1.2.100.*" DenyIP
SetEnvIF X-Forwarded-For "2.3.4.*" DenyIP
SetEnvIF X-Forwarded-For "123.123.*.*" DenyIP
..
..
..
..
..
SetEnvIF X-Forwarded-For "200.200.*.*" DenyIP

Order Deny,Allow
Deny from env=DenyIP

It is working, but about 200 lines long.

Question

How do I put all these lines (SetEnvIF X-Forwarded-For "<ip>" DenyIP) inside a TXT file, and then load it inside .htaccess so that I could do the same Deny from env=DenyIP again, please?

Note: For some reasons, I need it for the .htaccess approach. Not able to put inside VirtualHost Config settings.

Thank you all in advance.

Just like AWS EC2 style, how do i setup the way people SSH into my Server to be via the Server's Private Key (PEM?), instead of we uploading Clients Public Keys into Server?

Idea is to share the Server's Private Key (.pem?) to the Clients, so that whoever has that key would be straight away able to SSH into Server. So that there would be no overheads for me (as Server Admin) whenever there are new Client Users/Machines appeared.

That would also mean, as long as we have that Key, we can connect to Server from "any" Client Machines even it is a new one. (That also would probably allow us to go for the vacation without carrying the Particular Laptops, as long as we have the KEY and a random Computer. :D)

How to achieve that (AWS-like) method please?

I'm using "GlusterFS" Client, to mount the GlusterFS Volume on my Web Server. Below is the MOUNT command when i manually mount from commandline:

# mount -t glusterfs -o aux-gfid-mount gluster1:/gv0 /var/www/html

I don't know how to put that -o aux-gfid-mount option inside the /etc/fstab. So my fstab is still, lacking of that option:

gluster1:/gv0 /var/www/html/ glusterfs defaults,_netdev,fetch-attempts=5 0 0

How do i put that -o aux-gfid-mount option inside the fstab please?

In my Linux, i have httpd (Apache2), and i was curious about it's version released date. So i did:

# httpd -v

Server version: Apache/2.2.15 (Unix)
Server built:   Dec 15 2015 15:50:14

But when i google about "Apache 2.2.15 Released Date", according to Apache official version announcement archive, it was released since 06 Mar 2010 which is terribly OLD.

Ref:

So please let me understand which DATE is my Apache actually updated to?

Thank yo all! :)

The situation is quite typical. I'm using CentOS + Apache(s) behind a Load-balancer for the WebSites.

At this point, let's assume i don't have any access to the Load-balancer (or) let's forget about that one.

Currently i'm using .htaccess to BLOCK the IPs because the native iptables is not able to block the X-Forwarded-For kind of IP addresses (as far as i understand).

My current approach works anyway. But this still allows my Apache(s) to be hit, at least.

Is there a firewall (like: iptables) which can block the X-Forwarded-For IP/connections as well?

Thanks all in advance :)

My question may be a bit confusing, and sorry for that. Here is what i need:

Well, i have (let's say) 3 Websites hosted on one Apache (which are defined inside each VirtualHost settings. Then i currently log their Access Logs separately as in:

• /var/log/httpd/website-a.com.access_log
• /var/log/httpd/website-b.com.access_log
• /var/log/httpd/website-c.com.access_log

So it is all fine. But at the same time, i ALSO want them (all the access logs) inside the Apache's default localhost log file (i'm not sure how to call that one). May be:

• /var/log/httpd/access_log (I think this is default one)

In other words, is there a way to (how to) keep the Access Logs in the separated manner, and at the same time, keep all in the same combined file also?

So this will allow me to see each of the Access Logs in their separated files. And at the same time, i can see all the logs in one place also IF I NEED TO.

I need to understand about something in NFS (Server). Let say i have:

• Two Different Clouds (One Server on each)
• One NFS Server
• One Apache Server (as the NFS Client) with a NFS Mounted Directory (lets say at: /var/www/example.com/downloads/)
• That will become something like: http://www.example.com/downloads/myvideo.mp4
• Lets say: myvideo.mp4 is 1Gb in size.
• 3 different users (3 downloads via Website = 3Gb via Web)

Ok then here is the question for, when a user downloads that file:

• Of course i believe the Apache Server has to use Total 3Gb of Outbound Bandwidth.
• Then again at the same time, has the NFS Server to use Total 3Gb of Outbound Bandwidth also?

Because, what i'm currently understanding is like:
NFS Server---(Outbound: ? Gb)--->Apache Server---(Outbound: 3 Gb)--->Browser

Hoping for the expert's advices since i have to understand clearly and then calculate b/w costs.

Thanks all in advance.

I have nginx as a LB. And the 2 Apaches as the web servers. Lets say i got different domains:

• www.example.com
• checkout.example.com

Both domains will be in same 2 Apache Servers. But ofcoz under the different directories. And with the different VHost files on the Apache vhost file.

As something like below design:

          Nginx
            |
      -------------
      |           |
   Apache       Apache

Below is my current existing Nginx .conf file which is not working for the second Domain (checkout.example.com).

From NGINX (mysites.conf):

upstream serverpool {
  server 1.2.3.101:80 weight=1;
  server 1.2.3.102:80 weight=1;
}

server {
  listen 80;
  server_name www.example.com checkout.example.com;
  location / {
    proxy_pass http://serverpool;
  }
}

From both of 2 Apache Servers' same Vhost Files (httpd.conf):

<VirtualHost *:80>
   ServerName www.example.com
   DocumentRoot /var/www/html/www.example.com/
</VirtualHost>
<VirtualHost *:80>
   ServerName checkout.example.com
   DocumentRoot /var/www/html/checkout.example.com/
</VirtualHost>

But whenever i browse that (http://checkout.example.com), the Domain still comes up in browser .. but with the contents of (www.example.com), which is totally wrong.

What did i do wrong please?

I'm still kinda novice but now i'm on AWS/EC2 Redhat Machines. I have:

• 1x Load Balancer (Above)
• 2x EC2 Apache Web Servers (Below)

Something like:

    LB
    |
 -------
 |     |
Web   Web

Then now if i going to put the 1x Varnish infront of 2x Web Servers:

• What will the placements be?

Please suggest, and will it be like:

    LB
    |
   [V]
 -------
 |     |
Web   Web

Is it?
So here again my questions are:

• If yes, the original LB is not very useful already (i suppose). Will Varnish be the Load-balancer again here?
• Then, does it mean, I do NOT NEED default EC2 Load-Balancer anymore?

Please guide me the proper placement and share the logic.
Thank you so.

Suddenly one of my RHEL 6 Server [VM] (serving Apache on it also) went down and when i rebooted it, i saw several of following /var/log/messages as latest before it went down:

May  4 08:16:37 localhost xinetd[1928]: START: nrpe pid=19307 from=10.xx.xx.xx
May  4 08:16:37 localhost xinetd[1928]: EXIT: nrpe status=0 pid=19307 duration=0(sec)
May  4 08:17:08 localhost xinetd[1928]: START: nrpe pid=19363 from=10.xx.xx.xx
May  4 08:17:08 localhost xinetd[1928]: EXIT: nrpe status=0 pid=19363 duration=0(sec)
May  4 08:18:38 localhost xinetd[1928]: START: nrpe pid=19417 from=10.xx.xx.xx
May  4 08:18:38 localhost xinetd[1928]: EXIT: nrpe status=0 pid=19417 duration=0(sec)
May  4 08:18:40 localhost xinetd[1928]: START: nrpe pid=19420 from=10.xx.xx.xx
May  4 08:18:42 localhost xinetd[1928]: EXIT: nrpe status=0 pid=19420 duration=2(sec)
May  4 08:21:37 localhost xinetd[1928]: START: nrpe pid=19574 from=10.xx.xx.xx
May  4 08:21:37 localhost xinetd[1928]: EXIT: nrpe status=0 pid=19574 duration=0(sec)
May  4 08:21:38 localhost xinetd[1928]: START: nrpe pid=19577 from=10.xx.xx.xx
May  4 08:21:38 localhost xinetd[1928]: EXIT: nrpe status=0 pid=19577 duration=0(sec)
May  4 08:21:38 localhost xinetd[1928]: START: nrpe pid=19580 from=10.xx.xx.xx
May  4 08:21:38 localhost xinetd[1928]: START: nrpe pid=19581 from=10.xx.xx.xx
May  4 08:21:38 localhost xinetd[1928]: EXIT: nrpe status=0 pid=19580 duration=0(sec)
May  4 08:21:38 localhost xinetd[1928]: EXIT: nrpe status=0 pid=19581 duration=0(sec)
May  4 08:24:31 localhost xinetd[1928]: START: nrpe pid=19701 from=10.xx.xx.xx
May  4 08:24:31 localhost xinetd[1928]: EXIT: nrpe status=0 pid=19701 duration=0(sec)
May  4 08:24:49 localhost xinetd[1928]: START: nrpe pid=19713 from=10.xx.xx.xx
May  4 08:24:49 localhost xinetd[1928]: EXIT: nrpe status=0 pid=19713 duration=0(sec)
May  4 08:25:17 localhost xinetd[1928]: START: nrpe pid=19745 from=10.xx.xx.xx
May  4 08:25:17 localhost xinetd[1928]: EXIT: nrpe status=0 pid=19745 duration=0(sec)

• What does it mean please?

I'm on RHEL. How can i extend the log interval of SAR Reports to be more then a month? Normally the daily SAR reports are something like:

[root@server /]# cat /var/log/sa/sa
sa01      sa21      sa23      sa25      sa27      sa29      sa31      sar21     sar23     sar25     sar27     sar29     sar31     
sa20.bz2  sa22      sa24      sa26      sa28      sa30      sar20     sar22     sar24     sar26     sar28     sar30 

It means it stores daily reports for one month only.

• How do i store (for e.g) 3 months?

In RHEL, instead of using service network restart command, how can i restart a particular network interface, lets say "eth1", with only one command.

"Only one command" because that is the only interface where my ssh is working on also. So if i'm about to use: ifdown and then ifup, i will never be able to hit the ifup command as my ssh has been terminated once after ifdown eth1 command.

So there should be a single command which allows me to altogether bring down and then bring up the interface which is serving my current ssh connection. So i do not need to worry about connection totally lost to my server.

Any idea please?

From .htacces file, how to pass/redirect the FULL URL to another URL as GET Variable?

Like:

http://www.test.com/foo/bar.asp

will be redirected to:

http://www.newsite.com/?url=http://www.test.com/foo/bar.asp

With Full Url with Domain.
I tried:

RewriteEngine on
RedirectMatch 301 ^/(.*)\.asp http://www.newsite.com/?url=%{REQUEST_URI}

But it is going out like:

http://www.newsite.com/?url=?%{REQUEST_URI}

Actually i am trying to start my Kaltura service. The default "sphinx" came along together with Kaltura has crushed and then got uninstalled. Then i installed clean dedicated of "Sphinx" (source version from here) outside of Kaltura. I setup at /opt/sphinx/. Then when after setup .. i run:

~ # sudo /opt/kaltura/bin/sphinx/searchd

It is showing like:

Sphinx 2.0.2-dev (r2823)
Copyright (c) 2001-2011, Andrew Aksyonoff
Copyright (c) 2008-2011, Sphinx Technologies Inc (http://sphinxsearch.com)

using config file '/usr/local/etc/sphinx.conf'...
listening on all interfaces, port=9312
listening on all interfaces, port=9306
precaching index 'test1'
WARNING: index 'test1': preload: failed to open /var/data/test1.sph: No such file or directory; NOT SERVING
precaching index 'test1stemmed'
WARNING: index 'test1stemmed': preload: failed to open /var/data/test1stemmed.sph: No such file or directory; NOT SERVING
WARNING: multiple addresses found for 'localhost', using the first one (ip=127.0.0.1)
precaching index 'rt'
WARNING: last message repeated 1 times
WARNING: index 'rt': preload: failed to open /var/data/rt.lock: No such file or directory; NOT SERVING
precached 3 indexes in 0.000 sec

What are these means? I'm not really getting why.

Is it possible for single-instance of Solr to use across multiple Applications, Cross Domains? Like multiple CMSs such as Wordpress + Joomla + Drupal. Any searching from anywhere should be showing the Results from all of connected applications.

For example:
If i search a word marketing from the Wordpress Site, it should output the results from Wordpress itself and also the contents from the Joomla and Drupal sites.

Is it possible?