Apache :: HTTP Over SSL System - Authentication Authority
Oct 6, 2013
I have to create a structure in which there is a client, a server and an authentication authority.
The authentication authority verifies the identity of both the client and server before they can communicate, so that the client can access the content offered by the server. Everything must be made using HTTP with SSL (HTTPS).
For now I have installed xampp on my pc with ubuntu, I performed the initial configuration and was able to view a test page locally hosted by entering the URL of the virtual server "www.server.it" (added to the configuration of apache2).
Recently I cross compiled Apache2.4.6 to run in my ARM cpu, and I found that http can work well if only it's own port such as 80 is listened, but if add one more port to be listened in conf file, http fails. That means http is good, but after https is on, http stop working and https is good. At the same time, I built the same apache2.4.6 in X86 gcc4.6.1, http and https can both work well. In addition, I also tried apache1.3.33 in the same ARM system, http and https can work well too. It seems like that in apache2.4.6, http can't work well if there is one more port listened in ARM system. Actually http port is listening and is established when using netstat to monitor it.
root@freescale /usr/local/apache/bin$ netstat -a Active Internet connections (servers and established) Proto Recv-Q Send-Q Local Address Foreign Address State tcp 0 0 :::www :::* LISTEN tcp 0 0 :::443 :::* LISTEN tcp 0 0 (null):www (null):3705 ESTABLISHED tcp 0 0 (null):443 (null):3720 ESTABLISHED
On our production service, we've been getting numerous malformed POST requests to some of our CGI scripts that are showing up as 500 errors in our logs. They are malformed in the sense that the actual content length doesn't match the Content-Length specified in the request.
Here's the most trivial example I can come up with that reproduces the problem for us:
In addition to the 500 error in the access log, we see the corresponding error in the error log:
(70014)End of file found: Error reading request entity data
Based on the nature of the POST request and the error response, it does appear that Apache is doing the right thing here.
The POST never actually makes it as far as the script being targeted (/some_valid_alias in the above example); in other words, Apache returns 500 to the client, writes the error to the error log and never executes the script.
Is there a way to capture/avoid internal Apache errors like 70014, and return some other HTTP status besides 500 (like 403)? It's particularly annoying in our case, because our server sends us an email for all 500 errors.
So far, our best "defense" against these 500 errors is to disallow POST for these aliases, which normally just ignore the POST data anyway (when the request is not malformed):
I have a machine at home running the UniformServer wamp package on Win2K3 Standard server and I'm hosting a simple family picture gallery using Gallery 2.3. I put the gallery in a subdirectory of root so I could have a public homepage and a gallery that is password protected. I did this using .htaccess and .htpasswd.
My issue is when I browse to mydomain.com and click the link to the gallery [url], the password box pops up, and when I enter the appropriate credentials, the domain name reverts to the IP and another password box pops up. If I log in a second time, I'm fine. If I cancel that second login, it still lets me in, but prompts me to login that second time for every click going forward until I authenticate.
I have been trying to set ssl client authentication with Apache.I basically have a server certificate issued by a recognized CA. For the normal ssl authentication I use the following configuration (and it works fine):
I have a couple of directories on my server that require authentication (MySQL DBD and AuthUserFile). Both work fine with SSL off. When I use SSL on directories without authentication that also works fine. However, when I put the 2 together, authentication is by-passed. I cannot seem to get the configuration right to do both. Here is my VH conf file (sanitized):
I'd appreciate if you can point out some links, or if anyone is willing to help me for a fee, we can talk about it.
I have a web application called MyApp
- Each MyApp user has 5gb hosting, and a web interface to manage their files.
- Therefore MyApp user gets a user account on linux machine and has access to only one directory /repo/usr/<user_id> and nowhere else. (suexec?)
- If that MyApp user creates a subdomain from any folder inside his home folder (he can do that using web interface), that folder is readable by www-data user not writable.
- That myapp user is able to mount/unmount his own ftp drives using curlftpfs.
- In short, users can only mess with their own files and they have no access nor rights to any other file that is outside of their home dir.
In short, this is a kind of hosting company server setup. Right now, we will have to manage all this from -only- one powerful server.
Apache Suexec, so that each VHosts runs under there own username FTP for each of the vhosts.
I have made a username aplushost and FTP works fine when i login, however when i try and get Suexec to work it shows a 403 permision dined, even know the whole directroy path is with correct permsions.
"/home/aplushost/www"
However if i chown the directroy "aplushost" to apaches username , currently "nobody" i have tried with "apache" and many others the page is displayed correctly.
The weird thing is that the www directroy can still be set to the aplushost username and files work inside.
However due to changing the privalages of the folder aplushost ftp now fails to login due to the folder not being owned by the ftp user "aplushost".
So im stuck between only having one item working at a time.
i have put some content of my config files.
----------httpd.conf vhosts------------------ <VirtualHost 87.117.196.247> DocumentRoot "/home/aplushost/www" ServerName aplushost.co.uk SuexecUserGroup aplushost aplushost <Directory "/home/aplushost/www"> allow from all Options +Indexes </Directory> </VirtualHost>
If a User logs in with username only, all works correct. If a user use the DomainUsername format, the login is rejected with "user not found".Should it not work with both login styles ? Or is there a option to reformat or rewrite the username before authentication without the "Domain part ?
I was wondering if there's a simple way to implement some kind of authencation bruteforce protection in apache for windows? Right now my authentication never stops asking if user inputs the wrong credentials, this makes me think i'm vounrable for bruteforce attacks which could eventually get through if given the time.
In web application we are facing high vulnerability issue based on the session validation.
We can download the files from the server whenever we are passing the link even without login. The links are directly hit into the server and download the files any type of files extension such as .txt, .xml, .zip and so on.
Need the solution for this issue: How will we resolve the issue using validate the session in apache side?
Scenario as below:
When user manually passing the request if user logged they should access the files When user manually passing the request if user is not login they shouldn't access the files
Here both the scenarios they can access the files but we want to restrict when the request is coming to apache without login.
EX: [URL] ....
When i tried above link I can able to view the file in browser. Even able to download all different fies extension which are having in the under tomcat webapps dir.
How we can restrict this in apache code or any other files in apache side or is there any way to validate the request is logged one or not?.
I am setting up apache 2.4 as a service locally through localhost on a windows 2008 R2 standard server. I have set up SSL listening on port 443 and works correctly, however I am having trouble figuring out how to get apache to authenticate my CAC card.
I have downloaded the DOD certs and put them into various types of files including pem, base 64, der etc and I have yet to figure out a way to get the client certificates validated.
I have left out the information about the SSLCertificateChainFile, SSLCACertificateFile and SSLCARevocationPath as I am sure this must be where my problem is.
Is it possible to get this done through localhost. Here is the error I am getting in the error log.
[Wed Oct 29 11:37:05.675491 2014] [ssl:error] [pid xxxx:tid xxx] [client 127.0.0.1:59282] AH02039: Certificate Verification: Error (20): unable to get local issuer certificate
I am using a self created self signed server ceritficate. Here are some details from my httpd-ssl.conf file:
I am running apache currently on my QNAP server, and have enabled webserver and LDAP. We have set up users on LDAP. I have created a landing page for access from the internet. I want to configure Apache to authenticate the users using LDAP before granting access to the landing directory.
I have started this with the apache configuration below: My apache config file -
When I access my page, I get the authentication prompt. But when I enter my LDAP login and password, I get thrown out of the system with the error:
Internal Server Error
The server encountered an internal error or misconfiguration and was unable to complete your request.
Please contact the server administrator, admin@NAS and inform them of the time the error occurred, and anything you might have done that may have caused the error.
More information about this error may be available in the server error log.
Looks like my apache configuration is a problem as I am able access my LDAP and everything with LDAP seems to be working fine except Apache configuration to authenticate against LDAP.
A couple of directories on the website need authentication against LDAP. The setup has been working for many years and all of a sudden is giving some issues.
The page asks for authentication and once login details are entered, either it throws an error "Page has moved" or "Please refresh your browser or reload the page".
As far as I am aware no configuration has been changed for LDAP or for Apache. The website runs on Windows Server 2003.
I am very much new to Apache and cannot seem to find any errors logged.
Im having a problem where local authentication will not work when when the configured LDAP server is unavailble. When the ldap server is online I can authenticate fine against ldap and local file. However, when the ldap server is offline, I cannot authenticate with the user1 account. The browser just sits at a blank screen.
Ive searched a lot on this and found many examples, all very similar to my config below, but I still cannot failback authentication to local file when ldap is unavailable. Im running Apache/2.2.10. I have also tested this on 2.2.16 with the same results.
I have several applications that use authentication and expect REMOTE_USER to be set by Apache for authentication/authorization.
I am putting a reverse proxy, with shibboleth in front of these applications, on a separate server.
Currently, REMOTE_USER is not sent. I have tried a few things, and I am currently sending it inside another header, but I have some applications that are closed sources and this will not work.
Is this possible to do? I am running this with Apache 2.4.7, I believe, on the Windows platform.
I have a class project that we are working on where we have to configure and implement an Apache server with ssl using ldap for authentication. I have documentation of literally everything I have done in the configuration. Everything seems to be fine with the config that I can tell. The client gets a prompt for username and password when they access the server ip address. However, once the correct username and password are entered, then the client receives a 500 internal server error message instead of the webpage: "Internal Server Error...The server encountered an internal error or misconfiguration and was unable to complete your request.
I'm in the process of trying to pin down a couple errors I've been running into, and after installing and configuring SSL I have occasionally been receiving the following error:
Suddenly on a particular website (possibly following an easyapache php minor version update) if I specify a folder rather than a specific file I get a 503 response.
If I do the same on https it's ok. But http always gets a 503.
If I specify a file it's fine, but the folder always hits a 503.
I have removed the htaccess file and php.ini to rule that out and indeed they weren't factors - it seems like it goes wrong before reading the htaccess file.
I keep hearing that redirecting from http to https is not very secure [non-SSL to SSL]. Among other reasons, one reason is that the browser may continue to think it is communicating with non secure server and may not encrypt the data. Is it true? I hope not, I am using the following -
its possible to do a P2V migration of a Apache http server 2.2
Present environment:
Windows 2003 Apache http server 2.0.63
There are 2 webservers (running Apache) for load balancing. The backend server runs an application which uses an oracle database. Is a P2V migration of the web servers possible?
I made the idiotic move of not making a backup of the default configs and now I can't access my http website externally. ssl works fine, but I need http to work too.
