Jeudi 29 juillet 2010 4 29 /07 /Juil /2010 22:03
- Publié dans : Veille virale

I had recently to deal with a compromised computer. It was though supposed to be protected by a up to date antivirus with real policies.

 

The file has been detected using Spybot S&D: ZBot. It's been a long time ago since I did not find a malware which is not a spyware/adware/rogue AV, with Spybot. Anyway...

The computer was also protected by Clam for Windows, the 'in the cloud' version of Clam AV. See: http://www.clamav.net/lang/en/about/win32/ But it did not detect anything however.

Since I like to contribute to OpenSource projects, I thought of submitting this sample that Clam was not supposed to detect. Then I was astonished to see that the online form to submit samples to Clam was complaining the sample was 'already detected' and that I should check my own Clam updates...!

Here is the screenshot:

clam_zbot_support_260710_ann.jpg

 

What's going on with Clam In The Cloud? is it less efficient than the regular ClamAV?  I still do not have any answer...

 

By the way, I was kindda disappointed to see at first that McAfee was not planning to publish an extrat.dat:

mcafee_webimmune_echantillon_Zbot_230710_ann.jpg

 

Fortunately, they did publish an extra.dat later on. Why did the WebImmune portal say 'inconclusive' and 'no Extra', so?

 

To finish about Clam, I'm gonna try to check if there are real differences between the regular ClamAV version and the 'Clam in the cloud, for Windows'.

First, let's see what VT says about the sample:

VT_sdra64.exe_010810.jpg

So, VT already scanned the sample on the 23rd... the day I submitted the sample to McAfee.

And the detections summary array confirms that Clam was supposed to detect the sample:

VT_sdra64.exe_glob_010810.jpg

 

Let me remind you that, as other people like T. Zoller said, VT uses command line versions of AV engines. I know quite well ClamAV, and yes, it is the command line version that is available by default on Linux!

Thus, ClamScan most probably detected the sample, but not the Clam 'in the cloud' version... wheras the 'in the cloud' technology was said to offer better protection in real time... (I even tried to move the sample on the HDD to see if Clam in the could would detect it, but nothing happened).

According to that example, I would say that the 'in the cloud' version may offer better protection in real time, but can also miss samples that the regular version of Clam would just detect...

Par Philippe V.
Ecrire un commentaire - Voir les 0 commentaires
Mardi 20 juillet 2010 2 20 /07 /Juil /2010 21:30
- Publié dans : Veille sécurité

Voici un échantillon tout frais de hameçonnage ciblant SFR.

 

Cher client Sfr.

  
Ce mail vous a ete envoye suite a une faut comptable qui s est produite lors de nos factur mensuel .
En effet le 19  juillet 2010 la somme de (154,00) cent cinquante quatre Euros a ete indument preleve a cause d'un Probelem technique,un reversement en votre faveur sera effectue dans les plus brefs delais,a cet effet nous vous invitons a cliquer sur le lien ci-dessous et vous connectez pour fournir toute information susceptible d accelerer ce restitution .
le versement effectuer sera considere comme valide et aucune reclamation ne sera accepte.
Nous vous remercions de votre comprehension et nous nous excusons pour le desagrement encouru.
plus brefs delais,a cet effet nous vous invitons a cliquer sur le lien ci-dessous et vous connectez pour fournir toute information susceptible d accelerer ce restitution .
le versement effectuer sera considere comme valide et aucune reclamation ne sera accepte.
Nous vous remercions de votre comprehension et nous nous excusons pour le desagrement encouru.
Remplissez le formulaire de remboursement en cliquant sur le lien suivant.
 
  
  
Important :
Le versement effectue par Sfr sera porte sur votre prochain releve bancaire.
Nos clients Sfr beneficieront d un geste commercial.
Nous vous assurons de la confidentialite des informations fournies et Sfr se porte garant quant a la responsabilite juridique de ces transactions
--------------------------------------------------

Voici les entêtes du courriel :

Return-Path: <giosef@dazzler.unbit.it>
Received: from smtp.%serveur%.com ([unix socket]) by 
hermes.assonetworx (Cyrus v2.2.13-Debian-xxxxxx+lenny3) with LMTPA; 
Tue, 20 Jul 2010 12:42:46 +0200
X-Sieve: CMU Sieve 2.2
Received: from sabretooth.unbit.it ([81.174.68.19]) by 
smtp.%serveur%.com with esmtp (Exim 4.69) (envelope-from 
<giosef@dazzler.unbit.it>) id 1ObAHY-0004gq-Ne for 
philippe.vialle@%serveur%.com; Tue, 20 Jul 2010 12:42:46 +0200
Received: from dazzler.unbit.it (unknown [192.168.0.57]) by 
sabretooth.unbit.it (Postfix) with ESMTP id 578C5201D9ED for 
<philippe.vialle@%serveur%.com>; Tue, 20 Jul 2010 12:11:00 +0200 (CEST)
Received: by dazzler.unbit.it (Postfix, from userid 18372) id 
67EE520CFEE43; Tue, 20 Jul 2010 12:13:00 +0200 (CEST)
To: philippe.vialle@%serveur%.com
Subject: probleme technique
MIME-Version: 1.0
Content-type: text/html; charset=iso-8859-1
From: Service Client <Service@allmail.com>
Message-Id: <20100720102348.67EE520CFEE43@dazzler.unbit.it>
Date: Tue, 20 Jul 2010 12:13:00 +0200 (CEST)
               
                    
Au niveau filtrage SpamAssassion :
- DCC positif
- MIME_HTML_ONLY BODY  positif (mais donc score négatif)
- HTML_MESSAGE BODY positif
                  
                    
Quid des détections au niveau du lien inséré dans le courriel :
- Google Chrome : signalement de "site de phising"
- Safari : blocage partiel (en faisant F5, l'avertissement peut apparaître, mais pas à tous les coups...)
- Internet Explorer 8 : aucune alerte
- Mozilla Firefox et Shiretoko : avertissement de "site contrefait"
- Opera 10 : avertissement de "site frauduleux"
- barre d'outil Netcraft : avertissement classique
- McAfee trustedsource.org : site catégorisé "phishing", risque maximal
- CISCO Surfcontrol (mtas.surcontrol.com) : ne fonctionne pas :(
- Fortiguard : classifié en "information technology"...
- DrWeb URL analysis tool : URL injoignable, test impossible
- Finjan URL test : URL injoignable...
                     
                       
Quid des informations sur l'émetteur du courriel ?
Apparemment cela tourne autour de dazzler.unbit.it. Il semblerait qu'une machine interne soit compromise (192.168.0.57). L'adresse IP en résolution est : 81.174.68.57.
- IronPort Senderbase : très bonne réputation (cf. http://www.senderbase.org/senderbase_queries/detailip?search_string=81.174.68.57 )- 
            
               
A propos de l'adresse IP elle même : https://dns.l4x.org/81.174.68.57
Hébergement mutualisé : http://www.w3who.com/reverse-ip/81.174.68.57  Ceci aura probablement facilité la compromission du serveur.
              
Méfiance donc ! il est probable que les CERT assistant SFR sont déjà sur le pont...

Par Philippe V.
Ecrire un commentaire - Voir les 1 commentaires
Dimanche 11 juillet 2010 7 11 /07 /Juil /2010 21:54
- Publié dans : Bidouille informatique

Just for once (again?), I'm gonna deal with linux / Unix system administration., and in particular, migration.


This may be related to scurity, as you may say that integrity does concern system configuration, when it's about renewing a server or transfering a whole system configuration from a machine to another.

 

First of all, let's say that you have an old server, and you want to migrate it to a new one, with new hardware. If you create an image of the whole system and restores it on the new server, it may just not work because of hardware differences and/or special configuration elements.

 

To insure a transfert with integroty check, I'll suggest you to use RSync. Here is what could the command line look like:

rsync -avchz --bwlimit=150 --stats /etc remote_host:/sync_etc

The way of synchronizing is source - > destination (last argument).

The bwlimit limits the bandwidth, so that you would not pay extra money for this (huge) transfer, depending on your hosting contract.

 

The folders that you would at least need to synchronize between the two servers are:

- /etc

- /var

 

Special thanks: Roro, Duckx, Fredy, and Lucky ;)  the guys who helped me meeting Linux ;)

Par Philippe V.
Ecrire un commentaire - Voir les 0 commentaires
Jeudi 8 juillet 2010 4 08 /07 /Juil /2010 23:58
- Publié dans : Veille sécurité

Most of the time I would say and repeat: people should review their security systems and probably harden their configurations and system settings.

 

This time, my post is gonna deal with the contrary: how a hardened URL filtering system can impact productivity and vital assets.

You may read that security vendors tend to add a new way to block malicious / suspect content: security reputation.

I will not discuss the idea by itself right now. Let's just say that to protect from compromised systems, it could be a really interesting concept.

But what happens if the security reputation is wrong? That kind of measure is somewhat a score, which is most probably dynamic.

The firms that are being protected by that security reputation filtering system do not see the reputation score of their online applications and needed websites (assets).

Because the score is dynamic, it can be different today from yesterday.

Here is a part of a real story: it is about a vital portal, I mean a portal related to a vital function of a state (counter terrorism point of view, such as energy, communications, transportation, medical emergencies...).

One day that portal had its security reputation to be reviewed and reach "high security risk". In that case, most of the systems working with "security reputation filtering" automatically blocked any access to that website...

This was quite serious... We are not talking about Facebook being unreachable, not even Google... it is about a crucial web portal, government watched and related for a service that helps the country to run.

Were the firms informed before the website was automatically blocked? not at all, since it is an automatic check and update.

Could the firms easily bypass that filter? not really, if you consider that proxies are vital equipments and may not have their configuration radically changed so quickly, in emergency mode, and in production...

Could the firms monitor all the virtal websites/portals they know they need, to prevent that kind of situation? well if you consider 100 000 computers, several millions requests a day, and several thousands websites being whitelisted...  certainly not!

 

So I really warn any people that use or plan to use a technology close to security reputation score with automatic blocking. You may have real situations, with oproductivity loss and above all collateral damage...

Par Philippe V.
Ecrire un commentaire - Voir les 0 commentaires
Lundi 7 juin 2010 1 07 /06 /Juin /2010 15:51
- Publié dans : Veille virale

Just after the Adobe and French CERTA advisories, I wanted to talk a lil bit about the website that is said to host the Adobe 0day.


Here is Adobe's advisory:

http://www.adobe.com/support/security/advisories/apsa10-01.html


According to Symantec (see: http://www.symantec.com/business/security_response/writeup.jsp?docid=2010-060601-3020-99, the suscpicious website is:

google-analytics. dynalias.org.


It quite clearly seems to be a fake Google Analytics portal. Not sure that it does steal user's credentials anyway...


Please note that :

- Netcraft did not warn about it (at the time of writing)

- IronPort does not detect it

- Secure Computing (trustedsource.org) does not detect it

- Firefox 3.6 does not tell anything

- internet Explorer 8 neither

and a very few AV vendors are said to be able to detect the PDF...


Here is what the website looks like:


capture_googleAnalytics_dynalias_070610.jpg


The IP address 180.149.252.136 is apparently located in Hong Kong... see:

http://www.robtex.com/dns/google-analytics.dynalias.org.html

and blacklisted at least once!


So I strongly recommend to remain prudent with that domain.




Par Philippe V.
Ecrire un commentaire - Voir les 0 commentaires
Vendredi 4 juin 2010 5 04 /06 /Juin /2010 20:56
- Publié dans : Veille virale

Well this is not the first one, but at least I find it relevant since it is not being detected by (almost) any AV engines - I mean command line versions on VirusTotal.

 

Here is what the MSN message looks like:

 

msg_MSN_040610.jpeg

 

 

I clicked on the link aztec-casino.uk... the browser popped up and offered me to download a file named installcasino.exe.

 

Unfortunately for the bad guys, a BSD derivative kernel is kindda immune to Win 32 PE files... :)

 

According to VirusTotal, only 1 engine out of 41 detects the sample:

http://www.virustotal.com/fr/analisis/09b2b1ea08b19ff9a4e1c3609a39fa8f6ae60b8827260e8dd034630af754343c-1275677556

 

The only detection is an heuristic one. Please keep in mind that VirusTotal uses command line versions of AV engines, and this may reduce heuristic features or particular content dynamic analysis.

 

I'm waiting for an online sandbox analysis results.

 


What about URL filtering? not better either:

- nothing for McAfee TrustedSource:

http://www.trustedsource.org/en/feedback/query?sid=&p=&q=www.aztec-casino.uk.mn

- nothing for IronPort / SurfControl:

http://mtas.surfcontrol.com/MTASResults.asp  (says 'not in our list' at the time of writting).

 


What about domain informations?

- a bit weird according to Netcraft: UK or De?

http://toolbar.netcraft.com/site_report?url=http://www.aztec-casino.uk.mn

- brazilian IP address according to DomainCrawler? no WhoIs information...

http://www.domaincrawler.com/domains/view/www.aztec-casino.uk.mn

 

 

What about DNS?

> server 208.67.222.222
Default server: 208.67.222.222
Address: 208.67.222.222#53

 

> www.aztec-casino.uk.mn
Server:         208.67.222.222
Address:        208.67.222.222#53

Non-authoritative answer:

Name:   www.aztec-casino.uk.mn
Address: 188.40.70.45

OpenDNS and my ISP do agree about the IP resolution, therefore that should be correct.

 

Then I guess RIPE will be a pretty reliable about geo-localization:

http://www.db.ripe.net/whois?form_type=simple&full_query_string=&searchtext=+188.40.70.45&do_search=Search

and (this winner is): Germany.

 

 

 

 

 

Par Philippe V.
Ecrire un commentaire - Voir les 1 commentaires
Vendredi 30 avril 2010 5 30 /04 /Avr /2010 00:12
- Publié dans : Veille virale

I'm not gonna deal here with the real McAfee DAT 5958 issue by itself. What I find interesting is what's coming around this incident.

Some other AV vendors warned that users attempts t download the DAT that fixes the 5958 failure may be used to infect their computers.

I was honestly thinking about fake DAT packages: malware linked to a real exefile DAT from McAfee, or even just malwares called 'SDAT5959'...

But what I discovered is actually worse, to me.

Consider the following Google request: download mcafee DAT 5959. Quite natural and obvious, isn't it?

Here is the URL of the 5th page

 http://www.google.fr/search?hl=fr&safe=off&q=download+mcafee+DAT+5959&start=40&sa=N

 

But this is where the danger shows up. Let's have a look at the first link of the Google's page of results:

RQ_Google_tolstiy.co.cc_290410.jpg

 

The website domain name is: tolstiy.co.cc.

The Google preview of its content even includes the Google logo.. This should not be dangerous, right?

You may notice all the relevant keywords the website may need to be well indexed and appear at a good place in Google's results: 'sdat 5959 free download license mcafee superdat failure SDAT5959 EM. exe mcafee8.5i, McAfee®: 5960 Update '

This could help SEO hijacking (or poisonning) for sure!

Nonetheless the point is that the user will immediately be redirected to another website: endroiturlredirect.com

exploit_alertAV_endroiturlredirect.com_290410.jpg

 

Then the malicious part shows up. This pages hosts an exploit!

Avira prompted then a warning:

MSG_Avira_endroiturlredirect.com_290410-copie-1.jpg

 

A PDF exploit for a DAT update rescue... that's probably funny (or weird).

Therefore I strongly recommend to any users and admins to really pay attention to where they download updates (including antivirus ones), at any time, especially in case of emergencies.

 

More about the website: http://tolstiy.co.cc/

You have to pay attention to notice something quite strange.

I said that the thumbnail of it seems to include a Google's logo... yes but guess what, the Google logo, buttons and request bar are all a simple picture in fact!

And here is the URL of it: http://www.webopedia.com/quick_ref/img/google_screen001.jpg

And what about the WhoIs of it? http://www.domaincrawler.com/domains/view/tolstiy.co.cc

Hey, more interesting: the IP address seems to be a Brazilian one, and the rest of the WhoIS info appears to be protected by an anonymization system. Quite obscure, but a kindda habit in VX methods.


But what if I use Internet Explorer 8?

Surprinsingly (or not?) the page redirected me to: malware-checker-free.com. And I.E. 8 screamed about a phishing risk while accessing this website.

Here is a screenshot of what I saw:

MSG_IE8_malware-checker-free.com_290410.jpg


And what about other browsers? (PoC: Win 7, 64 bits, full patched).

- Safari (last version) did not alert me in anayway

- Firefox 32 & 64 bits (last versions): no alert

- Opera 10 (last version): no alert.

- Chrome (last version) : no alert.


So here is what an user could see if he does not use I.E.:

controlpanel_malware-checker-free.com_300410.jpg


And the (funny but) annoying part of it is an endless loop behind this popup:

msg_malware-checker-free.com_290410-copie-1.jpg


Whatever an user will do ('cancel', or 'OK'), the popup will come back, and furthermore will try to download an exefile on the computer.

This file is called: 'win_protection_update.exe'

Here is the VT results for it (let me remind that VT is a list of command line AV scanners, not the realtime protection they could offer in a regular installation):

http://www.virustotal.com/fr/analisis/415af935f5ce82f68f15ad133af4542813e34c40a7c5825fed9cdf0d2a46d304-1272584528

Ok so that's 20 out of 41 engines, not bad.


About the malicious URL by itself: http://malware -checker -free.com/secure1/?id=ololo


If you try to access it with only the FQDN, let's say malware-checker -free.com  you may be redirected to... Google. A bit funny.

But if you try to change subdolder and/or page, such like: http://malware -checker-free.com/test   here is what shows up: 

Apache/2.2.3 (CentOS) Server at malware-checker-free.com Port 80

 

Either the bad guys forgot to update (and secure) their web server, or they hacked a third party one to host their malicious page and files...



Last, if you look at the source code of the webpage http://malware -checker-free.com/secure1/?id=ololo   (thanks to Opera!), you may have an idea about how the bad guys tried to obfuscate their source code:


<!--

Page protected by ionCube - HTML/JavaScript Encoder

Copyright (c) 2003 RWJD.Com and ionCube Ltd.  All Rights Reserved.

Any analysis of this  source code,  embedded data  or file by any means and by any entity whether human or otherwise  to including but without  limitation to discover details  of internal operation, to  reverse  engineer, to  de-compile object code, or to modify  for the purposes  of modifying behavior or scope of their usage is forbidden.

-->

 

To finish with, a Google request will suggest that ionCube is a proprietary solution to "protect and license" the PHP pages... well, I'm not sure the bad guys did pay for the ionCube license (just guessing).



 


Par Philippe V.
Ecrire un commentaire - Voir les 0 commentaires
Mardi 20 avril 2010 2 20 /04 /Avr /2010 00:55
- Publié dans : Veille virale

Une fois n'est pas coutume, je n'étais même pas en train de faire de la veille, que l'un de mes comptes MSN a reçu en rafale deux messages, de deux contacts différents.

Evidemment, les contacts en question étaient censés être "hors ligne" au moment de l'envoi, et (bizarre vous avez dit bizarre... ?) il m'envoient le même message.

Voici un échantillon :

msg2_MSN_180410_annonym.jpg


On notera qu'il s'agit apparemment d'un lien pointant sur une photo... un procédé qui n'est pas nouveau.

Téméraire que je suis dans mes analyses virales, je m'empresse de cliquer sur le lien.

Tiens donc, à la place d'une photo, c'est un fichier ".src" qui arrive sur mon disque. Ho les vieux temps de la virologie sont de retour...

Cependant, cette fois-ci l'antivirus bippe directement (NOD32 ,et Antivir)

Mieux que cela, une analyse comparative sur VT donne un résultat encourageant : presque 1 moteur sur 2 détecte (en mode ligne de commande) le fichier :

http://www.virustotal.com/fr/analisis/5a38a64a407f6093f8fe3ce737fd9ebe35835e0a0eac42a301e6c86c4def7850-1271593655

Alors que dire, pour les mauvaises langues : est-ce les antivirus qui sont tous à l'heure cette fois-ci, ou la menace virale qui est obsolète ?...

Par Philippe V.
Ecrire un commentaire - Voir les 1 commentaires
Lundi 12 avril 2010 1 12 /04 /Avr /2010 21:46
- Publié dans : Veille virale

As you probably read on te web, at the same time people welcome new URL services such as goo.gl, others warn about new threats that come with them: bypassing URL filters using URL shorteners...


Here is a kindda new sample. Once again, over the MSN Network. 

message_msn_tinyurl_120410_annon.jpg

If you click on te URL, the tinyURL system will automatically redirect you to: http://www.camstranger.com/


Once again, I strongly warn any people to click on that link, unless you know what you do.

The website seems to be a kindda public chat, with webcam. AFAIK there is no viral component on it. But I would say anyway that this looks strange.

capture_camstranger.com_120410.jpg


My guess: some guys rented a BotNet and/or a stolen MSN accounts database to send a massive communication in order to promote that new public chat... 

I'll try to check that out later on :)

Par Philippe V.
Ecrire un commentaire - Voir les 0 commentaires
Mercredi 7 avril 2010 3 07 /04 /Avr /2010 22:48
- Publié dans : Veille sécurité

It's been a quite long time since I wanted to write this article. But taking into account the fact that I spent 3 hours at night to understand what was happening to my BIOS, I could not forget it, I guess...

How did all of that started? very simple. I thought of applying some of the best practices in laptop security: (BIOS) password at startup.

Very well, I entered the BIOS. I had a few difficulties, because the 'Care' button did not work that well, and the boot-splash did hide the key to press to enter the BIOS. Anyway, I'm more patient than that.

Once I had got into the BIOS, I went to the security tab. Setting a password up seemed to be quite simple, as usual.

Then, I started my laptop, like in a regular way. As usual, it remained on for a few days, without reboot. And obviously, the problem came out at the next reboot. Thanks Windows Update (automatic / forced) reboot at night, for that...

Not even scared, I saw the next evening the bootstrap stucked at the password check. So I entered my so said password. Yikes, it seemed to compute a lil bit, and then displayed a warning telling me that my password was wrong. I tried a second time, a third one... then forced reboot....

This time, I started to feel less at ease... 

My password was 7 letters long, with 2 more digits. I imagined the problem could be a keymap issue (Qwerty in the Bios, else Azerty). So I built all the combinations I could imagine: typos, and keymaps issues... 

3 hours later, I was still in front of a locked computer. Then, fortunately, a bit of 'password hardening experience' came to my mind: what if the Bios could not register my whole password?

After 5 minutes, bingo! Only the first 6 characters had been saved! 

But no warning told me that only 6 out of 9 characters were going to be saved... I find it quite abnormal and tricky!

I hope this will help other people, at night, in a rush... like I was...

Par Philippe V.
Ecrire un commentaire - Voir les 0 commentaires

Présentation

Recherche

Syndication

  • Flux RSS des articles

Créer un Blog

Créer un blog gratuit sur over-blog.com - Contact - C.G.U. - Rémunération en droits d'auteur - Signaler un abus - Articles les plus commentés