Stories
Slash Boxes
Comments

Dev.SN ♥ developers

posted by janrinok on Tuesday March 18 2014, @09:36PM   Printer-friendly

ancientt writes:

"'Researchers have documented an ongoing criminal operation infecting more than 10,000 Unix and Linux servers with malware that sends spam and redirects end users to malicious Web pages' as reported at Arstechnica, in an attack campaign being called Windigo.

It has been going on since 2011 and successfully hit Linux Foundation kernel.org servers and developers of cPanel. 'During its 36-month run, Windigo has compromised more than 25,000 servers with robust malware that sends more than 35 million spam messages a day and exposes Windows-based Web visitors to drive-by malware attacks."

A detailed PDF writeup is available from We Live Security.'

This discussion has been archived. No new comments can be posted.
Display Options Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: -1) by Anonymous Coward on Tuesday March 18 2014, @10:02PM

    by Anonymous Coward on Tuesday March 18 2014, @10:02PM (#18354)

    whois welivesecurity.com

    Domain Name: WELIVESECURITY.COM
          Registrar: CSL COMPUTER SERVICE LANGENBACH GMBH D/B/A JOKER.COM
          Whois Server: whois.joker.com
          Referral URL: http://www.joker.com/ [joker.com]
          Name Server: NS1.P06.DYNECT.NET
          Name Server: NS2.P06.DYNECT.NET
          Name Server: NS3.P06.DYNECT.NET
          Name Server: NS4.P06.DYNECT.NET
          Status: clientTransferProhibited
          Updated Date: 04-aug-2013
          Creation Date: 21-nov-2012
          Expiration Date: 21-nov-2014

    • (Score: 1) by Ethanol-fueled on Tuesday March 18 2014, @10:19PM

      by Ethanol-fueled (2792) on Tuesday March 18 2014, @10:19PM (#18360) Journal

      The attack campaign was called "Windigo." Probably carried out by a disgruntled former SGI IRIX developer who got pushed out for the Linux craze.

      • (Score: 2, Funny) by Anonymous Coward on Tuesday March 18 2014, @10:31PM

        by Anonymous Coward on Tuesday March 18 2014, @10:31PM (#18367)

        Sounds more like a Microsoft black op to me. WINdigo, no?

        • (Score: 1) by Ethanol-fueled on Tuesday March 18 2014, @10:36PM

          by Ethanol-fueled (2792) on Tuesday March 18 2014, @10:36PM (#18369) Journal

          Drive-by hacking of a Windows install is like shooting fish in a barrel. The beauty of Microsoft's business model is that they can with a straight face claim plausible deniability given their historically shoddy attitude towards security; not to mention their cushy and profitable relationship with the United States' security services.

      • (Score: 1) by carguy on Tuesday March 18 2014, @10:33PM

        by carguy (568) on Tuesday March 18 2014, @10:33PM (#18368)

        I see what you did there...
        https://en.wikipedia.org/wiki/SGI_Indigo [wikipedia.org]

    • (Score: 0) by Anonymous Coward on Wednesday March 19 2014, @01:19AM

      by Anonymous Coward on Wednesday March 19 2014, @01:19AM (#18411)

      Anonymous Coward retracting. www.joker.com is a real domain registrar. Now you know why we are anonymous.

      Things to check:
      OpenSSH backdoor

      ssh -G (should produce: ssh: illegal option -- G)

      Other tests:
      rpm -qi openssh-server (should have a signature entry.)
      ls -l /lib/libkeyutils* (The library file should be less than 10kb in size.)

      Other signs of compromise:

      Existence of a /home/ ./ directory hiding binaries and configration files of TinyDNS ...

      • (Score: 4, Interesting) by Appalbarry on Wednesday March 19 2014, @02:05AM

        by Appalbarry (66) on Wednesday March 19 2014, @02:05AM (#18422) Homepage Journal

        Joker.com is notorious for hosting and/or registering any kind of slimeball spammers. Various people have found their e-mail disappearing off the map when joker registered domains were blacklisted for evil activity.

        • (Score: 3, Interesting) by ls671 on Wednesday March 19 2014, @02:19AM

          by ls671 (891) on Wednesday March 19 2014, @02:19AM (#18432) Homepage

          That's funny. Who would want to have somebody with a name like joker.com as their registrar?

          I am still with networksolutions, When I registered my first .com domains I had no choice. Back then, they had a monopoly on all .com domains.

          They are far from perfect and are still a little more expensive than the godaddy, or joker.com around but it isn't worth the time for me now,

          Anybody know which registrar are good is I ever want to move from networksolutions?

          --
          Everything I write is lies, including this sentence.
          • (Score: 1) by TK on Wednesday March 19 2014, @03:54PM

            by TK (2760) on Wednesday March 19 2014, @03:54PM (#18673)

            Look up "sopa godaddy boycott" on your search engine of choice. There were quite a few other companies being touted as alternatives when that was going on.

            --
            The fleas have smaller fleas, upon their backs to bite them, and those fleas have lesser fleas, and so ad infinitum
          • (Score: 1) by toygeek on Wednesday March 19 2014, @04:24PM

            by toygeek (28) on Wednesday March 19 2014, @04:24PM (#18686) Homepage

            Namecheap.com has been treating me well the last couple of years :)

            --
            There is no Sig.
      • (Score: 2) by ls671 on Wednesday March 19 2014, @02:11AM

        by ls671 (891) on Wednesday March 19 2014, @02:11AM (#18426) Homepage

        ls -l /lib/libkeyutils*

        no such files in any of the machines I own.

        Also tested with /usr/lib

        --
        Everything I write is lies, including this sentence.
        • (Score: 3, Informative) by ticho on Wednesday March 19 2014, @02:34AM

          by ticho (89) on Wednesday March 19 2014, @02:34AM (#18437) Homepage

          It's in /lib64 on 64-bit systems (at least on CentOS).

        • (Score: 0) by Anonymous Coward on Wednesday March 19 2014, @08:30AM

          by Anonymous Coward on Wednesday March 19 2014, @08:30AM (#18502)

          I found /lib/x86_64-linux-gnu/libkeyutils.so.1.4

          on my ubuntu box. As near as I can tell, the most recent legitimate version of keyuitls is 1.5.9, so the 'fake' version supersedes that by being at a higher point in the path and by claiming a more recent version.

          I see reports of it going back to mid-Feb, so I don't think it's a hoax.

  • (Score: 5, Informative) by frojack on Tuesday March 18 2014, @11:48PM

    by frojack (1554) on Tuesday March 18 2014, @11:48PM (#18392)

    The report is interesting, but you want to skip to appendix 1 and run the simple detection tests.

    Just do it.

    --
    Discussion should abhor vacuity, as space does a vacuum.
    • (Score: 4, Insightful) by ls671 on Wednesday March 19 2014, @12:34AM

      by ls671 (891) on Wednesday March 19 2014, @12:34AM (#18404) Homepage

      "Just do it."

      Hmm, for some reason that raises a red flag and makes be not want to do it.

      What if appendix 1 was infected?

      --
      Everything I write is lies, including this sentence.
      • (Score: 2) by ls671 on Wednesday March 19 2014, @01:51AM

        by ls671 (891) on Wednesday March 19 2014, @01:51AM (#18419) Homepage

        What the heck? All you need to do to supposedly test if you are infected is verifying if your ssh program supports the -G option?

        Is it April fool yet?

        Anyway, if this is true, this would have been caught easily by routine md5 sums check on important executables like ssh.

        Who supports the -G option and what is it supposed to do?

        --
        Everything I write is lies, including this sentence.
        • (Score: 4, Insightful) by frojack on Wednesday March 19 2014, @02:33AM

          by frojack (1554) on Wednesday March 19 2014, @02:33AM (#18436)

          Its not supposed to do anything. Because - G is not supported.
          Its what it does when it spits an error that is indicative.

          But that specic test alone is unreliable. There are additional things to test.

          --
          Discussion should abhor vacuity, as space does a vacuum.
          • (Score: 0) by Anonymous Coward on Wednesday March 19 2014, @10:08AM

            by Anonymous Coward on Wednesday March 19 2014, @10:08AM (#18541)

            There are additional things to test.

            And those things would be...?

            • (Score: 2) by frojack on Wednesday March 19 2014, @01:42PM

              by frojack (1554) on Wednesday March 19 2014, @01:42PM (#18633)

              Mentioned in the pdf.

              --
              Discussion should abhor vacuity, as space does a vacuum.
    • (Score: 2) by tibman on Wednesday March 19 2014, @12:45AM

      by tibman (134) on Wednesday March 19 2014, @12:45AM (#18407)

      Looks like i'm clean : )

      --
      SN won't survive on lurkers alone. Write comments.
  • (Score: 3, Interesting) by gishzida on Wednesday March 19 2014, @12:21AM

    by gishzida (2870) on Wednesday March 19 2014, @12:21AM (#18401) Journal

    The Ars Technica article was a lot more help than trying to read the sludge of the ESET report.

    My old home web server it seems is infected. See my Journal [dev.soylentnews.org] for specifics...

    Thanks SN for the heads up!

    • (Score: 2) by frojack on Wednesday March 19 2014, @01:19AM

      by frojack (1554) on Wednesday March 19 2014, @01:19AM (#18410)

      There is information in the sludge about cleaning.
      You might give it a try.

      --
      Discussion should abhor vacuity, as space does a vacuum.
    • (Score: 2) by ls671 on Wednesday March 19 2014, @01:26AM

      by ls671 (891) on Wednesday March 19 2014, @01:26AM (#18412) Homepage

      " See my Journal for specifics..."

      Details about the failed tests would have been interesting. It could have false positives.

      Anyway, nowadays, it is preferable to have a fancy firewall that inspects packets + a web application firewall (WAF) if you want to do what you are doing. Also plan some time to configure and maintain your security layer properly. I always have run an IP firewall since ipfwadm but lately I installed mod_security WAF and it is amazing what is going on out there. Definitely worth in 2014.

      --
      Everything I write is lies, including this sentence.
      • (Score: 2, Informative) by gishzida on Wednesday March 19 2014, @02:26AM

        by gishzida (2870) on Wednesday March 19 2014, @02:26AM (#18434) Journal

        This particular version of SuSE had the "AppArmor" firewall installed... The gateway router [an ASUS N900] does not allow ssh from the outside and ssh was disable as a login on the server.

        I checked the system using all of the tests mentioned.

        The ssh -G test gives the "Illegal" operation error and when I ran the full test got "infected" [why not just save time and say echo "infected"?]

        There are no share memory spaces larger that about 700K and none with 666 permissions, some with 600

        The other tests come up null [i.e. not infected]

        curl, rpm -qi, ls -I, and an examination of /home reveal nothing out of the ordinary.

        opened the package manager and removed ssh and the X related ssh package. Re-ran the ssh -g test and got "ssh not installed"...

        I have not used ssh as a way of getting into this system [Why should I? I can walk to the console].

        I'm not sure if the system was actually infected or if this isn't just another NSA surprise security package that has surfaced. Think about that before you say "can't be..." no black ops organization is above a little black market work... as I recall the CIA has been accused [wikipedia.org] of running some side businesses... then again maybe it's just my ol' hippie paranoia meter running out of kilter. But I will tell you I haven't done much in the way of patching this thing in maybe five years.

        • (Score: 2) by ls671 on Wednesday March 19 2014, @02:50AM

          by ls671 (891) on Wednesday March 19 2014, @02:50AM (#18438) Homepage

          " Think about that before you say "can't be..." no black ops organization is above a little black market work..."

          Relax man, it is common in human behavior to run like headless chickens when a crisis occurs. I have seen it in some organization where people would think they are under control.

          Anyway, according from what I have read, yes I read TFA after my first post to you, if you type:

          $ ssh -G

          and it tells you "illegal option" or "unknown parameter" then you are NOT infected. No ssh client that I know of supports the -G option. It wouldn't be very clever for the hackers to do anyway.

          see my other post:

          http://dev.soylentnews.org/comments.pl?sid=731&cid=184 19 [dev.soylentnews.org]

          --
          Everything I write is lies, including this sentence.
          • (Score: 3, Informative) by ls671 on Wednesday March 19 2014, @03:20AM

            by ls671 (891) on Wednesday March 19 2014, @03:20AM (#18448) Homepage

            another poster has made a clarification:

            http://dev.soylentnews.org/comments.pl?sid=731&cid=184 19 [dev.soylentnews.org]

            So, here is my full output of ssh -G, even on a fresh install:
            ~$ ssh -G
            ssh: illegal option -- G
            usage: ssh [-1246AaCfgKkMNnqsTtVvXxYy] [-b bind_address] [-c cipher_spec]
                                  [-D [bind_address:]port] [-e escape_char] [-F configfile]
                                  [-I pkcs11] [-i identity_file]
                                  [-L [bind_address:]port:host:hostport]
                                  [-l login_name] [-m mac_spec] [-O ctl_cmd] [-o option] [-p port]
                                  [-R [bind_address:]port:host:hostport] [-S ctl_path]
                                  [-W host:port] [-w local_tun[:remote_tun]]
                                  [user@]hostname [command]

            yet another poster with a good point:
            http://dev.soylentnews.org/comments.pl?sid=731&cid=184 26 [dev.soylentnews.org]

            --
            Everything I write is lies, including this sentence.
            • (Score: 0) by Anonymous Coward on Wednesday March 19 2014, @05:38AM

              by Anonymous Coward on Wednesday March 19 2014, @05:38AM (#18471)

              So, here is my full output of ssh -G, even on a fresh install:

              Might be a little more helpful if you said fresh install of what!

              • (Score: 1) by tempest on Wednesday March 19 2014, @08:58AM

                by tempest (3050) on Wednesday March 19 2014, @08:58AM (#18515)

                Since ssh is cross platform, it says that on a fresh install of anything. I'm assuming he was using Linux, but it says that on FreeBSD too.

          • (Score: 1) by gishzida on Wednesday March 19 2014, @03:29AM

            by gishzida (2870) on Wednesday March 19 2014, @03:29AM (#18450) Journal

            No crisis here in the Chicken Little Data Center... Just an old geezer getting stuck in the sludge... I really do need to do a new install on this server... it seems to have grown barnacles.... well, maybe at least one of us has... :)

            Thanks for the reply and clarification.

  • (Score: 1) by SCY on Wednesday March 19 2014, @04:24AM

    by SCY (3006) on Wednesday March 19 2014, @04:24AM (#18456)

    Somebody around here who actually got infected by this?

    • (Score: 2) by ls671 on Wednesday March 19 2014, @05:15AM

      by ls671 (891) on Wednesday March 19 2014, @05:15AM (#18466) Homepage

      None really confirmed. I am still wondering if this is real...

      If it is; hello! fake sshd? come on, md5 or sha sums on your binaries once in a while anybody?

      It doesn't sound real. On the other hand, my logs of attacks against my infrastructure and the silliness of the attacks could make be beleive it is really happening.

      --
      Everything I write is lies, including this sentence.
      • (Score: 2) by ls671 on Wednesday March 19 2014, @05:27AM

        by ls671 (891) on Wednesday March 19 2014, @05:27AM (#18467) Homepage

        By the way, anybody heard of a bot attacking sites with the "COOK" HTTP method?

        Cooking is a well known term to take control of assets but I can't find anything about the COOK method being sent by some clients.

        37.19.151.14 - - [19/Mar/2014:02:46:41 -0400] "COOK /index.html...
        94.230.8.164 - - [19/Mar/2014:02:46:26 -0400] "COOK /index.html

        Normal log would look like:
        94.75.193.139 - - [19/Mar/2014:02:33:58 -0400] "GET /index.html

        Thanks,

        P.S. Of course, COOK requests get a 403

        --
        Everything I write is lies, including this sentence.
        • (Score: 2) by stderr on Wednesday March 19 2014, @10:17AM

          by stderr (11) on Wednesday March 19 2014, @10:17AM (#18546) Journal

          P.S. Of course, COOK requests get a 403

          Why is that "of course"?

          403 Forbidden
          The server understood the request, but is refusing to fulfill it. Authorization will not help and the request SHOULD NOT be repeated. If the request method was not HEAD and the server wishes to make public why the request has not been fulfilled, it SHOULD describe the reason for the refusal in the entity. If the server does not wish to make this information available to the client, the status code 404 (Not Found) can be used instead.

          Since your server (hopefully) didn't understand the request, 403 is the wrong status code to use.

          The quote about is from section 10.4.4 [ietf.org] of the RFC. Section 5.1.1 [ietf.org] says:

          An origin server SHOULD return the status code 405 (Method Not Allowed) if the method is known by the origin server but not allowed for the requested resource, and 501 (Not Implemented) if the method is unrecognized or not implemented by the origin server.

          --
          alias sudo="echo make it yourself #" # ... and get off my lawn!
          • (Score: 2) by ls671 on Wednesday March 19 2014, @11:27PM

            by ls671 (891) on Wednesday March 19 2014, @11:27PM (#18768) Homepage

            mod_security just generically returns 403 when it detects something wrong but this could be easily fine tuned.

            I specifically fine tuned a few cases to return other codes but I am mostly OK with generically returning 403s. I am quite aware of the different HTTP status codes.

            I think that the idea is to generically forbid access to the resource without giving any hints about what went wrong thus letting the attacker know less about your internal configuration.

            FYI, rfc2119 (Key words for use in RFCs to Indicate Requirement Levels) defines SHOULD as:

              3. SHOULD
              This word, or the adjective "RECOMMENDED", mean that there
                  may exist valid reasons in particular circumstances to ignore a
                  particular item, but the full implications must be understood and
                  carefully weighed before choosing a different course.

            --
            Everything I write is lies, including this sentence.
      • (Score: 2) by ls671 on Wednesday March 19 2014, @05:32AM

        by ls671 (891) on Wednesday March 19 2014, @05:32AM (#18468) Homepage

        "and the silliness of the attacks could make be beleive it is really happening"

        The "silliness" in the sense that it looks like one clan or competing clans testing if you are already infected by the would be other clan. The actual intrusion seems to still be done through things like "click on this link".

        --
        Everything I write is lies, including this sentence.
    • (Score: 0) by Anonymous Coward on Wednesday March 19 2014, @03:32PM

      by Anonymous Coward on Wednesday March 19 2014, @03:32PM (#18664)

      Whenever I hear "Linux exploited", my initial reaction is always "incompetent admin".
      My mind immediately goes to the episode where Jerry owns the most-touted lockset on the market...then along comes Kramer [google.com]. Surely I can't be alone here.

      -- gewg_

  • (Score: 3, Informative) by lubricus on Wednesday March 19 2014, @06:45AM

    by lubricus (232) on Wednesday March 19 2014, @06:45AM (#18482)

    Apparently, ssh behaves differently on an infected system, it can be checked with this:

    $ ssh -G 2>&1 | grep -e illegal -e unknown > /dev/null && echo "System clean" || echo "System infected"

    More detail and checks in the report.

    --
    ... sorry about the typos
    • (Score: 2) by stderr on Wednesday March 19 2014, @10:25AM

      by stderr (11) on Wednesday March 19 2014, @10:25AM (#18551) Journal

      That's a pretty naive test. I bet the next version of the malware will say "illegal option" to hide from that simple test.

      Here's an updated version of the "test":

      ssh -G 2>&1 | grep -e illegal -e unknown > /dev/null && echo "System might be clean or it might not be. Run a _REAL_ test" || echo "System infected"

      --
      alias sudo="echo make it yourself #" # ... and get off my lawn!
  • (Score: 1) by cosurgi on Wednesday March 19 2014, @02:52PM

    by cosurgi (272) on Wednesday March 19 2014, @02:52PM (#18651) Journal
    On debian you can just run:

    sudo aptitude install debsums
    # then as non-root user:
    debsums > result
    cat result | grep -v -e "OK$"


    This will catch all suspicious binaries.
    --
    #
    #\ @ ? [adom.de] Colonize Mars [kozicki.pl]
    #
    • (Score: 1) by toygeek on Wednesday March 19 2014, @04:31PM

      by toygeek (28) on Wednesday March 19 2014, @04:31PM (#18689) Homepage

      slightly cleaner, no intermediate file needed:

      debsums | grep -v -e "OK$"

      --
      There is no Sig.
      • (Score: 2) by ls671 on Wednesday March 19 2014, @11:45PM

        by ls671 (891) on Wednesday March 19 2014, @11:45PM (#18772) Homepage

        Sure, then, when you are done, do a:

        debsums | less

        To view the details of what has been checked or: debsums | grep -e "OK$"

        Only problem, you have just computed your sums multiple times.

        It isn't a bad idea at all to cache the output of operations like computing checksums on a bunch of files.

        --
        Everything I write is lies, including this sentence.