1

Got a complaint from one of our customer saying that they were unable to ssh into our CentOS 32 bit system. On console, even 'ls' was not possible. The reason being :

/bin/ls: /lib/ld-linux.so.2: bad ELF interpreter: No such file or directory

Customer insists that no drastic commands were executed on the system by anyone. And they have re-imaged the unit.

What could be the reason, that the dynamic loader itself can go missing?

On reboot the following is seen:

Pid: 1, comm: init Not tainted 2.6.32-573.3.1.el6.i686 #1                       
Call Trace:                                                                    
 [<c086c5c7>] ? panic+0x6e/0x122                                                
 [<c0464793>] ? do_exit+0x743/0x750                                             
 [<c04647dc>] ? do_group_exit+0x3c/0xa0                                         
 [<c0464851>] ? sys_exit_group+0x11/0x20                                       
 [<c0409bbf>] ? sysenter_do_call+0x12/0x28
 [<c0b00000>] ? startup_xen+0x0/0x11

Appreciate any comments regarding this case(presumably unique?) if we accept that a user with root privilege didn't meddle with the system?

Prabhu
  • 11
  • 2
    If you inspect the filesystem, does the file actually exist? This error generally means that (a) someone deleted ld-linux.so, or (b) they are attempting to run a binary for a different architecture (e.g., trying to run a 64bit binary on a 32bit system). If the customer has re-imaged the system, you've really lost your opportunity to perform further diagnosis. – larsks Jul 28 '16 at 12:55
  • @larsks Yes. Currently we have lost the system for good. We didn't have the opportunity to explore. Customer insists that nothing was done. I think it is the case of 'a' mainly. 'ls' command too failed.. – Prabhu Jul 28 '16 at 13:13

0 Answers0