Re: Question about a fault

From: Paul H. Hargrove (PHHargrove_at_lbl_dot_gov)
Date: Thu Mar 18 2010 - 19:08:11 PDT

  • Next message: Paul H. Hargrove: "Re: Question about a fault"
    Jun Zhu,
    
      From the "#" prompt in your example I would guess you are running as 
    root.  It is a common practice in Linux and other UNIX-like operating 
    systems to ignore the setting of the LD_PRELOAD environment variable 
    when running as root (or in any setuid or setgid executable) for 
    security reasons.  This is one possible explanation for the error 
    message you are seeing.
    
      Another possibility would be if "counting" is a 32-bit executable and 
    the first libcr_run.so.0 found is a 64-bit library (or the other way 
    around).  I can reproduce that case on my own workstation and get this 
    message (which I know was NOT the case on previous multilib systems I've 
    tested).  I wish I knew a way to avoid this problem, but Linux doesn't 
    appear to have separate LD_PRELOAD or LD_LIBRARY_PATH env vars for 32- 
    vs 64-bit libs.
    
    -Paul
    
    
    Jun Zhu wrote:
    > Hi all,
    >
    > When I test cr_run command it has the following errors.
    > BTW, I can use cr_checkpoint to checkpoint this process and cr_restart it.
    > I have tried the two solutions in FAQ. It is useless.
    >
    > # cr_run ./counting
    > ERROR: ld.so: object 'libcr_run.so.0' from LD_PRELOAD cannot be
    > preloaded: ignored.
    > ERROR: ld.so: object 'libcr_run.so.0' from LD_PRELOAD cannot be
    > preloaded: ignored.
    > ERROR: ld.so: object 'libcr_run.so.0' from LD_PRELOAD cannot be
    > preloaded: ignored.
    > ERROR: ld.so: object 'libcr_run.so.0' from LD_PRELOAD cannot be
    > preloaded: ignored.
    > ERROR: ld.so: object 'libcr_run.so.0' from LD_PRELOAD cannot be
    > preloaded: ignored.
    > Counting demo starting with pid 32498
    > Count = 0
    > Count = 1
    > Count = 2
    > Count = 3
    > Count = 4
    > Count = 5
    > ^C
    >   
    
    
    -- 
    Paul H. Hargrove                          PHHargrove_at_lbl_dot_gov
    Future Technologies Group                 Tel: +1-510-495-2352
    HPC Research Department                   Fax: +1-510-486-6900
    Lawrence Berkeley National Laboratory     
    

  • Next message: Paul H. Hargrove: "Re: Question about a fault"