Dear all,<br>We are running the latest version of the code on Fedora Core 7 using the latest MKL and ifc compiler. Our system is Intel(R) Core(TM)2 Quad CPU @ 2.40GHz.<br><br>Our case, containing 60 atoms, is converged selfconsistently, if we perform run_lapw -p interactively. However, we would run it in background using at daemon. But surprisingly, we have observed the following error at first cycle, when we perform run_lapw -p using "at now" command:<br>** testerror: Error in Parallel LAPW2<br><br>This is in the case that the at daemon works well for other lighter cases. Even it works for this heave case, if we reduce the Rkmax.<br><br>The RAM memory of our system is 2 GB, and the swap memorry is 4 GB. When the 4 cores of our system are involved to run, in parallel, the lapw1, almost 3.5 GB out of the 4 GB swap space is used.<br><br>We are surprised from where the problem of running the 60 atoms case only at background (and not interactively!)
comes.<br><br>Your,<br>S. Jalali.<br><br><br><br><b><i>yxl@email.jlu.edu.cn</i></b> wrote:<blockquote class="replbq" style="border-left: 2px solid rgb(16, 16, 255); margin-left: 5px; padding-left: 5px;"> Dear users,<br> Do you have ever met the same situation? In step kgen the number of k-points is <br>the same, if enter 'yes' in 'Shift k-mesh' then all things will be ok until convergence,<br> but if enter 'no' then lapw1 error: Cholesky INFO = 4715<br> 'SECLR4' - POTRF (Scalapack/LAPACK) failed.<br> I couldn't understand it, what's the difference between them. Any suggestions will<br>be appreciated.<br>Best wishes!<br> yours sincerely,<br> hongxia<br>_______________________________________________<br>Wien mailing
list<br>Wien@zeus.theochem.tuwien.ac.at<br>http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien<br></blockquote><br><p> 
<hr size=1>Moody friends. Drama queens. Your life? Nope! - their life, your story.<br> <a href="http://us.rd.yahoo.com/evt=48224/*http://sims.yahoo.com/">Play Sims Stories at Yahoo! Games. </a>