From: Adam P. <aph...@gm...> - 2014-05-22 19:42:33
|
Hi John, Have you tracked the memory of these jobs? Sounds like it may be exhausting system memory--that's generally the only way I see my cluster nodes bricked. If not, that's Mike Schatz's code, he may have more insight. Best, -Adam On Sun, May 18, 2014 at 10:24 PM, John Johnston <jo...@ms...> wrote: > Hello, > > I have a user that is attempting to use the "amosvalidate" utility on a > HPC cluster, and it appears to be crashing nodes. > > The process appears to proceed normally until reaching the "analyzeSNPs" > step with the command (obtained from the log file): > > analyzeSNPs -i -b/ /assembly.bnk -S -cumqv 40 -minsnps 2 -r -H > > assembly.snps > > Searching 223685 contigs > > At this point, log output ceases, the job terminates abnormally, and the > scheduling system on that node registers in a "down" state. We've seen > this > on multiple nodes. > > Other users have used other parts of the AMOS package over the last 3 > years on the system without these problems (though *not* specifically > amosvalidate). > AMOS is installed on a RHEL 6 system with the torque/PBS resource > manager and moab scheduler. > > The version in use is 3.1.0 released in 2011. > > Has anyone seen anything like this or might someone have any insight on > this issue? > > Thanks. > > > > ------------------------------------------------------------------------------ > "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE > Instantly run your Selenium tests across 300+ browser/OS combos. > Get unparalleled scalability from the best Selenium testing platform > available > Simple to use. Nothing to install. Get started now for free." > http://p.sf.net/sfu/SauceLabs > _______________________________________________ > AMOS-help mailing list > AMO...@li... > https://lists.sourceforge.net/lists/listinfo/amos-help > |