OrthoFinder
OrthoFinder copied to clipboard
davidemms
Bus error w/ large run
Hello,
I'm currently trying to run an orthofinder job w/ a large (115) number of proteomes and i'm having real trouble trying to get past the mcl stage:
/usr/bin/time -v orthofinder -p ${TEMP_DIR} -a 3 -t 40 -S mmseqs -f ./ > out_log_all_rerun_full.txt
Command terminated by signal 7 Command being timed: "orthofinder -p /app/data/tmp -a 3 -t 40 -S mmseqs -f ./" User time (seconds): 6540672.39 System time (seconds): 5824.53 Percent of CPU this job got: 3028% Elapsed (wall clock) time (h:mm:ss or m:ss): 60:02:58 Average shared text size (kbytes): 0 Average unshared data size (kbytes): 0 Average stack size (kbytes): 0 Average total size (kbytes): 0 Maximum resident set size (kbytes): 28483104 Average resident set size (kbytes): 0 Major (requiring I/O) page faults: 6953 Minor (reclaiming a frame) page faults: 909715452 Voluntary context switches: 4487919 Involuntary context switches: 11632318 Swaps: 0 File system inputs: 902573176 File system outputs: 2432579064 Socket messages sent: 0 Socket messages received: 0 Signals delivered: 0 Page size (bytes): 4096 Exit status: 0
The machine has a large (256 gb) amount of RAM, and i don't seem to hitting that cap... I've honestly hit a snag and i'm out of ideas. I'm able to run the first ~70 out of 115 through, but when i include more it seems to crash out. Are the page faults an issue?
Any ideas as to causes would be super useful, i'm at a loss at this point.
--Thank you very much
I'm going to bump this. I'm hitting Bus errors as well. My current run only has 20 samples and I'm still getting a bus error at around the mcl stage. Attached job log below....
I also have snapshots of running processes over time.
...This isn't my only dataset encountering this issue, but it is the smallest.
