Join GitHub today
GitHub is home to over 31 million developers working together to host and review code, manage projects, and build software together.Sign up
Resuming crashed job #160
I'm running a 3 iteration search job and it crushed towards the end of the prefiltering of the last iteration. I tried to run again the same command in order to resume from where I stopped, but it seems to me that it is starting from the beginning.
creates new folder inside the tmp folder and starts a new run
Steps to Reproduce (for bugs)
Are you sure you called
If anything changed in between the two calls MMseqs2 will not be able to reuse the previous run.
Btw what did the prefilter say when it crashed?
added a commit
Feb 7, 2019
So I figured out the puzzle. All parameters have to match, even ones that MMseqs2 resolves automatically such as thread count. The first run was with 48, the other with 56 threads. If you resubmit the job, such that it runs on the 48 thread machine again it should continue from the point it stopped correctly.
@martin-steinegger: should we make a list of parameters to always exclude from hashing (threads, verbosity, ...)?