These are chat archives for nextflow-io/nextflow

16th
Jan 2017
Félix C. Morency
@fmorency
Jan 16 2017 15:11
I see
Félix C. Morency
@fmorency
Jan 16 2017 15:19
"I frequently see processes re-run that I don't expect" → This is exactly the behaviour I'm seeing
Paolo Di Tommaso
@pditommaso
Jan 16 2017 15:29
Are u using the latest version?
Félix C. Morency
@fmorency
Jan 16 2017 15:30
0.23.0
Paolo Di Tommaso
@pditommaso
Jan 16 2017 15:31
Yes, that's the latest
Well, in principle it should not happen
There could be something weird in your code
Try to use the -dump-hashes option to identify the difference between two runs
Félix C. Morency
@fmorency
Jan 16 2017 15:33
Can I run this command post-runs?
Paolo Di Tommaso
@pditommaso
Jan 16 2017 15:35
Well what happens if u -resume it?
If processes are not cached you can compare two -resume runs
Félix C. Morency
@fmorency
Jan 16 2017 15:38
Ok. I will try that.
Denis Moreno
@Galithil
Jan 16 2017 16:21
Hi ! I'm looking into the --with_trace option while running on aws. I get very different results for the same process in terms of time, while cpu usage, ram usage, and IO stats are consistent. I don't exactly know how long each process runs for, and I was expecting the trace file to help with that, but I'm getting wary of its data. Do you have any input about this ?
Paolo Di Tommaso
@pditommaso
Jan 16 2017 16:25
Um, what time? duration or realtime?
If u are using Docker, the first may include the time to download the image