Weekly meeting for 2022-07-19

[denizeu]

What I accomplished last week

• Worked on adding a conditional to kneaddata rule:
• tried code in run before shell, tried adding params
• Created a branch for the auto resubmit issue, tried to use resources for this

What I plan to do next week

• [ ] Actually figure out code that works for pulling info from the cluster file
• [ ] Work on the auto resubmit issue

Is there anything that Kevin can help with

• Troubleshooting..
• How to get code to actually go into the cluster file for kneaddata
• Does resources need things to be specifically defined somewhere?
• Help me see if there is something I'm still not understanding properly

How did the last week go?

• [ ] I accomplished all or most of my goals
• [X] I used my time wisely
• [ ] I struggled with a problem, but solved it
• [X] I ran into an issue I can't solve on my own

[denizeu]

if have cluster config use value for kneaddata, if not go to default: set a manually entered default (1)

or else just always pass cluster config file pull into rule if able to use code that I wrote for cluster config (c8f83940da3c687967a7d9acb1f02d5734134033)

ex: if cluster.yaml exists use kneaddata info, if not use manually entered default: processors= 1

goal: user never modifies code, can modify config files

internal and API (user exposed to)

• expose the part of the design: only thing user does is changing config files

[denizeu]

to test: run cluster config also add --cluster-config cluster.yaml & add to README

Don't do anything more complicated than double values for everything each try

humann uses 12 hour as a start, 12 hour limit on cluster so can't pull more resources

[denizeu]

resources defined inside cluster config could be pulled, even from default if no rule name