Bo

So you are basically proposing the use of tasks to execute subworkflows (#476). In that case you can ``` [work] input: for_each=dict(i=range(10000)) R: process {i} [10] task: sos_run('work') ``` where...

Then a compromised solution would be ``` [work] task: input: for_each=dict(i=range(10000)) R: process {i} [10] sos_run('work') ``` which basically sends the entire step as a task, with all the substeps....

Note that, however, whereas we can now run tasks at substep level and skip passed substeps, running the entire step as task will not be able to achieve this so...

This means one of the workers has been killed by external force, and currently SoS exits without trying to recover from the error. Do you have any idea why it...

Recovering from a failed worker is very difficult because the master has to know who is doing what, use a ping-pong protocol to check if the status of workers, then...

> the new UKB release is only available through this platform Curious as why this is the case since we are interested in running UKB data on DNAnexus as well.

There are two possibilities. The first is to run sos scripts entirely on the platform by specifying python sos etc as dependencies. We will not be able to use our...

Also building and uploading docker images could be a more general solution. https://youtu.be/A_iki_50Ig0

`sessioninfo()` allows external information to be added as variable `sessioninfo` so I suppose this can be added manually? It is possible to allow `Py_Modules` etc to modify `sessioninfo` though.

That branch is progressing step by step to keep the current tests passing. Right now the slot manager is gone, signatures are written by a single thread, with #1056 as...