search

lsmo-epfl / aiida-lsmo

AiiDA workflows for the LSMO laboratory at EPFL
Other
9 stars 13 forks source link

add SL diag library specification in cp2k parameters settings #99

Open mpougin opened 1 year ago

mpougin commented 1 year ago

never CP2K versions might run in a deadlock when using the default ELPA library for diagonalization

mpougin commented 1 year ago

@kjappelbaum can you help with the docs error?

mpougin commented 1 year ago

@kjappelbaum can you help with the docs error?

The docs problem is fixed in #100

kjappelbaum commented 1 year ago

for the tests, I'd need to check manually and bisect the history. Just by looking at the logs, I have no clue. I initially suspected that there has been an update in aiida-cp2k, but there was apparently no other bugfix version ...

mpougin commented 1 year ago

for the tests, I'd need to check manually and bisect the history. Just by looking at the logs, I have no clue. I initially suspected that there has been an update in aiida-cp2k, but there was apparently no other bugfix version ...

Also, it all runs locally for me

ltalirz commented 1 year ago

note that aiida-lsmo uses aiida-testing, i.e. it will read results from the pre-computed test database if they don't exist.

did you try updating the test database? https://aiida-lsmo.readthedocs.io/en/latest/development.html#updating-the-test-data

ltalirz commented 1 year ago

Looking at the current commit history, it seems like one of the tests is flaky: https://github.com/lsmo-epfl/aiida-lsmo/issues/102

Here, 3 tests are failing, so two of those might be legitimate problems (edit: here you seem to be updating the inputs, so that will require an update of the test database)

In any case, I would suggest rebasing this branch on top of develop before continuing

mpougin commented 1 year ago

okay, thanks a lot for your input :)