secondaryDS containing colon breaks about panda-client HOT 7 CLOSED

# can be used as a separator for --secondaryDS in 0.6.10.

from panda-client.

Hi Tadashi,

we have another problem with secondaryDS. If we split the primary input dataset into using nFilesPerJob the secondary datasets get used only in the first job. Is that how is it supposed to work? We need to have 2 files as input for all the jobs. How can we do that?

thanks

cheers
alessandra

from panda-client.

I forgot to put an example of task that failed in this way https://bigpanda.cern.ch/task/16204845/

from panda-client.

I suppose you need --reusableSecondary=IN2,IN3

from panda-client.

@tmaeno can you explain what the difference would be?

from panda-client.

--reusableSecondary=REUSABLESECONDARY
                        A comma-separated list of secondary streams which
                        reuse files when all files are used

Without this (default), files are used only once.

from panda-client.

I've submitted a task with the extra parameter.

https://bigpanda.cern.ch/task/16219599/

from panda-client.