Hello,

I got two new tasks from Rosetta that seem to not have checkpoints enabled: I left them running up to about 25%, then I had to switch off my server
and after it was switched on again I could see those tasks were reset to 0% done.
They have similar name than others previously processed that I am almost sure were not having this issue, so I suppose these specific two are some way "defective".
Running to completion in one shot those tasks requires about 9h of elapsed and now I am at risk to break their deadline. Therefore I will probably have
to abort them after have spent already more than 5h CPU time (3+ hours first run and 2+ hours this second run).
Is there a way to report this kind of "malfunctions" to the project?
Otherwise, if that task behavior is expected by desing for that project, I would like to raise that, in my opinion, it is not fair to force contributors
to run tasks for so long time in a row.

Hello,

I got two new tasks from Rosetta that seem to not have checkpoints enabled: I left them running up to about 25%, then I had to switch off my server
and after it was switched on again I could see those tasks were reset to 0% done.
They have similar name than others previously processed that I am almost sure were not having this issue, so I suppose these specific two are some way "defective".
Running to completion in one shot those tasks requires about 9h of elapsed and now I am at risk to break their deadline. Therefore I will probably have
to abort them after have spent already more than 5h CPU time (3+ hours first run and 2+ hours this second run).
Is there a way to report this kind of "malfunctions" to the project?
Otherwise, if that task behavior is expected by desing for that project, I would like to raise that, in my opinion, it is not fair to force contributors
to run tasks for so long time in a row.

Hello,

just to provide more details:
all Rosetta task with name similar to (numbers may vary)
rb_10_15_140181_136445_ab_t000__robetta_cstwr_5.0_FT_IGNORE_THE_REST_03_10_2093785_210_0
do not save the status in case the computer running them is rebooted. If that happens they are reset to zero.
This force the contributor to keep them running for the required 9h in a row. Aborting them results of course in
"failed tasks" into statistics.
Since most part of tasks I receive from Rosetta seems to be of this type, I am wondering if I would be better to
unsubscribe the project. I don't like the idea to do that but I don't see other options (I don't want to leave my computers
running continuously for 9h).
Any hint?
I am surprised nobody else is affected by this issue. Maybe everybody is running them on cpus that much faster than mines
that they don't realize there is this issue?