Hello. I have been doing a rosetta@home task wich had at the beginning a estimation of 8 hours of work left. For some reason as progress went up, progress became slower and slower. (first 50% took about 4 hours, 50-90% took 12 hours, 90%-95% took 24 hours 95-99% took 48 hours and im now at 99.8% wich took 2.5 days) Am I doing something wrong somewhere or is there any problem ongoing or is this normal? (I have 5 tasks like this atm). Ive been running at 100% cpu all that time. And is there anything I can do to avoid this? Please ask if you need additional informations

It looks like you now have no Tasks as you aborted all of them.
Some Python Tasks do have a problem with not completing- generally if it takes more than 12 hours for a Task to complete, then it's worth aborting it. No need to abort any others at the same time unless the same thing occurs with them as well.

While they don't actually use a lot of RAM, the current python Tasks require a minimum of around 3GB per Task in order to run (earlier ones required 8GB, even though they too didn't actually use nearly that much). So depending on what values you have selected for your Computing preferences, the most you will be able to run at any given time on your system will probably be around 4 or so (maybe 5 at a maximum). At least until new Rosetta 4.20 work is released sometime next year.

---

Grant
Darwin NT

I'm having the same issue on all tasks anymore. All going over at least 1 1/2 days if you are lucky. 2 days if you aren't. Must be a problem someplace.

I have one such python/VB task. It's in progress for 2.5 days already (99.93% completion).
It reached the deadline and rosetta stopped sending new python/VB tasks for me due to the issue. It can be turned on easily, btw.
It's just funny moment - I want to complete the task :) Only 2:25 minutes left so, I tnihk it may be ready this evening... :D

PS: Task ID is "aagb-SAR_pp-NMABU-mACPenC13T-ACHC12T_pp_4_2716600_2" if someone need this information for investigation.

I was having this problem consistently for the rosetta python tasks. With an estimate of about 8 hours, the task runs to somewhere between 97% and 99%, and then stops making progress, even though BOINC manager shows that its status is Running (no errors). It can sit there like that for multiple days, preventing newer tasks from being started.

I'm tired of this problem of being stuck, and then aborting the task (causing my computation time to be wasted)), and I'm also tired of the significant disk space demanded by these tasks. I'm done with VirtualBox.

I have uninstalled VirtualBox so that I no longer get these tasks.

Hoi everybody! I have the same problem, every increment of the last percentage takes longer and longer. After 4 days some tasks exceeds the Deadline and "Timed out - no response". Three other tasks I cancelnd (each after more than one day) - it seems like a hopeless case.
An other Tasks didn't start, so it chanced the progress state to active, but didn't start counting or processing.
Thanks everyone for providing the opportunity to support science and research, despite some difficulties. =)

I'm having this problem too, where about 1 task in 10 does this... gets exponentially slower as it progresses. I have one now that's at 99.996%, 3d10h elapsed, estimated 9 seconds remaining. They do still progress to completion, but it will take a few more hours I know.

My question is should I just let these run to completion? My questions is... are task results completed beyond the deadline discarded by the server, or recorded anyway? My other questions are... If the result is discarded, is the information about these tasks taking so long to complete useful to the project in itself for identifying/fixing this issue? I have many questions about whether... the information is completely useless or ignored, why aren't the overdue tasks aborted automatically by BOINC instead of wasting days of computing time? What is the meaning of life? I'll go out and come in again....

If it's definitely completely useless I'll write a scheduled script to query the tasks and automatically abort those over the deadline immediately.

This issue has come up a lot in the past.
And the solution has been found, so follow these very simple instructions

If you have a task you think is stalled or taking a really long time to complete, click on it and select properties on the left.
If there's a large difference between CPU time and Elapsed time, then it's stalled and you can only abort it. They <never> restart.

Also, if later tasks are completing before earlier tasks, it's a clue to check Properties of those earlier tasks and, if they've stalled in the way described above, abort them.
This wastes the least amount of processing time.

It's not your fault and there's <nothing> you can do to correct it

---