Finally got my machine to start downloading tasks again (Problem at Boincstats) and now everything it's downloading is failing after a short computation time. Is there a way around this issue?

As mentioned a few posts earlier, the majority of the new Tasks will error out in a few seconds, but the rest will run OK.

As it is, it looks like that batch has all been sent out, so it'll just be resends again until another larger batch of work is released.



NB- since you are running more than one project, i would suggest reducing your cache size to a half day or less. At times like these where Rosetta is out of work, your system will do just Einstein, when Rosetta gets more work then it will do more of Rosetta until your Resource share settings are being met. No need to cache work.

eg
Computing preferences, Other,

           Store at least 0.4  days of work
Store up to an additional 0.01 days of work

---

Grant
Darwin NT

Finally got my machine to start downloading tasks again (Problem at Boincstats) and now everything it's downloading is failing after a short computation time. Is there a way around this issue?

Errors after very short computation times are usually due to errors in one of the input files. In that case, not much can be done except to have the project staff cancel every task that shares the defective input file. Getting them to notice the need to do do is usually quite difficult.

I saw two of the last batch I got fail quickly, but the other five have been running for an hour. This may mean that they have filtered out some, but not all, of the tasks with defective inputs.

Thank you Grant. I'll give it a go

HI,
Since 3 days no new tasks, is there a problem on my side ?

08/08/2021 11:13:17 | Rosetta@home | Project requested delay of 31 seconds
08/08/2021 11:13:52 | Rosetta@home | Sending scheduler request: To fetch work.
08/08/2021 11:13:52 | Rosetta@home | Requesting new tasks for CPU
08/08/2021 11:13:53 | Rosetta@home | Scheduler request completed: got 0 new tasks
08/08/2021 11:13:53 | Rosetta@home | No tasks sent
08/08/2021 11:13:53 | Rosetta@home | Project requested delay of 31 seconds
08/08/2021 11:28:33 | Rosetta@home | Sending scheduler request: To fetch work.
08/08/2021 11:28:33 | Rosetta@home | Requesting new tasks for CPU
08/08/2021 11:28:34 | Rosetta@home | Scheduler request completed: got 0 new tasks
08/08/2021 11:28:34 | Rosetta@home | No tasks sent

Thanks
Albert H.

---

HI,
Since 3 days no new tasks, is there a problem on my side ?

The project is presently out of work.
There are some resends, and a couple of small batches of work being released, but it's down to luck as to whether you get any or not.
Hopefully come the start of the new work week in the US more work will be loaded up at some stage and will be readily available.

---

Grant
Darwin NT

i got only resent ^^

and some GB10_3CL errored out too,

Since 3 days no new tasks, is there a problem on my side ?

The project is presently out of work.
There are some resends, and a couple of small batches of work being released, but it's down to luck as to whether you get any or not.
Hopefully come the start of the new work week in the US more work will be loaded up at some stage and will be readily available.

I managed to get 2.64 seconds total runtime from my last 40 errored tasks #Blessed
Time to give WCG its head until they fix it - sometime whenever

---

I also had nothing from Roseta for almost a week. Then I got a string or tasks that worked out as "errors while computing."

Must be something amiss with the new setup?

S. Gaber

I also had nothing from Roseta for almost a week. Then I got a string or tasks that worked out as "errors while computing."

No work here and problems on Ralph
Seems: "Closed for holiday" :-P

---

I've notified the researcher that submitted the jobs that are failing.

We are looking into the HTTP download issues at the moment.

---

Not a new problem. It often and it last many days. Not very serious! :-(
I had to go back to W,C,G.

Not a new problem. It often and it last many days. Not very serious! :-(
I had to go back to W,C,G.

W.C.G. is now almost out of tasks, so you may need to add another project.

I have lots of WCG w.u. in store. And yes I can join an other project. I can stop crunching altogether. All that does not solve the fact that this project runs out of W.U. just too often without explanation. That show how much they care about the crunchers. And no I do not know who (they) are.

Their job is to do research, not to create meaningless work, so sometimes there is down-time.

It is especially understandable at the moment with such big improvements being made in the field since AlphaFold 2's showing at CASP13 and more recently, since releasing their source code.

It would be good to have more information posted about what is happening behind the scenes and the different versions of Rosetta though, especially the vbox versions.

---

I got lots of work at the moment.

Albert

---

Their job is to do research, not to create meaningless work, so sometimes there is down-time.

It is especially understandable at the moment with such big improvements being made in the field since AlphaFold 2's showing at CASP13 and more recently, since releasing their source code.

It would be good to have more information posted about what is happening behind the scenes and the different versions of Rosetta though, especially the vbox versions.

I thought RoseTTAFold was one step ahead of AlphaFold 2 now?

I don't think so - I think RoseTTaFold was closing the gap, but the reality is probably more complicated than that. I would guess that there are lots of areas where the are differences, like training improving the training, as well as the modelling.

---

And it looks like we're out of work again.

---

Grant
Darwin NT

Tasks ready to send 0
Tasks in progress 169215

And now that it is the weekend, nothing will change.
RAH is beginning to make me wonder if its stable or not.
Bugs, no work for days, etc.
This is not the RAH I started with.

Tasks ready to send 0
Tasks in progress 169215

And now that it is the weekend, nothing will change.
RAH is beginning to make me wonder if its stable or not.
Bugs, no work for days, etc.
This is not the RAH I started with.

We've been told to expect periods of down time. It's even at the top of the project home page.

https://boinc.bakerlab.org/rosetta/forum_thread.php?id=14290
https://boinc.bakerlab.org/rosetta/

---