A few weeks ago, my two participating computers stopped updating tasks and computing them. One of the log files contains the following recent trailing report:

Wed 22 Aug 2012 12:06:09 PM EDT | rosetta@home | Started upload of lr_aa_bench_4dn0A_SAVE_ALL_OUT_IGNORE_THE_REST_53125_68_0_0
Wed 22 Aug 2012 12:06:13 PM EDT | rosetta@home | Finished upload of lr_aa_bench_4dn0A_SAVE_ALL_OUT_IGNORE_THE_REST_53125_68_0_0
Wed 22 Aug 2012 05:11:49 PM EDT | rosetta@home | update requested by user
Wed 22 Aug 2012 05:11:52 PM EDT | rosetta@home | Sending scheduler request: Requested by user.
Wed 22 Aug 2012 05:11:52 PM EDT | rosetta@home | Reporting 8 completed tasks, requesting new tasks for CPU and NVIDIA
Wed 22 Aug 2012 05:11:54 PM EDT | rosetta@home | Scheduler request completed: got 0 new tasks
Wed 22 Aug 2012 05:11:54 PM EDT | rosetta@home | No work sent
Wed 22 Aug 2012 05:11:54 PM EDT | rosetta@home | (reached daily quota of 8 results)
Wed 22 Aug 2012 05:45:08 PM EDT | rosetta@home | Sending scheduler request: To fetch work.
Wed 22 Aug 2012 05:45:08 PM EDT | rosetta@home | Requesting new tasks for CPU and NVIDIA
Wed 22 Aug 2012 05:45:10 PM EDT | rosetta@home | Scheduler request completed: got 0 new tasks
Wed 22 Aug 2012 05:45:10 PM EDT | rosetta@home | No work sent

What seeting needs adjusted to continue continuous job update and computation?

The project establishes the daily quota. It is 100 tasks per CPU per day.

If your machine has run through more than that... something in wrong. That is what the quota is for, to prevent a single machine from failing on a huge pile of work.

If you have solved the problem that was failing so many tasks, then within 24hours you will begin getting work again.

---

Rosetta Moderator: Mod.Sense

When I close then re-open BOINC manager, project communication is immediately deferred by 11:11:58 .... to my knowledge, I have never set a parameter that requests over 11hrs of communication deferral. How to set the scheduler RPC to a more reasonable value (5-10 minutes)?

Could this be related to the problem that Update just doesn't ... there is no upload of finished tasks, and no download of new ones.

I'm sensing the need to better understand the dialogue between the Manager and the client; that, and how various core system parameters are set by the Project Server.

---

The BOINC Manager establishes that wait timer. It basically waits increasingly longer when it hits the scheduler and gets no work back. The project establishes a minimum time between requests which is on the order of 5 minutes. Meaning that if you try to update to the project (i.e. hit the project scheduler) more than once in a 5 minute period, the subsequent requests will be rejected.

So, now that 5 minutes has passed, and you feel any network connectivity, firewall, Anti-Virus, authority, etc. issues are resolved, go ahead and go to the projects tab, select Rosetta and hit the update button.

...but there is some reason that it was failing and waiting longer periods of time. What messages are showing in BOINC Manager's event log when you update the project?

---

Rosetta Moderator: Mod.Sense

Hi,

I guess you have 64-bit Ubuntu and 64-bit BOINC running on it. I had this case and noticed that Rosetta@home got 100% computation errors on 64-bit BOINC. The workaround is to install 32-bit BOINC.
Run these lines in parent directory where you want to install BOINC. In my case it is /opt.

sudo apt-get remove --purge 'boinc*'
wget http://boinc.berkeley.edu/dl/boinc_7.0.28_i686-pc-linux-gnu.sh
chmod 755 boinc_7.0.28_i686-pc-linux-gnu.sh
./boinc_7.0.28_i686-pc-linux-gnu.sh
sudo apt-get install ia32-libs libstdc++6 libstdc++5 freeglut3 libwxgtk2.8-0:i386 libnotify-bin:i386
./BOINC/run_manager

While waiting the working 64-bit Rosetta@home implementation...

Regards, Markus