There is work now.

I had to add via boinccmd as it said invalid/bad password with BOINCTasks.

I got some tasks to download. But other tasks "transient HTTP error".

---

Reno, NV
Team: SETI.USA

I got some tasks to download. But other tasks "transient HTTP error".

I got some too on my Linux laptop, they are Linux only, and they are taking about 2+ hours each according to the initial estimates

Now scheduler error, when trying to report completed tasks.

4572 RamanujanMachine 10/16/2021 7:18:17 PM update requested by user
4573 RamanujanMachine 10/16/2021 7:18:22 PM [sched_op] sched RPC pending: Requested by user
4574 RamanujanMachine 10/16/2021 7:18:22 PM [sched_op] Starting scheduler request
4575 RamanujanMachine 10/16/2021 7:18:23 PM Sending scheduler request: Requested by user.
4576 RamanujanMachine 10/16/2021 7:18:23 PM Reporting 42 completed tasks
4577 RamanujanMachine 10/16/2021 7:18:23 PM Requesting new tasks for CPU
4578 RamanujanMachine 10/16/2021 7:18:23 PM [sched_op] CPU work request: 111290.19 seconds; 0.00 devices
4579 RamanujanMachine 10/16/2021 7:18:23 PM [sched_op] NVIDIA GPU work request: 0.00 seconds; 0.00 devices
4580 RamanujanMachine 10/16/2021 7:18:24 PM Scheduler request failed: Error 403
4581 RamanujanMachine 10/16/2021 7:18:24 PM [sched_op] Deferring communication for 03:58:51
4582 RamanujanMachine 10/16/2021 7:18:24 PM [sched_op] Reason: Scheduler request failed

---

Reno, NV
Team: SETI.USA

I notified the project from the non-BOINC page. They passed it on to someone who is in charge of the BOINC side, I think. I am not yet able to report tasks. Anyone else have this problem still?

Anyone else have this problem still?

Yes, I had that problem.
Reset the project and got new tasks
Scheduler started working fine.

New Project:
RamanujanMachine
https://rnma.xyz/boinc

There is an invite code here:
https://www.reddit.com/r/BOINC/comments/q7vqfs/the_ramanujan_machine_a_collection_of_algorithms/

No team creation yet.
Dunno how many tasks they'll create.

App:
Ramanujan Machine
Linux x86_64 only
suggests Ubuntu 20.04, but definitely needs GLIBC_2.29

Please add to WuProp - I (and othere here) already have hours

I spent a very frustrating 7 hours trying to get Fedora 29 (GLIBC 2.28) to upgrade to Fedora 31 (GLIBC 2.30+).
Getting rid of spurious bits of old Fedora distros which where stopping the upgrade dpwnloads, to being able to download the upgrade files but told that the system is not ready to upgrade? (Fedora is currently at version 34, so I am sure version 29 is ready for an upgrade).

Finally got the downloads (for the 4th time at 3.5 GB a hit) to again download, clean up old files, delete unwanted modules and a few other things.
A reboot got me back to Fedora 29? No 31?
When the upgrade finally started (all 6,658 files), it took over 2 hours to run and then did not reboot.

It just ran as Fedora 31, weird?

However I did all this to get some work from RamanujanMachine and find that it had none to give me and I am still waiting to see if what I did allows me to get and process work.

The things we do?

Conan

---

I'm good. I think nearly all my 261 pending have zombie as wingman.

Tasks ran out yesterday. Only thing left are timeouts/resends.

I'm good. I think nearly all my 261 pending have zombie as wingman.

Tasks ran out yesterday. Only thing left are timeouts/resends.

I got a few tasks overnight and this morning am over my 100 hours minimum. WOO HOO!!

As for getting or sending tasks back I've had no problems at all, now I didn't sit there and watch it all day long but everytime I went into the room it's in it was doing just fine. I'm running the latest version of Linux Mint with 16gb of ram and the cpu is an Intel(R) Core(TM) i5-2520M CPU @ 2.50GHz with 4 cores, I'm using all 4 cores to crunch along with wuprop and Ithena and it just works.

Right now it is running out of work though.

I created an Ubuntu VM for running this project. That’s how I solved the library problem. Easy enough. Better than forcing another system to use the wrong library.

But the 403 problem took some research. It’s an Apache configuration problem on the server. The buffer for incoming requests is way too small. SETI had this problem when people had a thousand tasks. This project runs into trouble with just four or five.

I worked around it by setting BOINC to run only two tasks, with no cache. Problem solved, but the VM can only do 48 hours per day.

---

Want to find one of the largest known primes? Try PrimeGrid. Or help cure disease at WCG.

I'm good. I think nearly all my 261 pending have zombie as wingman.

Tasks ran out yesterday. Only thing left are timeouts/resends.

I got a few tasks overnight and this morning am over my 100 hours minimum. WOO HOO!!

As for getting or sending tasks back I've had no problems at all, now I didn't sit there and watch it all day long but everytime I went into the room it's in it was doing just fine. I'm running the latest version of Linux Mint with 16gb of ram and the cpu is an Intel(R) Core(TM) i5-2520M CPU @ 2.50GHz with 4 cores, I'm using all 4 cores to crunch along with wuprop and Ithena and it just works.

Right now it is running out of work though.

I just did an update and got a new task, the Server Status page says 344 tasks available.

I'm getting the same 403 error and am unable to report completed tasks. The tasks do upload okay.

I'm getting the same 403 error and am unable to report completed tasks. The tasks do upload okay.

My previous message (two before yours) explains how I created a workaround solution to the problem.

You'll need to reset the project (there's no saving the tasks you currently have), set the cache to 0, and run no more than two cores. If there's more than about 2 tasks currently allocated to your computer, the server can't process the request.

It's a real pain, but this is the best we can do until they fix the server configuration.

---

Want to find one of the largest known primes? Try PrimeGrid. Or help cure disease at WCG.

That didn't change anything here. Now I have 1 wu locked up with the 403 error.

FWIW, I forwarded to the admin the 403 issue, and gave him pointers to this thread. Perhaps it will be fixed when we get more tasks. We'll see, I guess.

It sure would be easer if the project forum was turned on, or perhaps a discord channel.

---

Reno, NV
Team: SETI.USA

FWIW, I forwarded to the admin the 403 issue, and gave him pointers to this thread. Perhaps it will be fixed when we get more tasks. We'll see, I guess.

It must have worked because my laptop got a boatload of them last night

I exchanged email with the admin. He hopes to turn on stats export in a couple of weeks.

FWIW, I am still having the 403 error when I try to run more than two tasks at a time.

---

Reno, NV
Team: SETI.USA

I exchanged email with the admin. He hopes to turn on stats export in a couple of weeks.

FWIW, I am still having the 403 error when I try to run more than two tasks at a time.

I'm now out of them but I was running 4 at a time with no problems...that's strange

I'm also out at this point, but when the last work was issued I saw no further 403 errors and one of my machines was running 48 concurrently. The other machines where running 75% of threads without issue.

As well as iThena changing its name to iThena.Measurements (See missing hours thread),
there isa 2nd iThena project called iThena.Computational
One app so far, "iThena GRAPH", x86_64 Linux CPU only.
Some of us have hours already...